Dissertations / Theses: 'Accident detector'

1

Petito, Anthony Bruno 1967. "Design of a shield system for a hyper-pure germanium detector as a stack monitor for use in accident conditions at a nuclear power plant." Thesis, The University of Arizona, 1993. http://hdl.handle.net/10150/278343.

Full text

Abstract:

Collimator and shield configurations for two high-purity germanium detectors were designed for use during a loss of coolant accident at a boiling water reactor. The detectors will return information concerning stack releases to operators within a 15 minute time frame. Operating parameters for the stack monitors are defined by the United States Nuclear Regulatory Commission (USNRC) and a 24 hour source term generated by ORIGEN2. A lead collimator 0.4 cm in diameter, 20 cm in length for the high range detector and 2 cm in diameter, 20 cm in length for the low range detector was shown through a Monte Carlo code, MCNP4 to prevent high range detector saturation and provide enough low range detector response so good statistical data on stack releases result. A lead shield 20 cm thick was shown through MCNP4 to reduce the background radiation interference for both detectors to levels such that the detection of isotopes within the stack effluent is possible as required by the USNRC.

APA, Harvard, Vancouver, ISO, and other styles

2

Gomes, Vítor Emanuel Ornelas. "Smartphone based accident detection and eCall implementation." Master's thesis, Universidade de Aveiro, 2013. http://hdl.handle.net/10773/12835.

Full text

Abstract:

Mestrado em Engenharia Electrónica e Telecomunicações Intelligent Transportation Systems are emerging, to increase safety, e - ciency and comfort on roads. This intelligence is due to the fact that new technologies are being introduced in the most recent automobiles. As a result of this technological evolution, vehicular communication systems are being developed, to provide drivers with more information about the interventionists present in the roads they circulate. Predictions point that this information can increase safety and e ciency on roads. Presently, the Instituto de Telecomunica c~oes de Aveiro, is developing its own vehicular communication system, named HEADWAY, as a solution. HEADWAY DSRC 5.9 GHz vehicular communication system currently under development. Smartphones nowadays are very popular devices. This is due to the fact that they pack incredible hardware resources in a small and portable device and the possibility to third party developers, develop applications for them. This enables these devices to be used in di erent areas, depending only from the creativity of the developers. To diminish the number of fatalities due to road accidents, the European Commission has mandated the implementation of eCall in every new vehicle by 2015. In vehicles, the eCall aim to detect accidents and request accidents automatically. This dissertation targets, on the one hand, the development of an accident detection mechanism with eCall implementation. On the other hand it targets the integration of smartphones with HEADWAY, by developing an application that takes advantage of the system characteristics and demonstrates it. To achieve the proposed goals, an Android application was developed which acts as an HMI for HEADWAY, enables message exchange between vehicles, automatically detects accidents and proceeds with a help request. Most of the proposed goals where achieved, except the eCall implementation, which an alternative method was developed. Os Sistemas de Transporte Inteligentes estão a emergir, de forma a introduzir mais segurança, eficiência e conforto nas estradas. Esta inteligência deve-se ao facto de novas tecnologias estarem a ser introduzidas nos automóveis recentes. Como resultado da evolução tecnológica os sistemas de comunicação veiculares estão a ser desenvolvidos, com o objectivo de munir os condutores com informações relativas aos diferentes intervenientes da estrada onde circulam. Prevê-se que este tipo de informação leve a uma maior segurança e eficiência nas estradas. Actualmente no Instituto de Telecomunicações de Aveiro, está a decorrer um projecto que visa fornecer uma alternativa como sistema de comunicações veiculares. Este projecto tem o nome de HEADWAY. O HEADWAY é um sistema de comunicações veiculares DSRC 5.9 GHz, atualmente em desenvolvimento. Os smartphones hoje em dia já são dispositivos estabelecidos no mercado. Isto deve-se ao facto destes apresentarem um grande potencial, ao integrarem recursos de hardware incríveis num pequeno dispositivo e de permitirem o desenvolvimento de aplicações por terceiros. A criatividade dos programadores tem permitido a utilização destes dispositivos em diversas áreas. De forma a diminuir o número de mortes causadas por acidentes rodoviários, a Comissão Europeia, tornou obrigatório que em 2015 todos os novos carros estejam equipados com o sistema eCall, que visa a deteção de acidentes e pedido de ajuda ao 112 automáticos. Esta dissertação tem por um lado, o objectivo de desenvolver um detector de acidentes com implementação de eCall, e, por outro lado, integrar um smartphone com o HEADWAY, através do desenvolvimento de uma aplicação que tire partido das características deste sistema e assim o demonstre. Para cumprir os objectivos foi desenvolvida uma aplicação para Android que atua como HMI para o HEADWAY, facilita a troca de mensagens entre veículos, deteta automaticamente acidentes e procede com pedidos de ajuda. Na conclusão do projecto, verificou-se que os objectivos propostos foram na sua maioria concluídos, exceptuando a implementação da eCall ao 112, sendo desenvolvido um método alternativo.

APA, Harvard, Vancouver, ISO, and other styles

3

Černoch, Adam. "Vybrané způsoby zlepšení orientace řidiče v dopravním prostoru." Master's thesis, Vysoké učení technické v Brně. Ústav soudního inženýrství, 2014. http://www.nusl.cz/ntk/nusl-233053.

Full text

Abstract:

The aim of this thesis is to find suitable measures which will lead to the improvement of driver orientation in traffic spaceleading to lower the accident rate in the area. The theoretical part describes the traffic surveys that relate to the topic. Furthermore, the work mentioned detectors used for the implementation of automated traffic surveys. The theoretical part also describes methods that can be used to analyze the selected location and then design the ideal modification. At the end gives an overview of the various measures. In the practical part are different methodologies used for the analysis of selected intersections. Firstly, the analysis of observed conflict situations, including making conflicting diagram, under which was designed to measure. Then, the analysis of traffic accidents. Again, the result was a proposal for possible actions. The main objective was to compare and appreciation of both methodologies and proces design measures to improve driver orientation in the selected location.

APA, Harvard, Vancouver, ISO, and other styles

4

Lawoyin, Samuel. "Novel technologies for the detection and mitigation of drowsy driving." VCU Scholars Compass, 2014. http://scholarscompass.vcu.edu/etd/3639.

Full text

Abstract:

In the human control of motor vehicles, there are situations regularly encountered wherein the vehicle operator becomes drowsy and fatigued due to the influence of long work days, long driving hours, or low amounts of sleep. Although various methods are currently proposed to detect drowsiness in the operator, they are either obtrusive, expensive, or otherwise impractical. The method of drowsy driving detection through the collection of Steering Wheel Movement (SWM) signals has become an important measure as it lends itself to accurate, effective, and cost-effective drowsiness detection. In this dissertation, novel technologies for drowsiness detection using Inertial Measurement Units (IMUs) are investigated and described. IMUs are an umbrella group of kinetic sensors (including accelerometers and gyroscopes) which transduce physical motions into data. Driving performances were recorded using IMUs as the primary sensors, and the resulting data were used by artificial intelligence algorithms, specifically Support Vector Machines (SVMs) to determine whether or not the individual was still fit to operate a motor vehicle. Results demonstrated high accuracy of the method in classifying drowsiness. It was also shown that the use of a smartphone-based approach to IMU monitoring of drowsiness will result in the initiation of feedback mechanisms upon a positive detection of drowsiness. These feedback mechanisms are intended to notify the driver of their drowsy state, and to dissuade further driving which could lead to crashes and/or fatalities. The novel methods not only demonstrated the ability to qualitatively determine a drivers drowsy state, but they were also low-cost, easy to implement, and unobtrusive to drivers. The efficacy, ease of use, and ease of access to these methods could potentially eliminate many barriers to the implementation of the technologies. Ultimately, it is hoped that these findings will help enhance traveler safety and prevent deaths and injuries to users.

APA, Harvard, Vancouver, ISO, and other styles

5

Li, Yun Popescu Mihail. "Fall detection using sound sensors." Diss., Columbia, Mo. : University of Missouri--Columbia, 2009. http://hdl.handle.net/10355/6651.

Full text

Abstract:

Title from PDF of title page (University of Missouri--Columbia, viewed on March 10, 2010). The entire thesis text is included in the research.pdf file; the official abstract appears in the short.pdf file; a non-technical public abstract appears in the public.pdf file. Thesis advisor: Dr. Mihail Popescu. Includes bibliographical references.

APA, Harvard, Vancouver, ISO, and other styles

6

Balraj, Navaneethakrishnan. "AUTOMATED ACCIDENT DETECTION IN INTERSECTIONS VIA DIGITAL AUDIO SIGNAL PROCESSING." MSSTATE, 2003. http://sun.library.msstate.edu/ETD-db/theses/available/etd-10212003-102715/.

Full text

Abstract:

The aim of this thesis is to design a system for automated accident detection in intersections. The input to the system is a three-second audio signal. The system can be operated in two modes: two-class and multi-class. The output of the two-class system is a label of ?crash? or ?non-crash?. In the multi-class system, the output is the label of ?crash? or various non-crash incidents including ?pile drive?, ?brake?, and ?normal-traffic? sounds. The system designed has three main steps in processing the input audio signal. They are: feature extraction, feature optimization and classification. Five different methods of feature extraction are investigated and compared; they are based on the discrete wavelet transform, fast Fourier transform, discrete cosine transform, real cepstrum transform and Mel frequency cepstral transform. Linear discriminant analysis (LDA) is used to optimize the features obtained in the feature extraction stage by linearly combining the features using different weights. Three types of statistical classifiers are investigated and compared: the nearest neighbor, nearest mean, and maximum likelihood methods. Data collected from Jackson, MS and Starkville, MS and the crash signals obtained from Texas Transportation Institute crash test facility are used to train and test the designed system. The results showed that the wavelet based feature extraction method with LDA and maximum likelihood classifier is the optimum design. This wavelet-based system is computationally inexpensive compared to other methods. The system produced classification accuracies of 95% to 100% when the input signal has a signal-to-noise-ratio of at least 0 decibels. These results show that the system is capable of effectively classifying ?crash? or ?non-crash? on a given input audio signal.

APA, Harvard, Vancouver, ISO, and other styles

7

Hiemer, Marcus. "Model based detection and reconstruction of road traffic accidents." Karlsruhe : Univ.-Verl, 2005. http://deposit.ddb.de/cgi-bin/dokserv?idn=974366552.

Full text

APA, Harvard, Vancouver, ISO, and other styles

8

Hamdane, Hedi. "Improvement of pedestrian safety : response of detection systems to real accident scenarios." Thesis, Aix-Marseille, 2016. http://www.theses.fr/2016AIXM4091.

Full text

Abstract:

Le contexte général de cette recherche concerne la sécurité active des piétons. De nombreux systèmes embarqués dans les véhicules sont actuellement développés afin de détecter un piéton sur la chaussée et d’éviter une collision soit par une manœuvre de freinage d’urgence soit par une manœuvre de déport. La plupart de ces systèmes d’aide à la conduite sont basés sur des systèmes de détection (caméras, radars, etc). Ils analysent la scène en temps réel, puis effectuent un traitement d’images dans le but d’identifier un potentiel danger. Or il apparaît difficile de déterminer la pertinence de ces systèmes en termes de sécurité routière. L’'objectif général de ce travail est ainsi d’estimer cette pertinence en confrontant les systèmes à de multiples configurations d’accidents réels. La méthodologie consiste à tester les systèmes de détection des piétons dans les configurations d’accidents reconstruits en les associant à la cinématique des véhicules. Le test de performance de ces systèmes a été alors réalisé en vérifiant leurs compatibilités au regard de la chronologie des accidents; i.e. vérifier la possibilité d’'évitement des accidents. À partir de ces reconstructions d’accidents réels, une analyse a été réalisée afin de dégager les enjeux au niveau spatio-temporelle qui influencent la sécurité primaire du piéton The scope of this research concerns pedestrian active safety. Several primary safety systems have been developed for vehicles in order to detect a pedestrian and to avoid an impact. These systems analyse the forward path of the vehicle through the processing of images from sensors. If a pedestrian is identified on the vehicle trajectory, these systems employ emergency braking and some systems may potentially employ emergency steering. Methods for assessing the effectiveness of these systems have been developed. But, it appears difficult to determine the relevance of these systems in terms of pedestrian protection. The general objective of this research was to test the response of these systems in many accident configurations.The methodology consisted of coupling the vehicle dynamic behaviour with a primary safety system in order to confront these systems to real accident configurations. The relevance of these systems is studied by verifying the feasibility of deploying an autonomous emergency manoeuvre during the timeline of the accident and according to the vehicle dynamic capabilities: i.e. verifying the possibilities in terms of crash avoidance. From these accident reconstructions and simulation, factors relevant to the primary safety of pedestrians were deduced

APA, Harvard, Vancouver, ISO, and other styles

9

Khaghani, Farnaz. "A Deep Learning Approach to Predict Accident Occurrence Based on Traffic Dynamics." Thesis, Virginia Tech, 2020. http://hdl.handle.net/10919/98801.

Full text

Abstract:

Traffic accidents are of concern for traffic safety; 1.25 million deaths are reported each year. Hence, it is crucial to have access to real-time data and rapidly detect or predict accidents. Predicting the occurrence of a highway car accident accurately any significant length of time into the future is not feasible since the vast majority of crashes occur due to unpredictable human negligence and/or error. However, rapid traffic incident detection could reduce incident-related congestion and secondary crashes, alleviate the waste of vehicles’ fuel and passengers’ time, and provide appropriate information for emergency response and field operation. While the focus of most previously proposed techniques is predicting the number of accidents in a certain region, the problem of predicting the accident occurrence or fast detection of the accident has been little studied. To address this gap, we propose a deep learning approach and build a deep neural network model based on long short term memory (LSTM). We apply it to forecast the expected speed values on freeways’ links and identify the anomalies as potential accident occurrences. Several detailed features such as weather, traffic speed, and traffic flow of upstream and downstream points are extracted from big datasets. We assess the proposed approach on a traffic dataset from Sacramento, California. The experimental results demonstrate the potential of the proposed approach in identifying the anomalies in speed value and matching them with accidents in the same area. We show that this approach can handle a high rate of rapid accident detection and be implemented in real-time travelers’ information or emergency management systems. M.S. Rapid traffic accident detection/prediction is essential for scaling down non-recurrent conges- tion caused by traffic accidents, avoiding secondary accidents, and accelerating emergency system responses. In this study, we propose a framework that uses large-scale historical traffic speed and traffic flow data along with the relevant weather information to obtain robust traffic patterns. The predicted traffic patterns can be coupled with the real traffic data to detect anomalous behavior that often results in traffic incidents in the roadways. Our framework consists of two major steps. First, we estimate the speed values of traffic at each point based on the historical speed and flow values of locations before and after each point on the roadway. Second, we compare the estimated values with the actual ones and introduce the ones that are significantly different as an anomaly. The anomaly points are the potential points and times that an accident occurs and causes a change in the normal behavior of the roadways. Our study shows the potential of the approach in detecting the accidents while exhibiting promising performance in detecting the accident occurrence at a time close to the actual time of occurrence.

APA, Harvard, Vancouver, ISO, and other styles

10

Zino, John Frederick. "Monte Carlo based exposure rate response estimates for criticality accident detectors at the Savannah River site." Thesis, Georgia Institute of Technology, 1994. http://hdl.handle.net/1853/17554.

Full text

APA, Harvard, Vancouver, ISO, and other styles

11

Silva, Cares Matías Ignacio. "Diseño y construcción de un sistema para detectar, localizar y caracterizar accidentes automovilísticos." Tesis, Universidad de Chile, 2017. http://repositorio.uchile.cl/handle/2250/149475.

Full text

Abstract:

Ingeniero Civil Eléctrico Los accidentes automovilísticos son un grave problema mundial sin solución. La gran cantidad de heridos y muertes como resultado de choques en carretera o ciudad ha ido en aumento año tras año debido a la gran cantidad de viajes que ocurren en todo el mundo. Las estadísticas dicen que el 25\% de la gente involucrada en accidentes podría haberse salvado con una asistencia médica inmediata, es por eso que la empresa Sosmart Labs desarrolló una aplicación para detectar choques y notificar inmediatamente la ubicación y tiempo del suceso a cercanos del pasajero. Esta memoria se enmarca en el proyecto Sosmart Premium de la empresa, que consiste en diseñar y contruir un dispositivo electrónico que se ancle al vehículo y que realice la misma tarea de la aplicación Sosmart, además de implementar el sistema de comunicación y un proceso de post-procesamiento para caracterizar el accidente. El trabajo consistió en el diseño e implementación de un prototipo de dispositivo electrónico a través de la elección de componentes de mercado y fácil uso, en la que se optó por Arduino como controlador, un MPU9250 como sensor de aceleraciones y un FONA 2G como módulo para comunicaciones. Luego, se desarrolló la arquitectura completa implementando una plataforma web para recibir las notificaciones y también un programa de post-procesamiento para limpiar las señales y usando un programa de simulación 3D, recrear el accidente. El dispositivo electrónico posee un algoritmo de detección de choque autónomo que fue diseñado utilizando una base de datos de accidentes de tráfico de Estados Unidos y con ella, se desmostró que las magnitudes de aceleraciones en un choque real, poseen un valor alto que es fácil de detectar utilizando un acelerómetro de alto rango. En este caso, el MPU9250 demostró ser lo suficientemente útil para utilizar el algoritmo y discernir eficazmente de choques reales con falsos positivos. Por otro lado, el proceso notificación funcionó para los 3 canales propuestos: SMS, llamada y POST request a una plataforma web de monitoreo. Finalmente, el post-procesamiento consistió en la aplicación de filtros y zonas de histéresis para atenuar el ruido producto del sensor y con el uso de integraciones, obtener los desplazamientos y ángulos del sensor para simularlos en Unity3D. Concluyendo, se desarrolló un sistema de detección, notificación y simulación de accidenttes, en los que el sistema de detección obtuvo buenos resultados para accidentes de gran magnitud y se aislaron casos de falsos positivos, se creó la plataforma web para monitorear accidentes y en cuanto a la caracterización, el sistema desarrollado permite obtener una idea de la orientación del vehículo previo al accidente, pero no fue posible determinar la trayectoria real de este.

APA, Harvard, Vancouver, ISO, and other styles

12

Hiemer, Marcus [Verfasser]. "Model based detection and reconstruction of road traffic accidents / von Marcus Hiemer." Karlsruhe : Univ.-Verl. Karlsruhe, 2005. http://d-nb.info/974366552/34.

Full text

APA, Harvard, Vancouver, ISO, and other styles

13

Freibert, Emily Jane. "Assessing internal contamination levels for fission product inhalation using a portal monitor." Thesis, Georgia Institute of Technology, 2010. http://hdl.handle.net/1853/37184.

Full text

Abstract:

In the event of a nuclear power plant accident, fission products could be released into the atmosphere potentially affecting the health of local citizens. In order to triage the possibly large number of people impacted, a detection device is needed that can acquire data quickly and that is sensitive to internal contamination. The portal monitor TPM-903B was investigated for use in the event of a fission product release. A list of fission products released from a Pressurized Water Reactor (PWR) was generated and separated into two groups--Group 1 (gamma- and beta-emitting fission products) and Group 2 (strictly beta-emitting fission products.) Group one fission products were used in the previously validated Monte Carlo N-Particle Transport Code (MCNP) model of the portal monitor. Two MIRD anthropomorphic phantom types were implemented in the MCNP model--the Adipose Male and Child phantoms. Dose and Risk Calculation software (DCAL) provided inhalation biokinetic data that were applied to the output of the MCNP modeling to determine the radionuclide concentrations in each organ as a function of time. For each phantom type, these data were used to determine the total body counts associated with each individual gamma-emitting fission product. Corresponding adult and child dose coefficients were implemented to determine the total body counts per 250 mSv. A weighted sum of all of the isotopes involved was performed. The ratio of dose associated with gamma-emitting fission products to the total of all fission products was determined based on corresponding dose coefficients and relative abundance. This ratio was used to project the total body counts corresponding to 250mSv for the entire fission product release inhalation--including all types of radiation. The developed procedure sheets will be used by first response personnel in the event of a fission product release.

APA, Harvard, Vancouver, ISO, and other styles

14

Zhou, Dingshan Sam. "An integrated traffic incident detection model /." Full text (PDF) from UMI/Dissertation Abstracts International, 2000. http://wwwlib.umi.com/cr/utexas/fullcit?p9992952.

Full text

APA, Harvard, Vancouver, ISO, and other styles

15

Abas, Ashardi B. "Non-intrusive driver drowsiness detection system." Thesis, University of Bradford, 2011. http://hdl.handle.net/10454/5521.

Full text

Abstract:

The development of technologies for preventing drowsiness at the wheel is a major challenge in the field of accident avoidance systems. Preventing drowsiness during driving requires a method for accurately detecting a decline in driver alertness and a method for alerting and refreshing the driver. As a detection method, the authors have developed a system that uses image processing technology to analyse images of the road lane with a video camera integrated with steering wheel angle data collection from a car simulation system. The main contribution of this study is a novel algorithm for drowsiness detection and tracking, which is based on the incorporation of information from a road vision system and vehicle performance parameters. Refinement of the algorithm is more precisely detected the level of drowsiness by the implementation of a support vector machine classification for robust and accurate drowsiness warning system. The Support Vector Machine (SVM) classification technique diminished drowsiness level by using non intrusive systems, using standard equipment sensors, aim to reduce these road accidents caused by drowsiness drivers. This detection system provides a non-contact technique for judging various levels of driver alertness and facilitates early detection of a decline in alertness during driving. The presented results are based on a selection of drowsiness database, which covers almost 60 hours of driving data collection measurements. All the parameters extracted from vehicle parameter data are collected in a driving simulator. With all the features from a real vehicle, a SVM drowsiness detection model is constructed. After several improvements, the classification results showed a very good indication of drowsiness by using those systems.

APA, Harvard, Vancouver, ISO, and other styles

16

Putnam, Michael R. "Responding to Dangerous Accidents Among the Elderly: A Fall Detection Device with Zigbee-Based Positioning." DigitalCommons@CalPoly, 2012. https://digitalcommons.calpoly.edu/theses/859.

Full text

Abstract:

The following paper describes a fall detection and activity monitoring system with position detection based on Zigbee transceivers.The main objective is to reduce the time taken for emergency personnel to respond to falls among the elderly. Especially when the victim is unconscious or delirious, position tracking reduces location determination time within a busy hospital or nursing home environment and facilitates immediate treatment. Reduced response times correlate to decreased morbidity and mortality rates. Background is provided on the major wireless network advances currently deployed in a healthcare setting for asset and personnel tracking, etiology of falls, and several methods of detecting falls using sensors and image processing techniques. Data analysis proves that a precise coordinate tracking system was infeasible using the XBee RF module (based on the Zigbee protocol) due to environmental noise, a poor antenna construction and lack of precise signal strength measurements. A primitive scheme with lower resolution and higher reliability associating a single location with each Zigbee transceiver was employed. A pedometer function was added to the project to monitor the user’s daily activity and to potentially serve as a predictor of falls through the interpretation of mobility and gait patterns related to step counts.

APA, Harvard, Vancouver, ISO, and other styles

17

Martel, Christopher. "Development of a Mass Detection Technique to Detect Intakes of Radioactive Material and their Resulting Radiation Exposures Following a Large-Scale Radiological Release." Digital WPI, 2019. https://digitalcommons.wpi.edu/etd-dissertations/520.

Full text

Abstract:

Large-scale radiological accidents have resulted in intakes of radioactive materials by members of the public and occupational radiation workers. However, current methods to evaluate intakes are designed for small numbers of individuals and cannot be easily scaled for large populations as has occurred. A proposed method for high throughput volumes of people to identify and quantify intakes of radioactive material through urine radiobioassay is described. MATERIALS AND METHODS: The MCNP V6.0 software code was used to model the General Electric Hawkeye V3 Gamma Camera for gamma ray efficiency. Technitium-99m was used to validate the model. The model was used to calculate detection efficiencies and minimum detectable doses for Cobalt-60, Iodine-131, Cesium-137/Barium-137m and Iridium-192. RESULTS: Differences of 8% were observed between measurements of the detection efficiency for Technitium-99m and the MCNP modeled detection efficiency (11.1% vs. 12.0%, respectively). Calculations showed that a dose of 20 mSv could be detected using urine radiobioassay in 6, 3, 2, and 20 days post incident for Type F intakes of Cobalt-60, Iodine-131, Cesium-137/Barium-137m and Iridium-192 respectively. Approximately 1,152 urine samples could be analyzed in an eight-hour shift using a single gamma camera. CONCLUSIONS: The use of the gamma camera for urine radiobioassay allows for high throughput volumes of samples and has sufficient detection sensitivity to meet dose-based decision guidelines.

APA, Harvard, Vancouver, ISO, and other styles

18

Fouan, Damien. "Détection et caractérisation d'embolies gazeuses : application à la prévention des accidents de décompression." Thesis, Aix-Marseille, 2013. http://www.theses.fr/2013AIXM4763.

Full text

Abstract:

Les accidents de désaturation (ADD) résultent de la formation de microbulles dans les tissus lors de la décompression. De nombreux corps de métiers sont concernés par cette problématique : plongeurs, astronautes, tunneliers, personnel médical hyperbare... On observe que ces accidents peuvent survenir alors même que les tables de décompression sont respectées. La détection et la caractérisation des bulles de décompression présentent un intérêt diagnostic potentiel pour la prévention des ADD. Aujourd’hui, la détection de microbulles est réalisée par des personnels de santé expérimentés en utilisant des systèmes Doppler. Cependant, cette approche présente une grande dépendance à l'opérateur et ne permet pas d'obtenir une information quantitative (nombre, taille) sur la distribution des bulles circulantes. Par ailleurs, elle n'est pas adaptée à la détection de bulles tissulaires.Ces limitations conduisent à mettre en oeuvre une méthode ultrasonore bifréquentielle de détection et de caractérisation par mise en résonance des microbulles. Les contraintes de mesure temps-réel, de polydispersité en taille ([1 à 200 μm]) et de sursaturation des tissus imposent d'utiliser des ondes d'excitation de très faibles puissances mais très large bande.Les solutions mises en oeuvre visent d'une part à réduire la complexité de l'instrumentation et d'autre part à prendre en considération la dynamique de l'excitation. Par ailleurs, une solution originale, ayant fait l’objet d’un dépôt de brevet, est développée. Elle permet de s'affranchir de la mise en résonance tout en conservant un caractèrediscriminant séparant bulles et tissus Decompression sicknesses (DCS) are a consequence of microbubbles formation in tissues during decompression. Many fields are affected by this issue: divers, astronauts, tunneling, hyperbaric medical staff... It is observed that these accidents can occur even if the decompression tables are respected. The detection and characterization of decompression bubbles have a diagnostic potential for the prevention of DCS. Today, the detection of microbubbles is performed by experienced health workers using Doppler systems. However, this approach has a high dependence on the operator and does not provide quantitative information (number, size) about the distribution of circulating bubbles. Moreover, it is not suitable for the detection of stationary bubbles (tissue bubbles).These limitations lead to the development of a bi-frequency ultrasonic method for microbubbles detection and characterization by setting them into resonance. The constraints as real-time measurements, size polydispersity ([1 to 200 μm]) and saturation of tissues require the use of very low powerful excitation but high bandwidth waves.The solutions implemented are aimed firstly to reduce the complexity of the instrumentation and secondly to consider the dynamics of the excitation. In addition, an original solution, protected by a patent, has been developed. It allows to overcome the measurement of resonance while maintaining a discriminating character between bubbles and tissues

APA, Harvard, Vancouver, ISO, and other styles

19

GALAN, CATILINA EMMANUELLE CHARLETTE. "Faut-il prendre en compte l'age dans l'interpretation des resultats des tests psychotechniques utilises dans la detection de la propension aux accidents ? implications pratiques dans le domaine de la conduite des vehicules automobiles." Clermont-Ferrand 1, 1993. http://www.theses.fr/1993CLF1M036.

Full text

APA, Harvard, Vancouver, ISO, and other styles

20

Lundberg, Catarina. "Older drivers with cognitive impairments : issues of detection and assessment /." Stockholm, 2003. http://diss.kib.ki.se/2003/91-7349-590-5.

Full text

APA, Harvard, Vancouver, ISO, and other styles

21

Vock, Dominik. "Automatic segmentation and reconstruction of traffic accident scenarios from mobile laser scanning data." Doctoral thesis, Saechsische Landesbibliothek- Staats- und Universitaetsbibliothek Dresden, 2014. http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa-141582.

Full text

Abstract:

Virtual reconstruction of historic sites, planning of restorations and attachments of new building parts, as well as forest inventory are few examples of fields that benefit from the application of 3D surveying data. Originally using 2D photo based documentation and manual distance measurements, the 3D information obtained from multi camera and laser scanning systems realizes a noticeable improvement regarding the surveying times and the amount of generated 3D information. The 3D data allows a detailed post processing and better visualization of all relevant spatial information. Yet, for the extraction of the required information from the raw scan data and for the generation of useable visual output, time-consuming, complex user-based data processing is still required, using the commercially available 3D software tools. In this context, the automatic object recognition from 3D point cloud and depth data has been discussed in many different works. The developed tools and methods however, usually only focus on a certain kind of object or the detection of learned invariant surface shapes. Although the resulting methods are applicable for certain practices of data segmentation, they are not necessarily suitable for arbitrary tasks due to the varying requirements of the different fields of research. This thesis presents a more widespread solution for automatic scene reconstruction from 3D point clouds, targeting street scenarios, specifically for the task of traffic accident scene analysis and documentation. The data, obtained by sampling the scene using a mobile scanning system is evaluated, segmented, and finally used to generate detailed 3D information of the scanned environment. To realize this aim, this work adapts and validates various existing approaches on laser scan segmentation regarding the application on accident relevant scene information, including road surfaces and markings, vehicles, walls, trees and other salient objects. The approaches are therefore evaluated regarding their suitability and limitations for the given tasks, as well as for possibilities concerning the combined application together with other procedures. The obtained knowledge is used for the development of new algorithms and procedures to allow a satisfying segmentation and reconstruction of the scene, corresponding to the available sampling densities and precisions. Besides the segmentation of the point cloud data, this thesis presents different visualization and reconstruction methods to achieve a wider range of possible applications of the developed system for data export and utilization in different third party software tools.

APA, Harvard, Vancouver, ISO, and other styles

22

Dingus, Thomas A. "Development of models for detection of automobile driver impairment." Thesis, Virginia Tech, 1985. http://hdl.handle.net/10919/45721.

Full text

Abstract:

Two of the leading causes of automobile accidents are driver impairment due to alcohol and drowsiness. Apparently, a relatively large percentage of these accidents occur because drivers are unaware of the degree to which they are impaired due to these sources. The purpose of this research was to develop models which could detect driver impairment due to alcohol, drowsiness, or the combination of alcohol and drowsiness, and which could be practically implemented in an automobile. Such detection models, if successfully implemented in conjunction with a system to warn an impaired driver of his or her condition, could potentially save hundreds of lives each year. Six driver-subjects operated a computer controlled driving simulator during each of four conditions. The four conditions consisted of a control condition, an alcohol condition, a sleep-deprived condition, and a combination alcohol and sleep-deprived condition. Moderate levels of alcohol and sleep deprivation were used for this study. Nineteen performance and behavioral measures were collected during this study. Each measure was evaluated singly and in combination with other measures to determine potential value for detection of driver impairment. Detection models were then formulated using the most promising detection measures. The results indicated that a useful on-board drowsiness impairment detection device is possible and practical for highway driving. This device would also, in all likelihood, provide useful detection information regardless of whether low to moderate amounts of alcohol were present in a drowsy driver. The results also showed that on-board alcohol impairment detection may be possible at moderate to high BAC. Master of Science

APA, Harvard, Vancouver, ISO, and other styles

23

Taylor-Piliae, Ruth E., M. Jane Mohler, Bijan Najafi, and Bruce M. Coull. "Objective fall risk detection in stroke survivors using wearable sensor technology: a feasibility study." TAYLOR & FRANCIS LTD, 2016. http://hdl.handle.net/10150/621778.

Full text

Abstract:

Background: Stroke survivors often have persistent neural deficits related to motor function and sensation, which increase their risk of falling, most of which occurs at home or in community settings. The use of wearable technology to monitor fall risk and gait in stroke survivors may prove useful in enhancing recovery and/or preventing injuries. Objective: Determine the feasibility of using wearable technology (PAMSys (TM)) to objectively monitor fall risk and gait in home and community settings in stroke survivors. Methods: In this feasibility study, we used the PAMSys to identify fall risk indicators (postural transitions: duration in seconds, and number of unsuccessful attempts), and gait (steps, speed, duration) for 48 hours during usual daily activities in stroke survivors (n=10) compared to age-matched controls (n=10). A questionnaire assessed device acceptability. Results: Stroke survivors mean age was 70 +/- 8 years old, were mainly Caucasian (60%) women (70%), and not significantly different than the age-matched controls (all P-values >0.20). Stroke survivors (100%) reported that the device was comfortable to wear, didn't interfere with everyday activities, and were willing to wear it for another 48 hours. None reported any difficulty with the device while sleeping, removing/putting back on for showering or changing clothes. When compared to controls, stroke survivors had significantly worse fall risk indicators and walked less (P<0.05). Conclusion: Stroke survivors reported high acceptability of 48 hours of continuous PAMSys monitoring. The use of in-home wearable technology may prove useful in monitoring fall risk and gait in stroke survivors, potentially enhancing recovery.

APA, Harvard, Vancouver, ISO, and other styles

24

Головатий, Ігор Богданович, та Ihor Holovatiy. "Комп'ютерна система на основі нейромережі для виявлення зіткнення автомобілів". Bachelor's thesis, Тернопільський національний технічний університет імені Івана Пулюя, 2021. http://elartu.tntu.edu.ua/handle/lib/35429.

Full text

Abstract:

Кваліфікаційна робота присвячена розробці системи, що дозволяє визначати серйозні автомобільні зіткнення на відеоряді, записаному камерами дорожнього спостереження. Проведено огляд існуючих систем детектування дорожньо-транспортних пригод. Запропоновано спосіб вирішення проблеми за допомогою нейромережі, Наведено опис алгоритму детектування автокатастроф, здійснено пошук і підготовка вибірки. Проаналізовано алгоритми детектування об'єктів. Для детектування автомобіля на відео вибрано YOLOv3-детектор. Здійснено порівняння детекторів об'єктів для з'ясування впливу на кінцеву ефективність роботи системи в цілому. Реалізовано алгоритм детектування зіткнення автомобілів в режимі реального часу. Розроблювана система була протестована на реальних даних для визначення зіткнень автомобілів. Отримані практичні результати дозволяють стверджувати про ефективність використання розробки.. The qualification work deals with the development of a system that allows you to identify serious car collisions on a video recorded by surveillance cameras. A review of existing road accident detection systems was conducted. The way of the decision of a problem by means of a neural network is offered, the description of algorithm of detection of car accidents is given, search and preparation of sampling is carried out. Object detection algorithms are analyzed. A YOLOv3 detector is selected to detect the car on video. The object detectors are compared to determine the impact on the final efficiency of the system as a whole. The algorithm of car collision detection in real time is implemented. The developed system was tested on real data to determine car collisions. The obtained practical results allow us to assert the effectiveness of the development. Вступ. 1. Аналіз технічного завдання. 1.1 Огляд систем детектування ДТП. 1.2. Поняття ДТП. 2. Проектна частина. 2.1. Згорткові нейронні мережі. 2.2. Спосіб вирішення проблеми. за допомогою згорткових нейронних мереж. 2.3. Опис алгоритму детектування автокатастроф. 2.4. Пошук і підготовка вибірки. 2.5. Аналіз алгоритмів детектування об'єктів. 3. Практична частина. 3.1. Порівняння детекторів об'єктів для з'ясування впливу на кінцеву ефективність роботи системи в цілому. 3.2. Реалізація алгоритму детектування зіткнення автомобілів в режимі реального часу. 3.3. Отримані результати експериментів. 4. Безпека життєдіяльності, основи хорони праці. Висновки. Список використаних джерел

APA, Harvard, Vancouver, ISO, and other styles

25

Verma, Vasudha. "Development of a Neutron Flux Monitoring System for Sodium-cooled Fast Reactors." Doctoral thesis, Uppsala universitet, Tillämpad kärnfysik, 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-319945.

Full text

Abstract:

Safety and reliability are one of the key objectives for future Generation IV nuclear energy systems. The neutron flux monitoring system forms an integral part of the safety design of a nuclear reactor and must be able to detect any irregularities during all states of reactor operation. The work in this thesis mainly concerns the detection of in-core perturbations arising from unwanted movements of control rods with in-vessel neutron detectors in a sodium-cooled fast reactor. Feasibility study of self-powered neutron detectors (SPNDs) with platinum emitters as in-core power profile monitors for SFRs at full power is performed. The study shows that an SPND with a platinum emitter generates a prompt current signal induced by neutrons and gammas of the order of 600 nA/m, which is large enough to be measurable. Therefore, it is possible for the SPND to follow local power fluctuations at full power operation. Ex-core and in-core detector locations are investigated with two types of detectors, fission chambers and self-powered neutron detectors (SPNDs) respectively, to study the possibility of detection of the spatial changes in the power profile during two different transient conditions, i.e. inadvertent withdrawal of control rods (IRW) and one stuck rod during reactor shutdown (OSR). It is shown that it is possible to detect the two simulated transients with this set of ex-core and in-core detectors before any melting of the fuel takes place. The detector signal can tolerate a noise level up to 5% during an IRW and up to 1% during an OSR.

APA, Harvard, Vancouver, ISO, and other styles

26

Muzammel, Muhammad. "Développement d'un système d'avertissemment sonore, validé par EEG, basé sur des approches vision et acoustique pour la detection de véhicules approchants des véhicules moteur deux roues." Thesis, Bourgogne Franche-Comté, 2018. http://www.theses.fr/2018UBFCK019/document.

Full text

Abstract:

Dans de nombreux pays, le taux de mortalité des motocyclistes est beaucoup plus élevé que celui des autres conducteurs de véhicules. Parmi de nombreux autres facteurs, les collisions arrière des motocyclettes contribuent fortement à ces décès de motards. Les systèmes de détection de collision peuvent être utilisés pour minimiser ces accidents mortels. Cependant, la plupart des systèmes de détection de collision existants n'identifient pas le type de danger potentiel auquel sont exposés les motocyclistes. Chaque système d'alerte de collision utilise une technique de détection de collision distincte, ce qui limite ses performances et rend impératif l'étude de son efficacité. Malheureusement, aucun travail de ce type n'a été signalé dans ce domaine particulier pour les motocyclistes. Par conséquent, il est important d'étudier la réponse physiologique du motocycliste contre ces systèmes d'alerte de collision. Dans cette recherche, une méthode de détection et de classification des véhicules approchant par l'arrière est présentée. Pour la détection de collision, une approche basee vision et la technique basee sur le son ont été utilisées. Pour les techniques visuelles et acoustiques, des caractéristiques d'apparence et de spectre de puissance ont été utilisées, respectivement, pour détecter le véhicule qui s'approche à l'extrémité arrière de la motocyclette. En ce qui concerne la classification des véhicules, seule une technique acoustique est utilisée; un spectre de puissance acoustique et des caractéristiques énergétiques sont utilisés pour classer les véhicules qui approchent. Deux types d'ensembles de données, à savoir des ensembles de données acquises durant ce travail (obtenues en plaçant une caméra à l'arrière d'une motocyclette) et des ensembles de données disponibles telechargeables (pour la détection visuelle et pour la classification audio des véhicules) sont utilisés pour la validation. La méthodologie proposée a permis de détecter et de classer les véhicules pour des ensembles de données acquises durent cette these. De même, pour les ensembles de données disponibles , le taux positif vrai le plus élevé et le taux de détection faux le plus faible ont été atteints par rapport aux méthodes de l etat de l art. En outre, une étude physiologique basée sur le potentiel lié à l'événement (ERP) a été réalisée sur les motocyclistes afin d'étudier leurs réponses vis-à-vis du système d'alerte de collision arrière. Deux types d'avertissements auditifs (c'est-à-dire verbal et buzzer) sont utilisés pour ce système d'avertissement. Pour étudier la réponse des motocyclistes, les composantes N1, N2, P3 et N400 ont été extraits des données d'électroencéphalographie (EEG). Ces systèmes d avertissement ont montré des effets positifs au niveau des neuronal sur les motocyclistes et réduisent leur temps de réaction et les ressources attentionnelles nécessaires pour traiter correctement la cible. En résumé, le système d'avertissement de collision par l'arrière proposé avec des avertissements verbaux auditifs augmente considérablement la vigilance du motocycliste et peut être utile pour éviter les scénarios possibles de collision arrière In many countries, motorcyclist fatality rate is much higher than that of other vehicle drivers. Among many other factors, motorcycle rear-end collisions are also contributing to these biker fatalities. Collision detection systems can be used to minimize these fatalities. However, most of the existing collision detection systems do not identify the type of potential hazard faced by motorcyclists. Every collision warning system used a distinctive collision detection technique, which limits its performance and makes it imperative to study its effectiveness. Unfortunately, no such work has been reported in that particular domain for motorcyclists. Therefore, it is important to study the physiological response of the motorcyclist against these collision warning systems. In this research, a rear end vehicle detection and classification method is presented for motorcyclists. For collision detection, vision technique and acoustic technique have been used. For visual and acoustic techniques, appearance features and power spectrum have been used, respectively, to detect the approaching vehicle at the rear end of the motorcycle. As for the vehicle classification, only an acoustic technique is utilized; an acoustic power spectrum and energy features are used to classify the approaching vehicles. Two types of datasets which are comprised of self-recorded datasets (obtained by placing a camera at the rear end of a motorcycle) and online datasets (for vision-based vehicle detection and for audio based vehicle classification techniques) are used for validation. Proposed methodology successfully detected and classified the vehicle for self-recorded datasets. Similarly, for online datasets, the higher true positive rate and less false detection rate has been achieved as compared to the existing state of the art methods. Moreover, an event-related potential (ERP) based physiological study has been performed on motorcyclists to investigate their responses towards the rear end collision warning system. Two types of auditory warnings (i.e., verbal and buzzer) are used for this warning system. To study the response of the motorcyclists, the N1, N2, P3, and N400 components have been extracted from the Electroencephalography (EEG) data. These introduced systems have shown positive effects at neural levels on motorcyclists and reduce their reaction time and attentional resources required for processing the target correctly. In summary, the proposed rear-end collision warning system with auditory verbal warnings significantly increases the alertness of the motorcyclist and can be helpful to avoid the possible rear-end collision scenarios

APA, Harvard, Vancouver, ISO, and other styles

27

Wu, Dan. "Quantification des causes des accidents de deux / trois-roues motorisés et de leurs conséquences corporelle (approche épidémiologique)s." Thesis, Lyon, 2018. http://www.theses.fr/2018LYSE1175/document.

Full text

Abstract:

Contexte : les deux-roues motorisés (2RM) sont en forte évolution et la vulnérabilité de leurs usagers est de plus en plus manifeste. En 2013, dans le monde, les usagers de 2RM ont compté pour 23 % de la mortalité routière, devant les piétons (22 %). Dans ce contexte, la sécurité des usagers de 2RM est devenue une préoccupation sérieuse dans la plupart des pays. Contrairement à beaucoup de problèmes de santé publique, les principaux facteurs de l'insécurité routière en 2RM sont connus. L'enjeu d'aujourd'hui est surtout de quantifier leur effet sur la survenue d'accidents et d'évaluer les effets des quelques dispositifs de protection disponibles. Les objectifs de cette thèse sont de mieux connaitre les problèmes d'interaction entre les conducteurs de 2RM et les automobilistes selon différentes situations accidentelles (en particulier en intersection), de quantifier les effets des facteurs à l'origine de la survenue des accidents issus de la perte de contrôle du 2RM, et de mesurer l'efficacité des dispositifs de protection portés par les usagers de 2RM (casque intégral, blouson de moto, pantalon de moto, gants, bottes ou chaussures montantes et protection dorsale). Données : la réalisation des objectifs de la thèse s'appuie sur les données du projet VOIESUR, les données du Registre des accidents de la circulation du Rhône et les données recueillies au moyen d'une enquête postale réalisée auprès de 7 148 usagers de 2RM accidentés entre 2010 et 2014 identifiés par le Registre du Rhône. Résultats : Le problème de détectabilité des motos au niveau des intersections est confirmé. Dans la même situation d'interaction entre deux véhicules en intersection, un automobiliste a plus de difficulté à détecter une moto qu'à détecter un véhicule léger. Les facteurs suivants sont associés à la survenue des accidents de type perte de contrôle : alcool, type de moto, jour de l'accident, sinuosité de la route, adhérence de la route et vitesse. Nous soulignons le fait qu'une mauvaise adhérence de la route augmente considérablement le risque de perte de contrôle, en particulier lorsque la dégradation de l'adhérence est inattendue pour les motocyclistes en raison de la présence de gravillons, de corps gras, de nids de poule, etc. sur la route. En cas d'accident, les vêtements dédiés à la moto protègent les usagers des blessures. Ils sont efficaces pour éviter les dermabrasions et les plaies cutanées, mais ne semblent pas protéger des blessures plus graves telles que fracture, luxation ou entorse, sauf pour les bottes ou chaussures montantes qui réduisent le risque de fracture de la cheville ou du pied. Concernant les plaques dorsales, aucune efficacité n'a été montrée. Les casques intégraux protègent mieux le visage de l'usager du 2RM comparés aux non-intégraux, alors qu'aucune différence de protection du crâne ou du cerveau n'est mise en évidence entre les deux types de casques. Conclusion : Nos résultats mettent en évidence le problème de détectabilité d'un 2RM par les automobilistes en intersection et l'importance de l'état de la chaussée dans les accidents associés à une perte de contrôle du 2RM par son conducteur. Il est fortement recommandé pour les automobilistes de bien faire attention aux véhicules prioritaires venant de la gauche ou d'en face avant de tourner à gauche, particulièrement aux 2RM venant d'en face. L'amélioration de l'état de la chaussée et une signalisation adaptée représentent aussi un enjeu fort pour la sécurité des usagers de 2RM. Nos résultats soutiennent la recommandation faite par les organismes de sécurité routière du port plus systématique des équipements vestimentaires chez les usagers de 2RM. Enfin, nos résultats encouragent l'utilisation d'un casque intégral chez les usagers de 2RM afin de mieux protéger le visage Background: Motorized two-wheelers (MTW) are evolving rapidly and the vulnerability of their users is becoming more obvious. MTW users were the most vulnerable road users, accounting for 23 % of all road traffic fatalities worldwide in 2013, in front of pedestrians (22 %). In this context, the safety of MTW users has become a serious concern in most countries. In contrast to many public health problems, the main factors of MTW road safety have been identified. The issue today is to quantify their effects on the occurrence of accidents and to measure the effects of protective devices. The present thesis aims to better understand the interaction problems between MTW drivers and automobilists according to different accident configurations (in particular at intersection), to identify and assess the effect of critical factors on the risk of MTW loss-of-control crashes, and to measure the effectiveness of protective clothing (motorcycle jacket, trousers, gloves, knee-high or ankle boots, back protection) for MTW users. Data: This study used MTW accident data collected in a French project VOIESUR, injury data from the Rhône Registry, plus data collected by means of a postal survey which was conducted among 7148 MTW riders injured between 2010 and 2014 and identified in the Rhône Registry. Results: We confirmed the problem of motorcycle detection for other road users. In the similar conditions, the motorists have more difficulty to detect a motorcycle than a car oncoming, in particular at intersections. The following factors are associated with the occurrence of motorcycle loss-of-control accidents: alcohol use, motorcycle type, weekend vs. weekday, road alignment, road adhesion and traveling speed. We emphasize that poor road adhesion significantly increases the risk of losing control, especially when deteriorated road adhesion is encountered unexpectedly, due to the presence of loose gravel, ice, oil, potholes, etc. on the roadway. In case of accident, motorcycle clothing protects users from injury. It can protect riders against injuries such as dermabrasion and laceration, but not against more serious injuries, such as fracture and sprain, except for boots, which reduce foot-and-ankle fracture risk. No effect of dorsal protectors was shown. Full-face helmets provide significantly greater protection against facial injury than do other helmets. However, no significant difference of protection against skull or brain injury is found between the two types of helmets. Conclusion: Our results highlight the problem of MTW detection for motorists at intersections and the important role of road conditions in the occurrence of accidents resulting from loss of control of MTW. It is recommended for motorists to pay extra attention to priority vehicles oncoming from their left or the opposite direction, before turning left at an intersection, especially to MTW. Regular road maintenance and immediate installment of appropriate warning signs concerning road deterioration are also highly recommended in consideration of MTW safety. Finally, our results support road safety organizations’ recommendation that protective clothing be worn: this should be more systematic for MTW users, the same for the use of full-face helmets

APA, Harvard, Vancouver, ISO, and other styles

28

Weerasuriya, Sujeeva A. "An application of artificial neural networks in freeway incident detection." Scholar Commons, 1998. http://purl.fcla.edu/fcla/etd/SFE0000011.

Full text

APA, Harvard, Vancouver, ISO, and other styles

29

Powell, Ronald. "Evaluating cognitive screening as a possible solution to reducing accidents and improving workplace productivity through early preventive detection of fatigue-impairment in the construction industry." Thesis, University of Bath, 2012. https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.558907.

Full text

Abstract:

Fatigue is emerging as a significant concern in the workplace principally focused on its relationship to accidents and lost productivity. Construction work exposes workers to many hazards and if safety programmes are not effective, accidents will result. Based on the sector’s safety performance, workers are not being adequately protected and improvement is needed. Fatigue-related impairment has been identified as a subject of concern for all workplaces yet it is not yet a focus within construction and few operational studies have been undertaken to develop tools to assist with identification and control of this workplace impairment. This research started with an assessment of the management of impairment within the global construction industry as well as an evaluation of tools that might assist in identification and classification of fatigue levels. In particular, cognitive tests were studied and shown to have sensitivity to natural changes in alertness in an operational setting. A small battery of cognitive tests was compared and showed that cognitive tests based on reaction times were possible candidates to help identify fatigue-related impairment in real time. The top performing tests were then used as possible surrogate measures for fatigue. To finally assess their performance capability their output was compared to estimations from an advanced actigraph-fed fatigue model. 100 volunteer workers each wore an actigraph for a month each to collect information on their personal sleep/wake cycles and activity whilst periodically doing the cognitive tests. The data from the actigraphs was analyzed by proprietary software to determine individual performance effectiveness. It was found that output from these simple, quick, and low cost tests significantly correlated with the most advanced actigraph-fed fatigue model. It is concluded that cognitive tests can be used as screens for fatigue-related impairment in the workplace. All primary parameters used for modelling showed extremely high significance (Pr( >Chisq) < 2.2e-16) in correlation to fatigue-based effectiveness results and could be developed into a screening tool for fatigue-related impairment in the construction industry as part of a fatigue management programme.

APA, Harvard, Vancouver, ISO, and other styles

30

Consuegra, Rengifo Nathan Adolfo. "Detection and Classification of Anomalies in Road Traffic using Spark Streaming." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-238733.

Full text

Abstract:

Road traffic control has been around for a long time to guarantee the safety of vehicles and pedestrians. However, anomalies such as accidents or natural disasters cannot be avoided. Therefore, it is important to be prepared as soon as possible to prevent a higher number of human losses. Nevertheless, there is no system accurate enough that detects and classifies anomalies from the road traffic in real time. To solve this issue, the following study proposes the training of a machine learning model for detection and classification of anomalies on the highways of Stockholm. Due to the lack of a labeled dataset, the first phase of the work is to detect the different kind of outliers that can be found and manually label them based on the results of a data exploration study. Datasets containing information regarding accidents and weather are also included to further expand the amount of anomalies. All experiments use real world datasets coming from either the sensors located on the highways of Stockholm or from official accident and weather reports. Then, three models (Decision Trees, Random Forest and Logistic Regression) are trained to detect and classify the outliers. The design of an Apache Spark streaming application that uses the model with the best results is also provided. The outcomes indicate that Logistic Regression is better than the rest but still suffers from the imbalanced nature of the dataset. In the future, this project can be used to not only contribute to future research on similar topics but also to monitor the highways of Stockholm. Vägtrafikkontroll har funnits länge för att garantera säkerheten hos fordon och fotgängare. Emellertid kan avvikelser som olyckor eller naturkatastrofer inte undvikas. Därför är det viktigt att förberedas så snart som möjligt för att förhindra ett större antal mänskliga förluster. Ändå finns det inget system som är noggrannt som upptäcker och klassificerar avvikelser från vägtrafiken i realtid. För att lösa detta problem föreslår följande studie utbildningen av en maskininlärningsmodell för detektering och klassificering av anomalier på Stockholms vägar. På grund av bristen på en märkt dataset är den första fasen av arbetet att upptäcka olika slags avvikare som kan hittas och manuellt märka dem utifrån resultaten av en datautforskningsstudie. Dataset som innehåller information om olyckor och väder ingår också för att ytterligare öka antalet anomalier. Alla experiment använder realtidsdataset från antingen sensorerna på Stockholms vägar eller från officiella olyckor och väderrapporter. Därefter utbildas tre modeller (beslutsträd, slumpmässig skog och logistisk regression) för att upptäcka och klassificera outliersna. Utformningen av en Apache Spark streaming-applikation som använder modellen med de bästa resultaten ges också. Resultaten tyder på att logistisk regression är bättre än resten men fortfarande lider av datasetets obalanserade natur. I framtiden kan detta projekt användas för att inte bara bidra till framtida forskning kring liknande ämnen utan även att övervaka Stockholms vägar.

APA, Harvard, Vancouver, ISO, and other styles

31

Didier, Romain. "Athérosclérose : approche translationnelle de la détection, caractérisation, et du traitement de la plaque athéromateuse." Thesis, Brest, 2018. http://www.theses.fr/2018BRES0030.

Full text

Abstract:

L’athérosclérose est une pathologie artérielle chronique évoluant sur plusieurs années ou dizaines d’années. Sa traduction clinique est variable, pouvant rester longtemps asymptomatique, ou induire des symptômes d’effort directement lié à la réduction du calibre de l’artère (angor, claudication des membres inférieurs). Malgré cette évolution en apparence lente et progressive, les manifestations cliniques aiguës de cette pathologie sont souvent de survenue brutale. La formation de thrombus en regard des lésions athéromateuses dites déstabilisées représente la composante physiopathologique la plus fréquemment rencontrée. Cette complication aigüe de la plaque athéromateuse est responsable d’événements cardio-vasculaires majeurs impactant la morbi-mortalité, représentés par les syndromes coronariens aigus, les accidents vasculaires cérébraux ischémiques, et les ischémies aigues de membre inférieur. Après une revue des connaissances actuelles sur la plaque athéromateuse, nous étudierons à l’aide d’un modèle animal d’athérosclérose, son évolution, ses caractéristiques morphologiques, et nous testerons l’impact d’un traitement au long cours par statine en prévention primaire sur les plaques athéromateuses et sur les composants de la paroi vasculaire. Puis, dans une approche clinique, nous nous intéresserons à la durée optimale du traitement par double antiagrégant plaquettaire instauré après un traitement par angioplastie percutanée de plaque athéromateuse. Dans un second temps, nous détaillerons l’évolution des pratiques d’angioplastie coronaire en analysant les facteurs ayant contribué à limiter l’utilisation des stents non actifs (dit nus) au cours des 10 dernières années. Enfin, nous étudierons les principaux facteurs restant associés à la survenue d’un accident vasculaire cérébral post angioplastie percutanée, correspondant à l’une des principales complications majeures de ces procédures Atherosclerosis is a chronic arterial disease that progresses over several years or decades. Its clinical translation is variable, and may remain asymptomatic for a long time, or induce symptoms of stress directly related to the reduction in size of the artery (angina, chronic limb ischemia). Despite this seemingly slow and progressive evolution, the acute clinical manifestations of this pathology often occur suddenly. Thrombus formation in regard to the “destabilized” atheromatous lesions is the most frequently physiopathological components. This acute complication of atheromatous plaque is responsible of major cardiovascular events impacting the morbi-mortality, mostly represented by acute coronary syndromes, ischemic strokes, and acute lower limb ischemia. After a review of current knowledge on atheromatous plaque, we will study using an animal model of atherosclerosis, its evolution, its morphological characteristics, and we will test the impact of a long-term statin treatment in primary prevention on atheromatous plaques and on vascular wall components. Then, in a clinical approach, we will look at the optimal duration of double platelet antiaggregant after angioplasty. In a second step, we will detail the evolution of coronary angioplasty practices by analyzing the main factors that have contributed to limiting the use of bare metal stent over the past 10 years. Finally, we will analyze the major factors still associated with the occurrence of a stroke after percutaneous angioplasty

APA, Harvard, Vancouver, ISO, and other styles

32

Dakkoune, Amine. "Méthodes pour l'analyse et la prévention des risques d'emballement thermique Zero-order versus intrinsic kinetics for the determination of the time to maximum rate under adiabatic conditions (TMR_ad): application to the decomposition of hydrogen peroxide Risk analysis of French chemical industry Fault detection in the green chemical process : application to an exothermic reaction Analysis of thermal runaway events in French chemical industry Early detection and diagnosis of thermal runaway reactions using model-based approaches in batch reactors." Thesis, Normandie, 2019. http://www.theses.fr/2019NORMIR30.

Full text

Abstract:

L’histoire des événements accidentels dans les industries chimiques montre que leurs conséquences sont souvent graves sur les plans humain, environnemental et économique. Cette thèse vise à proposer une approche de détection et de diagnostic des défauts dans les procédés chimiques afin de prévenir ces événements accidentels. La démarche commence par une étude préalable qui sert à identifier les causes majeures responsables des événements industriels chimiques en se basant sur le retour d’expérience (REX). En France, selon la base de données ARIA, 25% des évènements sont dus à l’emballement thermique à cause d’erreurs d’origine humaine. Il est donc opportun de développer une méthode de détection et de diagnostic précoce des défauts dus à l’emballement thermique. Pour cela nous développons une approche qui utilise des seuils dynamiques pour la détection et la collecte de mesures pour le diagnostic. La localisation des défauts est basée sur une classification des caractéristiques statistiques de la température en fonction de plusieurs modes défectueux. Un ensemble de classificateurs linéaires et de diagrammes de décision binaires indexés par rapport au temps sont utilisés. Enfin, la synthèse de l'acide peroxyformique dans un réacteur discontinu et semi-continu est considérée pour valider la méthode proposée par des simulations numériques et ensuite expérimentales. Les performances de détection de défauts se sont révélées satisfaisantes et les classificateurs ont démontré un taux de séparabilité des défauts élevés The history of accidental events in chemical industries shows that their human, environmental and economic consequences are often serious. This thesis aims at proposing an approach of detection and diagnosis faults in chemical processes in order to prevent these accidental events. A preliminary study serves to identify the major causes of chemical industrial events based on experience feedback. In France, according to the ARIA database, 25% of the events are due to thermal runaway because of human errors. It is therefore appropriate to develop a method for early fault detection and diagnosis due to thermal runaway. For that purpose, we develop an approach that uses dynamical thresholds for the detection and collection of measurements for diagnosis. The localization of faults is based on a classification of the statistical characteristics of the temperature according to several defectives modes. A multiset of linear classifiers and binary decision diagrams indexed with respect to the time are used for that purpose. Finally, the synthesis of peroxyformic acid in a batch and semi batch reactor is considered to validate the proposed method by numerical simulations and then experiments. Faults detection performance has been proved satisfactory and the classifiers have proved a high isolability rate of faults

APA, Harvard, Vancouver, ISO, and other styles

33

gupta, Devansh. "Smart-Scooter Rider Assistance System using Internet of Wearable Things and Computer Vision." Case Western Reserve University School of Graduate Studies / OhioLINK, 2021. http://rave.ohiolink.edu/etdc/view?acc_num=case1619611136736967.

Full text

APA, Harvard, Vancouver, ISO, and other styles

34

Chaccour, Kabalan. "Elaborating the Actimetric Profile of Fall Sensitive Patients for Early Detection of Fall Incidents." Thesis, Bourgogne Franche-Comté, 2017. http://www.theses.fr/2017UBFCA016/document.

Full text

Abstract:

La croissance et le vieillissement sont inévitables pour la race humaine. Chez les personnes âgées, le vieillissement est souvent accompagné par de nombreuses formes de maladies et de dangers dont les chutes qui affectent la qualité de vie et qui posent un enjeu socio-économique. Mais les chutes sont évitables. Les acteurs de santé, les scientifiques et les chercheurs combinent actuellement des efforts pour développer des systèmes de détection et de prédiction des chutes. Dans le contexte de la prédiction des chutes, l'objectif de cette thèse est d'élaborer le profile actimétrique des patients sensibles aux chutes afin de les alerter d'une possible chute. Ceci consiste principalement à développer un système capable de surveiller les paramètres de la marche des personnes durant leurs activités quotidiennes avec un minimum d'intrusivité. Dans une première contribution, nous avons proposé une classification générique des systèmes liés à la chute en fonction du déploiement de leurs capteurs. Nous avons distingué les systèmes portables, les systèmes non-portables et les systèmes qui combinent les deux. En se basant sur cette classification, nous avons proposé notre plateforme WMFL v1.0 dans une deuxième contribution. WMFL combine une chaussure équipée par des capteurs de force avec des dalles où nous avons intégrés des capteurs optiques infrarouges. La fusion de ces deux systèmes assure une prévention à l'intérieure et à l'extérieure des locaux. WMFL peut être aussi déployées dans une clinique. Dans une troisième contribution, nous avons proposé une méthode de prédiction des chutes en se basant sur l'analyse du déplacement du centre de pression (projeté du centre de masse) sur la surface plantaire du pied durant la marche. La méthode utilise la fenêtre glissante spatio-temporelle pour alerter le patient d'une chute potentielle et pour déterminer le risque de chute à la fin de la marche Growth is the normal change of the human body and getting old is inevitable to human race. As a result, elderly people are subject to many forms of diseases and dangers among which falls are considered very serious in terms of quality of life and socio-economic costs. But falls can be manageable. Health practitioners, scientists and researchers currently combine efforts to develop systems capable of detecting and predicting falls. In the context of fall prediction, the goal of this thesis is to elaborate the actimetric profile of fall sensitive patients to alert them from a potential fall. It mainly consists of developing a system capable of monitoring gait and balance parameters during their daily activities with minimum intrusiveness. These are usually assessed in clinical settings using high-cost tools. In our first contribution, we proposed a generic classification of fall-related systems based on their sensors deployment. These are classified as Wearable, Non-Wearable and Fusion Systems. Based on the generic classification, we proposed the WMFL v1.0 platform in our second contribution. WMFL fuses a Foot Wear Force Sensing device with an Ambient system using IR-sensing floor tiles. The platform can be deployed at homes or in clinics. It ensures an indoor-outdoor protection. In a third contribution, we proposed an early fall detection approach to determine the risk of falling by analyzing the displacement of the Center of Pressure projecting the amount of sway of the Center of Mass on the foot plantar surface. The method uses the spatio-temporal sliding window to alert the patient of a potential fall

APA, Harvard, Vancouver, ISO, and other styles

35

Wang, Bihao. "Geometrical and contextual scene analysis for object detection and tracking in intelligent vehicles." Thesis, Compiègne, 2015. http://www.theses.fr/2015COMP2197/document.

Full text

Abstract:

Pour les véhicules intelligents autonomes ou semi-autonomes, la perception constitue la première tâche fondamentale à accomplir avant la décision et l’action. Grâce à l’analyse des données vidéo, Lidar et radar, elle fournit une représentation spécifique de l’environnement et de son état, à travers l’extraction de propriétés clés issues des données des capteurs. Comparé à d’autres modalités de perception telles que le GPS, les capteurs inertiels ou les capteurs de distance (Lidar, radar, ultrasons), les caméras offrent la plus grande quantité d’informations. Grâce à leur polyvalence, les caméras permettent aux systèmes intelligents d’extraire à la fois des informations contextuelles de haut niveau et de reconstruire des informations géométriques de la scène observée et ce, à haute vitesse et à faible coût. De plus, la technologie de détection passive des caméras permet une faible consommation d’énergie et facilite leur miniaturisation. L’utilisation des caméras n’est toutefois pas triviale et pose un certain nombre de questions théoriques liées à la façon dont ce capteur perçoit son environnement. Dans cette thèse, nous proposons un système de détection d’objets mobiles basé seule- ment sur l’analyse d’images. En effet, dans les environnements observés par un véhicule intelligent, les objets en mouvement représentent des obstacles avec un risque de collision élevé, et ils doivent être détectés de manière fiable et robuste. Nous abordons le problème de la détection d’objets mobiles à partir de l’extraction du contexte local reposant sur une segmentation de la route. Après transformation de l’image couleur en une image invariante à l’illumination, les ombres peuvent alors être supprimées réduisant ainsi leur influence négative sur la détection d’obstacles. Ainsi, à partir d’une sélection automatique de pixels appartenant à la route, une région d’intérêt où les objets en mouvement peuvent apparaître avec un risque de collision élevé, est extraite. Dans cette zone, les pixels appartenant à des objets mobiles sont ensuite identifiés à l’aide d’une approche plan+parallaxe. À cette fin, les pixels potentiellement mobiles et liés à l’effet de parallaxe sont détectés par une méthode de soustraction du fond de l’image; puis trois contraintes géométriques différentes: la contrainte épipolaire, la contrainte de cohérence structurelle et le tenseur trifocal, sont appliquées à ces pixels pour filtrer ceux issus de l’effet de parallaxe. Des équations de vraisemblance sont aussi proposées afin de combiner les différents contraintes d’une manière complémentaire et efficace. Lorsque la stéréovision est disponible, la segmentation de la route et la détection d’obstacles peuvent être affinées en utilisant une segmentation spécifique de la carte de disparité. De plus, dans ce cas, un algorithme de suivi robuste combinant les informations de l’image et la profondeur des pixels a été proposé. Ainsi, si l’une des deux caméras ne fonctionne plus, le système peut donc revenir dans un mode de fonctionnement monoculaire ce qui constitue une propriété importante pour la fiabilité et l’intégrité du système de perception. Les différents algorithmes proposés ont été testés sur des bases de données d’images publiques en réalisant une évaluation par rapport aux approches de l’état de l’art et en se comparant à des données de vérité terrain. Les résultats obtenus sont prometteurs et montrent que les méthodes proposées sont efficaces et robustes pour différents scénarios routiers et les détections s’avèrent fiables notamment dans des situations ambiguës For autonomous or semi-autonomous intelligent vehicles, perception constitutes the first fundamental task to be performed before decision and action/control. Through the analysis of video, Lidar and radar data, it provides a specific representation of the environment and of its state, by extracting key properties from sensor data with time integration of sensor information. Compared to other perception modalities such as GPS, inertial or range sensors (Lidar, radar, ultrasonic), the cameras offer the greatest amount of information. Thanks to their versatility, cameras allow intelligent systems to achieve both high-level contextual and low-level geometrical information about the observed scene, and this is at high speed and low cost. Furthermore, the passive sensing technology of cameras enables low energy consumption and facilitates small size system integration. The use of cameras is however, not trivial and poses a number of theoretical issues related to how this sensor perceives its environmen. In this thesis, we propose a vision-only system for moving object detection. Indeed,within natural and constrained environments observed by an intelligent vehicle, moving objects represent high risk collision obstacles, and have to be handled robustly. We approach the problem of detecting moving objects by first extracting the local contextusing a color-based road segmentation. After transforming the color image into illuminant invariant image, shadows as well as their negative influence on the detection process can be removed. Hence, according to the feature automatically selected onthe road, a region of interest (ROI), where the moving objects can appear with a high collision risk, is extracted. Within this area, the moving pixels are then identified usin ga plane+parallax approach. To this end, the potential moving and parallax pixels a redetected using a background subtraction method; then three different geometrical constraints : the epipolar constraint, the structural consistency constraint and the trifocaltensor are applied to such potential pixels to filter out parallax ones. Likelihood equations are also introduced to combine the constraints in a complementary and effectiveway. When stereo vision is available, the road segmentation and on-road obstacles detection can be refined by means of the disparity map with geometrical cues. Moreover, in this case, a robust tracking algorithm combining image and depth information has been proposed. If one of the two cameras fails, the system can therefore come back to a monocular operation mode, which is an important feature for perception system reliability and integrity. The different proposed algorithms have been tested on public images data set with anevaluation against state-of-the-art approaches and ground-truth data. The obtained results are promising and show that the proposed methods are effective and robust on the different traffic scenarios and can achieve reliable detections in ambiguous situations

APA, Harvard, Vancouver, ISO, and other styles

36

Quadros, Thiago de. "Development and evaluation of an elderly fall detection system based on a wearable device located at wrist." Universidade Tecnológica Federal do Paraná, 2017. http://repositorio.utfpr.edu.br/jspui/handle/1/2619.

Full text

Abstract:

A queda de idosos é um problema de saúde mundial. Todos os anos, cerca de 30% dos idosos com 65 anos ou mais são vítimas de quedas. Além disso, as consequências de uma queda podem ser fisiológicas (e.g. fraturas ósseas, ferimentos musculares) e psicológicas, como a perda de autoconfiança, levando a novas quedas. Uma solução para este problema está relacionada com ações preventivas (e.g. adaptação de mobília) aliadas a sistemas de detecção de quedas, os quais podem notificar familiares e serviços médicos de urgência. Como o tempo de espera por socorro após uma queda está relacionado com a severidade das consequências dela, esses sistemas devem oferecer elevada acurácia e detecção em tempo real. Embora existam várias soluções para isso na literatura (a maioria relacionada com dispositivos vestíveis), poucas delas estão relacionadas a dispositivos de punho, principalmente por causa dos desafios existentes para essa configuração. Considerando o punho como um local mais confortável, discreto e aceitável para uso de um dispositivo (menos associado com o estigma do uso de uma solução médica), este trabalho propõe o desenvolvimento e avaliação de uma solução baseada nessa configuração. Para isso, diferentes sensores (acelerômetro, giroscópio e magnetômetro) foram combinados com diferentes algoritmos, baseados em métodos de limiar e aprendizado de máquina, visando definir os melhores sinais e abordagem para a detecção de quedas. Esses métodos consideraram informações de aceleração, velocidade, deslocamento e orientação espacial, permitindo o cálculo de componentes verticais do movimento. Para o treino e avaliação dos algoritmos, dois protocolos diferentes foram empregados: um primeiro envolvendo 2 voluntários (homens, 27 e 31 anos) simulando um total de 80 sinais de queda e 80 de não-queda, e um segundo envolvendo 22 voluntários (14/8 homens/mulheres, idade média: 25,2 ± 4,7) simulando um total de 396 sinais de queda e 396 de não-queda. Uma análise exaustiva de diferentes sinais e parâmetros de configuração foi executada para cada método. O melhor algoritmo baseado em limiar considerou sinais de aceleração vertical e velocidade total, alcançando 95,8% de sensibilidade e 86,5% de especificidade. Por outro lado, o melhor algoritmo de aprendizagem de máquina foi o baseado no método K-Nearest Neighbors, considerando informações de aceleração, velocidade e deslocamento verticais combinadas com os ângulos de orientação espacial: 100% de sensibilidade e 97,9% de especificidade. Os resultados obtidos permitem enfatizar a relevância de algoritmos de aprendizagem de máquina para sistemas de detecção de queda vestíveis localizados no punho quando comparados a algoritmos baseados em limiar. Esta conclusão oferece grande contribuição para a pesquisa de detectores de quedas similares, sugerindo a melhor abordagem para novos desenvolvimentos. Falls in the elderly age are a world health problem. Every year, about 30% of people aged 65 or older become victims of fall events. The consequences of a fall may be physiological (e.g. bone fractures, muscular injuries) and psychological, including the loss of self-confidence by fear of falling, which leads to new falls. A solution to this problem is related to preventive actions (e.g. adapting furniture) allied to fall detection systems, which can alert family members and emergency medical services. Since the response time for help is related to the fall's consequences and severity, such systems must offer high accuracy and real-time fall detection. Although there are many fall detection solutions in literature (most part of them related to wearable devices), few of them are related to wrist-worn devices, mainly because of the existing challenges for this configuration. Considering the wrist as a comfortable, discrete and acceptable place for an elderly wearable device (less associated to the stigma of using a medical device), this work proposes the development and evaluation of a fall detection solution based on this configuration. For this, different sensors (accelerometer, gyroscope and magnetometer) were combined to different algorithms, based on threshold and machine learning methods, in order to define the best signals and approach for an elderly fall detection. These methods considered acceleration, velocity and displacement information, relating them with wrist spatial orientation, allowing the calculation of the vertical components of each movement. For the algorithms' training and evaluation, two different protocols were employed: one involving 2 volunteers (both males, ages of 27 and 31) performing a total of 80 fall and 80 non-fall events simulation, and the other involving 22 volunteers (14/8 males/females, ages mean: 25.2 ± 4.7) performing a total of 396 fall and 396 non-fall events simulation. An exhaustive evaluation of different signals and configuration parameters was performed for each method. The best threshold-based algorithm employed the vertical acceleration and total velocity signals, achieving 95.8% and 86.5% of sensitivity and specificity, respectively. On the other hand, the best machine learning algorithm was based on the K-Nearest Neighbors method employing the vertical acceleration, velocity and displacement information combined with spatial orientation angles: 100% of sensitivity and 97.9% of specificity. The obtained results allow to emphasize the relevance of machine learning algorithms for wrist-worn fall detection systems instead of traditional threshold-based algorithms. These results offer great contributions for the research of similar wearable fall detectors, suggesting the best approach for new developments.

APA, Harvard, Vancouver, ISO, and other styles

37

Podola, David. "Systém pro asistenci při nepřehledných dopravních situacích." Master's thesis, Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií, 2019. http://www.nusl.cz/ntk/nusl-400960.

Full text

Abstract:

T-intersections are one of the most common places where collisions happen. An intelligent traffic mirror is one the possible solutions to reduce the accident rate. The mirror detects the situation around the intersection, process the data and provides the driver with an information, whether the situation is safe and the driver can enter the junction safely. The aim of the thesis is a feasibility study of reliable detection of non-stationary objects based on cameras. The core of the intended product – the detection algorithm – detected the object on short distance from the camera reliably but as the distance was growing, the detection quality degraded. One of the possible solutions to achieve better detection results on longer distances may be achieved by using a camera with greater zoom. Based on the example improvement proposal, the feasibility of the solution based on optical methods was finally confirmed.

APA, Harvard, Vancouver, ISO, and other styles

38

Mousse, Ange Mikaël. "Reconnaissance d'activités humaines à partir de séquences multi-caméras : application à la détection de chute de personne." Thesis, Littoral, 2016. http://www.theses.fr/2016DUNK0453/document.

Full text

Abstract:

La vision artificielle est un domaine de recherche en pleine évolution. Les nouvelles stratégies permettent d'avoir des réseaux de caméras intelligentes. Cela induit le développement de beaucoup d'applications de surveillance automatique via les caméras. Les travaux développés dans cette thèse concernent la mise en place d'un système de vidéosurveillance intelligente pour la détection de chutes en temps réel. La première partie de nos travaux consiste à pouvoir estimer de façon robuste la surface d'une personne à partir de deux (02) caméras ayant des vues complémentaires. Cette estimation est issue de la détection de chaque caméra. Dans l'optique d'avoir une détection robuste, nous avons fait recours à deux approches. La première approche consiste à combiner un algorithme de détection de mouvements basé sur la modélisation de l'arrière plan avec un algorithme de détection de contours. Une approche de fusion a été proposée pour rendre beaucoup plus efficiente le résultat de la détection. La seconde approche est basée sur les régions homogènes de l'image. Une première ségmentation est effectuée dans le but de déterminer les régions homogènes de l'image. Et pour finir, nous faisons la modélisation de l'arrière plan en se basant sur les régions. Une fois les pixels de premier plan obtenu, nous faisons une approximation par un polygone dans le but de réduire le nombre d'informations à manipuler. Pour l'estimation de cette surface nous avons proposé une stratégie de fusion dans le but d'agréger les détections des caméras. Cette stratégie conduit à déterminer l'intersection de la projection des divers polygones dans le plan de masse. La projection est basée sur les principes de l'homographie planaire. Une fois l'estimation obtenue, nous avons proposé une stratégie pour détecter les chutes de personnes. Notre approche permet aussi d'avoir une information précise sur les différentes postures de l'individu. Les divers algorithmes proposés ont été implémentés et testés sur des banques de données publiques dans le but de juger l'efficacité des approches proposées par rapport aux approches existantes dans l'état de l'art. Les résultats obtenus et qui ont été détaillés dans le présent manuscrit montrent l'apport de nos algorithmes Artificial vision is an involving field of research. The new strategies make it possible to have some autonomous networks of cameras. This leads to the development of many automatic surveillance applications using the cameras. The work developed in this thesis concerns the setting up of an intelligent video surveillance system for real-time people fall detection. The first part of our work consists of a robust estimation of the surface area of a person from two (02) cameras with complementary views. This estimation is based on the detection of each camera. In order to have a robust detection, we propose two approaches. The first approach consists in combining a motion detection algorithm based on the background modeling with an edge detection algorithm. A fusion approach has been proposed to make much more efficient the results of the detection. The second approach is based on the homogeneous regions of the image. A first segmentation is performed to find homogeneous regions of the image. And finally we model the background using obtained regions

APA, Harvard, Vancouver, ISO, and other styles

39

Fayad, Moustafa. "Health care platform development based on multimedia sensors." Electronic Thesis or Diss., Université Paris Cité, 2021. http://www.theses.fr/2021UNIP5204.

Full text

Abstract:

Tous les pays sont confrontés à des changements démographiques importants à une échelle sans précédents. Les seniors constituent le segment de la population mondiale qui augmente le plus rapidement. Ceci entraine une augmentation des personnes âgées dépendantes et vulnérables marquées par une perte d’autonomie et maladie chronique. Ces complications augmentent les taux de mortalité et morbidité dans nos sociétés. Ces dernières années, les conséquences socio-économiques sur les seniors et sa famille, et le fait des tarifs élevées dans les établissements spécialisés d’hébergement ont attiré l’attention sur le maintien à domicile en servant des nouvelles technologies fiables et pas cher. Mais développer un système de soins de santé critique pour les personnes âgées est un véritable défi qui ne peut être relevé qu’en garantissant la fiabilité et la haute performance. Dans le cadre du futur système de soins de santé nommée Family Heroes, nous sommes intéressées dans cette thèse à sa fiabilité et sa sécurité. Ainsi, Nous avons étudié la détection automatique des chutes à l’aide de la caméra Kinect. Plus précisément, nous avons proposé une méthodologie combinant la méthode CATWOE et des notions UML pour analyser et modéliser notre système. Nous avons montré les étapes à suivre pour prendre en compte efficacement les exigences et les contraintes du système dans le processus de développement. Ainsi, nous avons utilisé les concepts MARTE pour faire face aux contraintes de temps dans la phase de modélisation. Ensuite, pour assurer la fiabilité de notre système, nous avons proposé une approche de vérification formelle basée sur les modèles UML/MARTE présentés dans l'étape de modélisation. Nous les avons modélisés avec des automates temporisés et déployé le vérificateur de modèles UPPAAL pour spécifier et vérifier les propriétés. Pour assurer la sécurité de notre système, nous avons présenté une approche formelle pour traiter les problèmes de cyber-attaques qui peuvent survenir dans le système Family Heroes. Ainsi, nous avons adopté un scénario de cyberattaques liés à ce contexte et proposé de reconfigurer notre système pour faire face aux cyberattaques. De plus, une vérification formelle a été proposée pour s'assurer de la fiabilité de la reconfiguration proposée du système. Les résultats ont montré le respect des propriétés (L’absence de deadlock, l’accessibilité, la sécurité et la vivacité) dans Family Heroes.Pour la détection automatique des chutes, nous avons utilisé le capteur multimédia de la technologie Kinect. Nous avons proposé une nouvelle approche basée sur des seuils qui analyse la variation de hauteur, les angles du haut du corps au cours du cycle de mouvement, et l'inactivité de la personne au sol. De plus, nous avons proposé un nouveau modèle léger basé sur Deep Learning LSTM utilisant des caractéristiques géométriques. Ce modèle léger est conçu pour fonctionner sur des appareils limités comme ceux de Raspberry Pi. Les résultats sont très prometteurs All countries are facing significant demographic changes on an unprecedented scale. Seniors are the fastest-growing segment of the world's population. It increases dependent and vulnerable older adults marked by loss of autonomy and chronic illness. These complications increase mortality and morbidity rates in our societies. In recent years, the socio-economic consequences on seniors and their families and the high prices in specialized accommodation establishments have drawn attention to home support by using new reliable and inexpensive technologies. However, developing a critical healthcare system for the elderly is a real challenge that can only be met by ensuring reliability and high performance. Within the framework of the future health care system named Family Heroes, we are interested in this thesis in its reliability and safety. Thus, we studied the automatic detection of falls using the Kinect camera. More precisely, we have proposed a methodology combining the CATWOE method and UML notions to analyze and model our system. We have shown the steps to take to consider system requirements and constraints in the development process effectively. Thus, we used the MARTE concepts to face the time constraints in the modeling phase. Then, to ensure the reliability of our system, we proposed a formal verification approach based on the UML / MARTE models presented in the modeling stage. We modeled them with timed automata and deployed the UPPAAL model checker to specify and verify properties. To ensure the security of our system, we have presented a formal approach to deal with cyber-attack issues that may arise in the Family Heroes system. Thus, we adopted a scenario of cyberattacks linked to this context and proposed reconfiguring our system to face cyberattacks. In addition, formal verification has been proposed to ensure the reliability of the proposed solution. The results showed the respect of the properties (No deadlock, accessibility, security, and liveliness) in Family Heroes.For automatic fall detection, we used the multimedia sensor of Kinect technology. We have proposed a new approach based on thresholds that analyzes the variation in height, the angles of the upper body during the cycle of movement, and the person's inactivity on the ground. In addition, we have proposed a new lightweight model based on Deep Learning LSTM using geometric features. This lightweight model is designed to work on limited devices like the Raspberry Pi. The results are very promising

APA, Harvard, Vancouver, ISO, and other styles

40

Shaafi, Aymen. "Secured and trusted remote wireless health monitoring systems for assisted living of elderly people." Electronic Thesis or Diss., Université Paris Cité, 2021. http://www.theses.fr/2021UNIP5208.

Full text

Abstract:

Le vieillissement de la population est l'un des problèmes clés pour la grande majorité de nombreux pays. Le nombre de personnes âgées souffrant de multiples maladies et nécessitant une surveillance continue de leurs signes vitaux augmente chaque jour, entraînant des coûts de santé supplémentaires. Les systèmes de santé modernes en médecine gériatrique nécessitent souvent la présence de personnes âgées à l'hôpital, ce qui est en conflit avec leur exigence d'indépendance et d'intimité. Les développements récents sur la télésurveillance e-santé offrent une large gamme de solutions. Cependant, la plupart des appareils sont conçus pour une détection médicale spécifique et fonctionnent indépendamment les uns des autres. Il y a toujours un manque de cadre intégré avec une interopérabilité élevée et un support de surveillance en ligne continu pour une analyse de corrélation plus approfondie. Cette thèse est une étape vers un système de collecte de données à distance, complet et continu pour les personnes âgées présentant divers types de problèmes de santé. Notre esprit de recherche est motivé par la demande immédiate d'un système de surveillance de la santé à distance sans fil sécurisé et fiable pour les personnes âgées en résidence assistée, combinant diverses sources de données. Pour créer un système aussi complet, nous le divisons en sous-systèmes, afin de le rendre réalisable et facile à mettre en œuvre, nous permettant ainsi de mettre à jour chaque sous-système individuellement dans les études futures sans affecter les autres sous-systèmes intégrés. L'accent est mis sur un système complet de surveillance à distance de l'e- santé. La liste des principales contributions contient (1) proposer une nouvelle approche pour la sécurité des appareils surveillés et proposer une solution pour prévenir les attaques MiTM et réduire la consommation d'énergie, (2) nous proposons une détection de chute fiable, (3) étudier et développer une nouvelle méthode de reconnaissance des activités quotidiennes des patients âgés surveillés, (4) proposer une approche pour améliorer la fiabilité du système et réduire les fausses alarmes et les interventions inutiles, (5) proposer et développer un algorithme de conversion de la langue des signes en texte utilisant une analyse de fusion multi-capteurs. En conséquence, nous prévoyons de fournir un système de surveillance avec une précision fiable dans la détection d'événements anormaux et de déclencher une alarme lors de la détection de tels événements pour demander de l'aide et de l'assistance Aging population is one of the key problems for the vast majority of many countries. The number of elderly people who suffer from multiple diseases and need continuous monitoring of their vital signs increases everyday, resulting in additional healthcare costs. Modern healthcare systems in geriatric medicine often require elderly presence at the hospital which conflict with their demand for independence and privacy. Recent developments on remote e-health monitoring, provides a wide range of solutions. However, most of the devices are designed for specific medical sensing and operate independently from each other. There is still a lack of integrated framework with high interoperability and continuous online monitoring support for further correlation analysis. This thesis is a step towards a remote, complete, and continuous data gathering system for elderly people with various types of health problems. Our research spirit is motivated by immediate demand in a secured and trusted remote wireless health monitoring System for assisted living Elderly people, combining various data sources. To create such a complete system we divide it into subsystems, in order to make it feasible and easy to implement, thus allowing us to update each subsystem individually in the future studies without affecting other integrated subsystems. The main focus is on a complete remote e-health monitoring system. The list of main contributions contains (1) propose a new approach for security of monitored devices and propose a solution to prevent MiTM attacks and reduce energy consumption, (2) we propose reliable fall detection,(3) investigating and developing a novel recognition method of daily activities for monitored elderly patient, (4) propose an approach to enhance the reliability of the system and to reduce false alarms and unnecessary interventions, (5) propose and develop a sign language to text converter algorithm using multi-sensor fusion analysis. As a result, we expect to provide a monitoring system with reliable accuracy in the detection of abnormal events, and raise an alarm upon detection of such events to seek help and assistance

APA, Harvard, Vancouver, ISO, and other styles

41

Zhou, Dingfu. "Vision-based moving pedestrian recognition from imprecise and uncertain data." Thesis, Compiègne, 2014. http://www.theses.fr/2014COMP2162/document.

Full text

Abstract:

La mise en oeuvre de systèmes avancés d’aide à la conduite (ADAS) basée vision, est une tâche complexe et difficile surtout d’un point de vue robustesse en conditions d’utilisation réelles. Une des fonctionnalités des ADAS vise à percevoir et à comprendre l’environnement de l’ego-véhicule et à fournir l’assistance nécessaire au conducteur pour réagir à des situations d’urgence. Dans cette thèse, nous nous concentrons sur la détection et la reconnaissance des objets mobiles car leur dynamique les rend plus imprévisibles et donc plus dangereux. La détection de ces objets, l’estimation de leurs positions et la reconnaissance de leurs catégories sont importants pour les ADAS et la navigation autonome. Par conséquent, nous proposons de construire un système complet pour la détection des objets en mouvement et la reconnaissance basées uniquement sur les capteurs de vision. L’approche proposée permet de détecter tout type d’objets en mouvement en fonction de deux méthodes complémentaires. L’idée de base est de détecter les objets mobiles par stéréovision en utilisant l’image résiduelle du mouvement apparent (RIMF). La RIMF est définie comme l’image du mouvement apparent causé par le déplacement des objets mobiles lorsque le mouvement de la caméra a été compensé. Afin de détecter tous les mouvements de manière robuste et de supprimer les faux positifs, les incertitudes liées à l’estimation de l’ego-mouvement et au calcul de la disparité doivent être considérées. Les étapes principales de l’algorithme sont les suivantes : premièrement, la pose relative de la caméra est estimée en minimisant la somme des erreurs de reprojection des points d’intérêt appariées et la matrice de covariance est alors calculée en utilisant une stratégie de propagation d’erreurs de premier ordre. Ensuite, une vraisemblance de mouvement est calculée pour chaque pixel en propageant les incertitudes sur l’ego-mouvement et la disparité par rapport à la RIMF. Enfin, la probabilité de mouvement et le gradient de profondeur sont utilisés pour minimiser une fonctionnelle d’énergie de manière à obtenir la segmentation des objets en mouvement. Dans le même temps, les boîtes englobantes des objets mobiles sont générées en utilisant la carte des U-disparités. Après avoir obtenu la boîte englobante de l’objet en mouvement, nous cherchons à reconnaître si l’objet en mouvement est un piéton ou pas. Par rapport aux algorithmes de classification supervisée (comme le boosting et les SVM) qui nécessitent un grand nombre d’exemples d’apprentissage étiquetés, notre algorithme de boosting semi-supervisé est entraîné avec seulement quelques exemples étiquetés et de nombreuses instances non étiquetées. Les exemples étiquetés sont d’abord utilisés pour estimer les probabilités d’appartenance aux classes des exemples non étiquetés, et ce à l’aide de modèles de mélange de gaussiennes après une étape de réduction de dimension réalisée par une analyse en composantes principales. Ensuite, nous appliquons une stratégie de boosting sur des arbres de décision entraînés à l’aide des instances étiquetées de manière probabiliste. Les performances de la méthode proposée sont évaluées sur plusieurs jeux de données de classification de référence, ainsi que sur la détection et la reconnaissance des piétons. Enfin, l’algorithme de détection et de reconnaissances des objets en mouvement est testé sur les images du jeu de données KITTI et les résultats expérimentaux montrent que les méthodes proposées obtiennent de bonnes performances dans différents scénarios de conduite en milieu urbain Vision-based Advanced Driver Assistance Systems (ADAS) is a complex and challenging task in real world traffic scenarios. The ADAS aims at perceiving andunderstanding the surrounding environment of the ego-vehicle and providing necessary assistance for the drivers if facing some emergencies. In this thesis, we will only focus on detecting and recognizing moving objects because they are more dangerous than static ones. Detecting these objects, estimating their positions and recognizing their categories are significantly important for ADAS and autonomous navigation. Consequently, we propose to build a complete system for moving objects detection and recognition based on vision sensors. The proposed approach can detect any kinds of moving objects based on two adjacent frames only. The core idea is to detect the moving pixels by using the Residual Image Motion Flow (RIMF). The RIMF is defined as the residual image changes caused by moving objects with compensated camera motion. In order to robustly detect all kinds of motion and remove false positive detections, uncertainties in the ego-motion estimation and disparity computation should also be considered. The main steps of our general algorithm are the following : first, the relative camera pose is estimated by minimizing the sum of the reprojection errors of matched features and its covariance matrix is also calculated by using a first-order errors propagation strategy. Next, a motion likelihood for each pixel is obtained by propagating the uncertainties of the ego-motion and disparity to the RIMF. Finally, the motion likelihood and the depth gradient are used in a graph-cut-based approach to obtain the moving objects segmentation. At the same time, the bounding boxes of moving object are generated based on the U-disparity map. After obtaining the bounding boxes of the moving object, we want to classify the moving objects as a pedestrian or not. Compared to supervised classification algorithms (such as boosting and SVM) which require a large amount of labeled training instances, our proposed semi-supervised boosting algorithm is trained with only a few labeled instances and many unlabeled instances. Firstly labeled instances are used to estimate the probabilistic class labels of the unlabeled instances using Gaussian Mixture Models after a dimension reduction step performed via Principal Component Analysis. Then, we apply a boosting strategy on decision stumps trained using the calculated soft labeled instances. The performances of the proposed method are evaluated on several state-of-the-art classification datasets, as well as on a pedestrian detection and recognition problem.Finally, both our moving objects detection and recognition algorithms are tested on the public images dataset KITTI and the experimental results show that the proposed methods can achieve good performances in different urban scenarios

APA, Harvard, Vancouver, ISO, and other styles

42

王士承. "Impatc of Detector on Egress Time During a Building Fire Accident." Thesis, 2001. http://ndltd.ncl.edu.tw/handle/13455617824119691522.

Full text

Abstract:

碩士 元智大學 機械工程研究所 89 We have been discussing fires for the few years in Taiwan, we find that many people die from inhaling too much smoke. But fire produces dense smoke and diffusion velocity is limited. It takes one or two minutes for smoke to fill the space of an average building. So the time for people to make their escape is rather short. Because devices of detector are installed, people can start escape earlier. Therefore the design and device of detector are very important inside buildings. This paper utilizes CFAST zone model and PHOENICS field model to simulate the fire accident that occurred in Sun-Trun city. Important parameters are obtained to analyze how the energy is transmitted and smoke is spread inside a building. The effects of fire detectors on the fire accident are also evaluated.

APA, Harvard, Vancouver, ISO, and other styles

43

Chen, Chih-Sheng, and 陳智聖. "The Design and Implementation of a Pocket-based Fall Accident Detector on Android Mobile Platforms." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/962435.

Full text

Abstract:

碩士 國立臺北科技大學 電腦與通訊研究所 101 We propose in this dissertation a portable pocket fall accident detection system on Android-based mobile devices. When a fall accident event is detected, the system will automatically generate a flash light as well as a sound continuously, and the emergency center will be notified so that injuries can get assistance immediately. Unlike the other existing fall accident detectors that have to be worn or fastened on the user''s body in a particular way, we just put the mobile device in the pocket, resulting in a better convenience and feasible for practical usage. With the built-in tri-axial accelerometer and electronic compass in the Android-based mobile device, the information about the user''s activity can be easily retrieved, and then analyzed by the proposed algorithm. When a fall accidents event is detected, the user’s current position acquired by the global positioning system (GPS) will be sent to the rescue center via the 3G communication system so that the user can get medical help immediately. Considering the limited computing resources in a mobile device, we propose in this dissertation an algorithm by using a finite state machine cascaded with a support vector machines for the detection of a fall accident event. Based on the concept of a finite sate machine, the features acquired in the proposed system will be examined in a sequential manner. Once the corresponding feature is verified by the current state, it can proceed to next stage; otherwise, the system will reset to the initial state and waiting for the appearance of another feature sequence. With the proposed approach, the computational burden can be alleviated significantly. Moreover, as we will see in the experiment that the interference caused by putting the device in the pocket can be successfully conquered and a distinguished fall accident detection accuracy up to 96% on the sensitivity and 99.71% on the specificity can be obtained when a set of 400 test actions in eight different kinds of activities are estimated by using the proposed approach which justifies the superiority of the proposed algorithm.

APA, Harvard, Vancouver, ISO, and other styles

44

Wu, Si-Chen, and 吳思陳. "Android-based Travel Itinerary Planning and Accident Detection System." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/77144191851873628662.

Full text

Abstract:

碩士 國立東華大學 資訊工程學系 101 The purpose of this essay is to create an integrated smart phone with multi-functions in travel planning and accident detection. A car equipped with the integrated smart phone will automatically activate its help-seeking function to send out a message for help by its accident detection function once an accident happens. The message with the indication on the location and the accident factors as well as the image recording for the entire accident will help end users to verify what caused the accident. In addition to navigation, accident detection, and automatically help-seeking features, the integrated smart phone will also be able to provide the instant travel information inquiry and trip automatically planning guide with its travel planning function. The users can receive the travel information from their phones before arriving in the destination and adjust the travel itinerary with its trip automatically planning guide function to meet their need for perfect getaways.

APA, Harvard, Vancouver, ISO, and other styles

45

Chang, Chao-Wei, and 張晁偉. "Accident Video Detection from YouTube Videos for Self-driving Cars." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/543884.

Full text

Abstract:

碩士 國立臺灣科技大學 電子工程系 107 As the perception technology of autonomous driving cars received attention in recent years, more and more researchers have investigated this issue. Specifically, to avoid car accident in road scenes is one of the critical issues for autonomous driving cars. In order to avoid car accidents, the autonomous driving cars should own the capability to detect car accidents in advance. To this end, in this thesis, we propose a novel three-stage classication architecture for dash-cam videos. First, the CNN-LSTM networks are used to detect out-of-control vehicles. Next, taking advantage of state-of-the-art object detection and object tracking schemes, we leverage the ratio of Intersection-over-union (IoU) of two bounding boxes from two vehicles to detect car accidents at the front-view images. Finally, cooperating with the inverse perspective transformation, we can confirm a car accident by using the occupancy map in the bird-view. The major contribution of this thesis is to propose a simple but eective car accident detection system based on neural networks, which can determine which frame a car accident occurs, point out which cars cause a accident and record their driving trajectories as well.

APA, Harvard, Vancouver, ISO, and other styles

46

Huang, Chih-Chi, and 黃至麒. "A Smart Phone-based Pocket Fall Accident Prediction and Detection System." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/9nt77z.

Full text

Abstract:

碩士 國立臺北科技大學 電子工程系研究所 104 We propose in this thesis an algorithm about fall prediction and detection system which can be operated on the smart phone. Because of the advanced technology in MEMS, we can get gravity and angular velocity signal by accelerometer and gyroscope inside the smartphone and analyze the signal. We construct an portable pocket-based fall detection system. In the past, most of the research about fall detection system are focused on after occurrence processing. In other words, these systems just start to process the signal and finish after human has been impact. However, we propose an algorithm can process the signal in real time. When fall event occur, system not only start to compute the input signal but finish detection before body hit the ground. This system is a fall prediction system. We hope we can predict the fall event and prevent the injury. It must be a significant develop in home cares. If we want to achieve fall prediction system on the mobile phone, time consumption and operation count must be reduced. For this reason, support vector machine (SVM) is a good choice. Furthermore, we can use signal processing skill including filtering, quantization, down sample and binarization to support the main system. In our fall prediction system, we have 91.6% in sensitivity and 96.6% in specificity. On the other hand, we also have a fast and good accuracy algorithm for fall detection after impact to repair the low performance in fall prediction. Afterward fall detection is a hierarchy structure to determine the feature by signal process, Schmitt trigger and standard signal comparing. Without frequency transform, just low operation but get 95% in sensitivity and 92.8% in specificity.

APA, Harvard, Vancouver, ISO, and other styles

47

Hsieh, Yun-Chi, and 謝云麒. "Traffic Accident Detection System Design Based on Vehicle Edge Motion Analysis." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/a6agz5.

Full text

Abstract:

碩士 國立雲林科技大學 電子工程系 103 With rapid development of the city and popularity of the vehicle, there are many vehicles on the road. Unfortunately, complex traffic of bring more and more traffic accident. If there is no witness there, it will not be dealt with in time that will cause injury. Therefore, by using traffic monitoring system, it can detect the accident immediately, informing related personnel to solve the problem and rescue people. It is very difficult to detect vehicle accidents, especially in various road conditions. We must keep higher correct detection rate and lower false alarm rate. Common method of accident detection is by tracking vehicle to determining the changes of vehicle's position, direction, area, and acceleration of vehicles in order to detect accident. Tracking result has great influence on accident detection. It is a great challenge to detect accident when vehicles occluded. Turning vehicle may occuring false alarm. This paper proposes a traffic accident detection system design based on vehicle edge motion analysis. It is effective solution for failure detecion caused by the turning vehicles. We selecting the direction of moving objects based on edges as the feature, and detecting accident by analyzing the varied direction in the blocks. Experimental results show that our proposed algorithm in various road scenarios can effectively detect vehicle accident, the detection rate achieves 85%, and the correct detection rate achieves 77%, the error rate decreased only 0.0008%. Our method is better than the other methods.

APA, Harvard, Vancouver, ISO, and other styles

48

Teizer, Jochen. "Real-time spatial modeling to detect and track resources on construction sites." Thesis, 2006. http://hdl.handle.net/2152/2939.

Full text

APA, Harvard, Vancouver, ISO, and other styles

49

Yen-LunHsu and 許晏綸. "Implement Car Accident Detection System on Dashboard Camera by Faster R-CNN." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/86t64c.

Full text

Abstract:

碩士 國立成功大學 資訊管理研究所 107 The invention of the vehicle shortens the distance between people, but it brings considerable danger. Car accidents will cause huge social costs and harm people's lives and property. Therefore, in recent years, research on car accidents has gradually increased. Taiwan is a densely populated country, and most people use motorcycles as their means of transportation. Thus, Taiwan's car accident casualty rate is one of the highest in the world. It is important to deal with problems related to car accidents in Taiwan. Most of the relevant literature, which analyzes the sensor information on the vehicle to determine whether there is a car accident, cannot preserve the circumstances, so it is impossible to judge the traffic accident responsibility. We use the deep learning object detection method, Faster Region-based Convolutional Neural Network(Faster R-CNN), to instantly identify car accidents. And we implement it on the Android system. We write apps through Android Studio, which can judge car accidents while photographing.

APA, Harvard, Vancouver, ISO, and other styles

50

Hamdane, Hédi. "Improvement of pedestrian safety: response of detection systems to real accident scenarios." Thesis, 2016. http://hdl.handle.net/2440/119694.

Full text

Abstract:

The scope of this research concerns pedestrian active safety. Several primary safety systems have been developed for vehicles in order to detect a pedestrian and to avoid an impact. These systems analyse the forward path of the vehicle through the processing of images from sensors. If a pedestrian is identified on the vehicle trajectory, these systems employ emergency braking and some systems may potentially employ emergency steering. Methods for assessing the effectiveness of these systems have been developed. But, it appears difficult to determine the relevance of these systems in terms of pedestrian protection. The general objective of this research was to test the response of these systems in many accident configurations. The first step consisted of gathering a sample of a hundred of accidents involving vehicles with pedestrians. These accidents were provided from accident databases of two laboratories LMA and CASR. Data of these accidents were recorded in sufficient detail from in-depth investigation which enables reconstructing the trajectory of the vehicle and pedestrian prior to the collision. The second step was to analyse qualitatively and quantitatively the data of the selected accidents. These accidents were reconstructed to simulate the pre-crash conditions. From this accident reconstruction, factors relevant to the primary safety of pedestrians were deduced. The next step consisted of coupling the vehicle dynamic behaviour with a primary safety system in order to confront these systems to real accident configurations. The potential of these systems is studied by verifying the feasibility of deploying an autonomous emergency manoeuvre during the timeline of the accident and according to the vehicle dynamic capabilities: i.e. verifying the possibilities in terms of crash avoidance. Based on this procedure, three modelling methods were developed: a first method testing a system to each accident configuration and two others using graphs of evaluation from a parametric study realised on a generic system. The results of the three methods were then discussed. Finally, as a perspective, the last study will approach crash mitigation. As a consequence of an active safety system response, the vehicle impact speed is reduced. The effect of speed reduction on variations in impact conditions will be then addressed to measure the potential safety impact of these systems on pedestrian protection. Thesis (Ph.D.) -- University of Adelaide, School of Mechanical Engineering, 2016.

APA, Harvard, Vancouver, ISO, and other styles

Dissertations / Theses on the topic 'Accident detector'

Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles