Dissertations / Theses on the topic 'Road cameras'

Road traffic accidents continue to be a public health problem and are a global issue due to the huge financial burden they place on families and society as a whole. Speed has been identified as a major contributor to the severity of traffic accidents and there is the need for better speed management if road traffic accidents are to be reduced. Over the years various measures have been implemented to manage vehicle speeds. The use of speed cameras and vehicle activated signs in recent times has contributed to the reduction of vehicle speeds to various extents. Speed cameras use punitive measures whereas vehicle activated signs do not so their use depends on various factors. Engineers, planners and decision makers responsible for determining the best place to mount a speed camera or vehicle activated sign along a road have based their decision on experience, site characteristics and available guidelines (Department for Transport, 2007; Department for Transport, 2006; Department for Transport, 2003). These decisions can be subjective and indications are that a more formal and directed approach aimed at bringing these available guidelines together in a model will be beneficial in making the right decision as to where to place a speed camera or vehicle activated sign is to be made. The use of optimisation techniques have been applied in other areas of research but this has been clearly absent in the Transport Safety sector. This research aims to contribute to speed reduction by developing a model to help decision makers determine the optimum location for a speed control device. In order to achieve this, the first study involved the development of an Empirical Bayes Negative Binomial regression accident prediction model to predict the number of fatal and serious accidents combined and the number of slight accidents. The accident prediction model that was used explored the effect of certain geometric and traffic characteristics on the effect of the severity of road traffic accident numbers on selected A-roads within the Nottinghamshire and Leicestershire regions of United Kingdom. On A-roads some model variables (n=10) were found to be statistically significant for slight accidents and (n=6) for fatal and serious accidents. The next study used the accident prediction model developed in two optimisation techniques to help predict the optimal location for speed cameras or vehicle activated signs. Pattern Search and Genetic Algorithms were the two main types of optimisation techniques utilised in this thesis. The results show that the two methods did produce similar results in some instances but different in others. Optimised results were compared to some existing sites with speed cameras some of the results obtained from the optimisation techniques used were within proximity of about 160m. A validation method was applied to the genetic algorithm and pattern search optimisation methods. The pattern search method was found to be more consistent than the genetic algorithm method. Genetic algorithm results produced slightly different results at validation in comparison with the initial results. T-test results show a significant difference in the function values for the validated genetic algorithm (M= 607649.34, SD= 1055520.75) and the validated pattern search function values (M= 2.06, SD= 1.17) under the condition t (79) = 5.15, p=0.000. There is a role that optimisation techniques can play in helping to determine the optimum location for a speed camera or vehicle activated sign based on a set of objectives and specified constraints. The research findings as a whole show that speed cameras and vehicle activated signs are an effective speed management tool. Their deployment however needs to be carefully considered by engineers, planners and decision makers so as to achieve the required level of effectiveness. The use of optimisation techniques which has been generally absent in the Transport Safety sector has been shown in this thesis to have the potential to contribute to improve speed management. There is however no doubt that this research will stimulate interest in this rather new but high potential area of Transport Safety.

Roads can negatively impact the survival of wildlife populations through additional mortality from road kill and population fragmentation caused by road avoidance behaviors. The 11.9 mile section of State Route 101 between the towns of San Luis Obispo and Atascadero, CA, USA, cross a mountain lion movement corridor and an area important to maintaining ecological connectivity between protected lands in the Los Padres National Forest to the north and south. I examined the spatial patterns and landscape and roadway factors associated with road kill occurrence for six taxa; large mammals, mesocarnivores, squirrels, rabbits, birds and raptors. Between 1 May 2009 and 30 June 2010 road kills were documented using vehicle-based surveys. Small mammals were the most common road kill (58.3%), followed by mesocarnivores (10.9%), birds (10.6%), rabbits (5.1%), large mammals (3.3%) and raptors (3.2%). Twenty-nine large mammal road kills were observed during the survey period; eighteen mule deer, six black bears and five feral pigs. Road kill was highest in the middle of the survey area between the top of Cuesta Grade and the southern edge of Atascadero and lowest along the Cuesta Grade. I modeled road kill occurrence using logistic regression to determine which landscape and roadway characteristics were associated with road kill locations. Large mammal and mesocarnivore road kills were more likely to occur near riparian corridors. Mesocarnivore and squirrel road kills were associated with locations with greater roadside tree cover. Squirrel and rabbit road kills were more likely to occur along sections of the road with large grassy center medians. I documented animal activity patterns around the roadway during three survey periods (summer 2009, fall 2009 and spring 2010) using remote cameras placed on game trails and underpasses along the roadway. Mule deer displayed crepuscular activity patterns with peaks in activity in the morning between 05:00h and 07:00h and in the evening between 16:00h and 18:00h. Mesocarnivores generally displayed a nocturnal activity patterns with the majority of activity occurring between 18:00h and 06:00h. I used logistic regression to determine if there was a relationship between animal activity patterns and traffic patterns while controlling for time of day, day of the week, and season. Mule deer and mesocarnivore activity patterns varied significantly by time of day and mule deer activity also varied significantly by season; however only mesocarnivore activity varied significantly in relation to traffic volume suggesting that mesocarnivores are less activity when traffic volume is high. Using traffic volume and animal activity patterns I calculated a collision potential value for both mule deer and mesocarnivores. Collision potential for mule deer was high in the morning, between 06:00h and 08:00h, and in the evening, between 16:00h and 18:00h in all three seasons. Collision potential for mesocarnivores was high in the evening in fall 2009 (18:00h and 21:00) and spring 2010 (17:00h), and high in the morning in summer 2009 (09:00h).

Las lesiones por tráfico son un importante problema de salud pública en España. En 2004, el gobierno español estableció la seguridad vial como una prioridad política, y creó el Plan Estratégico de Seguridad Vial 2005-2008, que propone una serie de medidas dirigidas a disminuir el impacto de las lesiones por tráfico en España. Los objetivos de la tesis son revisar las intervenciones de seguridad vial que se han demostrado efectivas en reducir las lesiones y mortalidad por tráfico y evaluar el impacto en morbilidad y mortalidad por tráfico del conjunto de medidas implementadas en España a partir del año 2004 y de algunas de las medidas implementadas, concretamente el permiso por puntos, la reforma del Código Penal y los radares. El diseño de todos los estudios de evaluación consistió en estudios de series temporales interrumpidas. Las poblaciones de estudio fueron el número de colisiones, conductores involucrados en colisiones con lesionados y personas lesionadas por tráfico en España entre los años 2000 y 2008. Las fuentes de información fueron bases de datos de policía y hospitalarias. Se ajustaron modelos de regresión Quasi-Poisson, controlando la tendencia temporal y la estacionalidad. Los estudios incluidos en la tesis sugieren que la priorización de la seguridad vial en el año 2004 supuso un cambio en la tendencia de las lesiones por tráfico en España, y fue especialmente efectiva en reducir el número de lesionados graves. Entre las intervenciones incluidas en el Plan Estratégico de Seguridad Vial 2005-2008, se evaluó la efectividad de los radares, el permiso por puntos y la criminalización de una serie de comportamientos de tráfico – mediante la reforma del Código Penal –, medidas que redujeron el impacto de las lesiones por tráfico en España. Sin embargo, la revisión de la literatura incluida en la tesis identificó diversas medidas efectivas de seguridad vial, como el permiso de conducir gradual, que todavía no han sido implementadas y que podrían reducir todavía más el número de personas lesionadas en las carreteras españolas. Será necesario realizar esfuerzos importantes y adjudicar suficientes recursos para mantener el nivel de seguridad vial alcanzado. Además, se deberán implementar más medidas efectivas de seguridad vial para reducir el todavía inaceptablemente elevado número de personas lesionadas o muertas en las carreteras españolas cada día.
Road traffic injuries are an important public health problem in Spain. In 2004, the Spanish government established road safety as a political priority, and created the Road Safety Strategic Programme 2005-2008, which proposes a series of actions aimed to reduce the burden of traffic injuries in Spain. The objectives of the present thesis are to review the road safety interventions which have proven to be effective in reducing road traffic deaths and injuries, and to assess the impact on traffic morbidity and mortality of overall road safety interventions implemented in Spain from the year 2004 on and of specific road safety interventions implemented, specifically the penalty points system, the reform of the Penal Code and speed cameras. The design of all the intervention evaluation studies consisted in interrupted time-series studies. The number of injury crashes, drivers involved in injury collisions, and people injured in traffic collisions in Spain between the years 2000-2008 were the study populations. Police and hospital registries were used as sources of information. Quasi-Poisson regression models were adjusted, controlling for time trend and seasonal patterns. The studies included in the present thesis suggest that the prioritisation of road safety in the year 2004 changed the trend of road traffic injuries in Spain, being especially effective in reducing the number of seriously injured people. Among the interventions included in the Road Safety Strategic Programme 2005-2008, speed cameras, the penalty points system, and the criminalisation of a set of road behaviours - by means of reforming the Penal Code – were assessed for effectiveness and were observed to reduce the burden of traffic injuries in Spain. Nevertheless, the literature review included in the thesis identified several effective road safety interventions, such as the graduated licensing system, that have not been implemented as yet, and which could further reduce the number of people injured on the Spanish roads. Important efforts and enough resources will be needed to maintain the level of road safety achieved. Furthermore, additional effective road safety measures should be implemented to reduce the still unacceptably high number of people injured or killed on the Spanish roads every day.

Roads pose two central problems for wildlife: wildlife-vehicle collisions (WVCs) and habitat fragmentation. Wildlife exclusion fencing can reduce WVCs but can exacerbate fragmentation. In Chapter 1, I summarize the relevant studies addressing these two problems, with a focus on large mammals in North America. Chapters 2 and 3 summarize field assessments of technologies to reduce WVCs and maintain connectivity, specifically jumpout ramps and underpasses, along Highway 101 near San Luis Obispo, CA. In a fenced highway, some animals inevitably breach the fence and become trapped, which increases the risk of a wildlife-vehicle collision. Earthen escape ramps, or “jumpouts”, can allow the trapped animal to escape the highway corridor. Few studies have quantified wildlife use of jumpouts, and none for >2 years. We used wildlife cameras to quantify wildlife use of 4 jumpouts from 2012-2017. Mule deer were 88% percent of our detections and jumped out 20% of the time. After accounting for pseudoreplication, 33% of the events were independent events, and 2 groups of deer accounted for 41% of all detections at the top of the jumpout. Female deer were 86% of the detections and were much more likely than males to return to the jumpout multiple times. This is the first study to document use of jumpouts for more than 3 years, the first to account for pseudoreplication, and the first to quantify differences in jumpout use between male and female mule deer. We recommend a jumpout height between 1.75m-2m for mule deer to increase the jumpout success rate. Chapter 3 addresses factors that may affect the use of undercrossings by mule deer and other wildlife. Wildlife crossings combined with wildlife exclusion fencing have been shown to be the most effective method to reduce wildlife-vehicle collisions while maintaining ecological connectivity. Although several studies have quantified wildlife use of undercrossings, very few have exceeded 24 months, and the factors affecting carnivores use of the undercrossings remain unclear. We quantified mule deer, black bear, mountain lion, and bobcat use of 11 undercrossings along Highway 101 near San Luis Obispo, California from 2012-2017. We constructed zero-inflated Poisson general linear models on the monthly activity of our focal species using underpass dimensionality, distance to cover, substrate, human activity, and location relative to the wildlife exclusion fence as predictor variables. We accounted for temporal variation, as well as spatial variation by quantifying the landscape resistance near each undercrossing. We found that deer almost exclusively used the larger underpasses whereas the carnivores were considerably less selective. Bears used undercrossings more that were within the wildlife exclusion fence, whereas mountain lion activity was higher outside the wildlife exclusion fence. Bobcat activity was highest and most widespread, and was negatively associated with distance to cover. Regional connectivity is most important for bear and mountain lion, and the surrounding habitat may be the most important predictor for their use of undercrossings. We recommend placing GPS collars on our focal species to more clearly document fine-scale habitat selection near the highway.

O uso de sistemas de assistência ao motorista (DAS) baseados em visão tem contribuído consideravelmente na redução de acidentes e consequentemente no auxílio de uma melhor condução. Estes sistemas utilizam basicamente uma câmera de vídeo embarcada (normalmente fixada no para-brisa) com o propósito de extrair informações acerca da rodovia e ajudar o condutor num melhor processo de dirigibilidade. Pequenas distrações ou a perda de concentração podem ser suficientes para que um acidente ocorra. Este trabalho apresenta uma proposta para o desenvolvimento de algoritmos para extrair informações sobre a sinalização em rodovias. Mais precisamente, serão abordados algoritmos de calibração de câmera explorando a geometria da pista, de extração da marcação de pintura (sinalização horizontal) e detecção e identificação de placas de trânsito (sinalização vertical). Os resultados experimentais indicam que o método de calibração de câmera alcançou bons resultados na obtenção dos parâmetros extrínsecos com erros inferiores a 0:5 . O erro médio encontrado nos experimentos com relação a estimativa da altura da câmera foi em torno de 12 cm (erro relativo aproximado de 10%), permitindo explorar o uso da realidade aumentada como uma possível aplicação. A acurácia global para a detecção e reconhecimento da sinalização horizontal (marcas seccionadas, contínuas e mistas) foi acima de 96% perante uma diversidade de situações apresentadas, tais como: sombras, variação de iluminação, degradação do asfalto e pintura. O uso da câmera calibrada para a detecção da sinalização vertical contribui para delimitar o espaço de varredura da janela deslizante do detector, bem como realizar a procura por placas em uma única escala para cada região de busca, caracterizada pela distância ao veículo. Os resultados apresentados reportam uma taxa global de classificação de aproximadamente 99% para o sinal de proibido ultrapassar, considerando-se uma base de dados limitada a 962 amostras.
The use of driver assistance systems (DAS) based on computer vision has helped considerably in reducing accidents and consequently aid in better driving. These systems primarily use an embedded video camera (usually fixed on the windshield) for the purpose of extracting information about the highway and assisting the driver in a better handling process. Small distractions or loss of concentration may be sufficient for an accident to occur. This work presents the development of algorithms to extract information about traffic signs on highways. More specifically, this work will tackle a camera calibration algorithm that exploits the geometry of the road track, algorithms for the extraction of road marking paint (lane markings) and detection and identification of vertical traffic signs. Experimental results indicate that the proposed method for obtaining the extrinsic parameters achieve good results with errors of less than 0:5 . The average error in our experiments, related to the camera height, were around 12 cm (relative error around 10%). Global accuracy for the detection and classification of road lane markings (dashed, solid, dashed-solid, solid-dashed or double solid) were over 96%. Finally, our camera calibration algorithm was used to reduce the search region and to define the scale of a slidingwindow detector for vertical traffic signs. The use of the calibrated camera for the detection of traffic signs contributes to define the scanning area of the sliding window and perform a search for signs on a unique scale for each region of interest, determined by the distance to the vehicle. The results reported a global classification rate of approximately 99% for the no overtaking sign, considering a limited of 962 samples.

Recently, sensors such as radars and cameras have been widely used in automotives, especially in Advanced Driver-Assistance Systems (ADAS), to collect information about the vehicle's surroundings. Stereo cameras are very popular as they could be used passively to construct a 3D representation of the scene in front of the car. This allowed the development of several ADAS algorithms that need 3D information to perform their tasks. One interesting application is Road Surface Preview (RSP) where the task is to estimate the road height along the future path of the vehicle. An active suspension control unit can then use this information to regulate the suspension, improving driving comfort, extending the durabilitiy of the vehicle and warning the driver about potential risks on the road surface. Stereo cameras have been successfully used in RSP and have demonstrated very good performance. However, the main disadvantages of stereo cameras are their high production cost and high power consumption. This limits installing several ADAS features in economy-class vehicles. A less expensive alternative are monocular cameras which have a significantly lower cost and power consumption. Therefore, this thesis investigates the possibility of solving the Road Surface Preview task using a monocular camera. We try two different approaches: structure-from-motion and Convolutional Neural Networks.The proposed methods are evaluated against the stereo-based system. Experiments show that both structure-from-motion and CNNs have a good potential for solving the problem, but they are not yet reliable enough to be a complete solution to the RSP task and be used in an active suspension control unit.

Road conditions are a critical issue for road users as, if not given sufficient attention, they may threaten users’ lives. The environmental parameters, such as snowy, icy, dry and wet, are important in relation to the condition of roads. This is particularly true in relation to the northern regions and greatest concern should be in relation to snowy and icy situations. In this thesis, a system based on an InGaAs area scan sensor utilizes NIR technology to detect water or ice on the road so as to enable drivers to avoid slippery road conditions. The conditions caused by freezing water on road surface are particularly dangerous and are not easy to observe and it is hope that this project will boost traffic safety. The system is able to assist road maintenance personnel in forecasting and detecting slippery road conditions during winter road maintenance (WRM). The system, which is based on FPGA, has functionalities that display the captured images on an HDMI monitor and send the images to the software on a host PC via the UART protocol. An interface board, which carries the sensor and which connects to the FPGA board, is developed for NIR sensor. VHDL implementation and PC software design are the works included in the project. Besides, this device is exploited utilizing InGaAs image sensor. According to its features, it can be applied in other applications which will also be discussed. Finally, experiments are conducted in order to investigate the system’s operation with the variation of temperature.

Over the past several decades, research efforts in the development of self-driving vehicles have drastically improved accompanying technologies. Since the challenges held by Defense Advanced Research Projects Agency, the autonomous driving industry has increased significantly, and almost all the automotive companies have started to develop the technologies to deploy autonomous driving vehicles in the real world. Even though a lot of companies have been making efforts to achieve fully automated vehicles, the current technologies are not mature enough to be deployed in the real world yet, because self-driving vehicles need to respond to uncontrolled environments, such as moving objects, pedestrians, traffic lights, and unexpected work-zones. Among these uncontrolled environments, this thesis focuses on understanding road information and estimating states of traffic lights. Given that all of the traffic control devices are regularized in colors, color is one of the most significant features to be recognized. In order to accomplish such necessary a vision task, self-driving vehicles must incorporate cameras. Despite the fact that traffic control devices have their own regularized color and cameras can see those devices, they are still difficult to detect and recognize by autonomous vehicles. One of the biggest problems is that the color of those devices can be captured differently based on illumination. In this thesis, we investigate the problem of recognizing static objects using a monocular camera to assist self-driving vehicles in perceiving traffic control devices. The perception system, specifically a camera, should recognize the objects robustly regardless of the environment. Throughout this thesis, we exploit different color spaces and apply machine learning to reduce color variance. Also, we develop algorithms which compensate for illumination changes by considering the Sun position, to further improve the road sign recognition. Furthermore, we improve a traffic light state estimation which performs robustly under various illumination conditions. We deploy and demonstrate all of the algorithms in an autonomous vehicle.

A complete prototype system for measuring vehicle lateral position has been set up during the course of this master’s thesis project. In the development of the software, images acquired from a back-ward looking video camera mounted on the roof of the vehicle were used.

The problem of using computer vision to measure lateral position can be divided into road marking detection and lateral position extraction. Since the strongest characteristic of a road marking image are the edges of the road markings, the road marking detection step is based on edge detection. For the detection of the straight edge lines a Hough based method was chosen. Due to peak spreading in Hough space, the difficulty of detecting the correct peak in Hough space was encountered. A flexible Hough peak detection algorithm was developed based on an adaptive window that takes peak spreading into account. The road marking candidate found by the system is verified before the lateral position data is generated. A good performance of the road marking tracking algorithm was obtained by exploiting temporal correlation to update a search region within the image. A camera calibration made the extraction of real-world lateral position information and yaw angle data possible.

This vision-based method proved to be very accurate. The standard deviation of the error in the position detection is 0.012 m within an operating range of ±2 m from the image centre. During continuous road markings the rate of valid data is on average 96 %, whereas it drops to around 56 % for sections with intermittent road markings. The system performs well during lane change manoeuvres, which is an indication that the system tracks the correct road marking. This prototype system is a robust and automatic measurement system, which will benefit VTI in its many driving behaviour research programs.

La perception de scènes routières est un domaine de recherche très actif. Cette thèse se focalise sur la détection et le suivi d’objets par fusion de données d’un système multi-capteurs composé d’un télémètre laser, une caméra fisheye et un système de positionnement global (GPS). Plusieurs étapes de la chaîne de perception sont ´ étudiées : le calibrage extrinsèque du couple caméra fisheye / télémètre laser, la détection de la route et enfin la détection et le suivi d’obstacles sur la route.Afin de traiter les informations géométriques du télémètre laser et de la caméra fisheye dans un repère commun, une nouvelle approche de calibrage extrinsèque entre les deux capteurs est proposée. La caméra fisheye est d’abord calibrée intrinsèquement. Pour cela, trois modèles de la littérature sont étudiés et comparés. Ensuite, pour le calibrage extrinsèque entre les capteurs,la normale au plan du télémètre laser est estimée par une approche de RANSAC couplée `a une régression linéaire `a partir de points connus dans le repère des deux capteurs. Enfin une méthode des moindres carres basée sur des contraintes géométriques entre les points connus, la normale au plan et les données du télémètre laser permet de calculer les paramètres extrinsèques. La méthode proposée est testée et évaluée en simulation et sur des données réelles.On s’intéresse ensuite `a la détection de la route à partir des données issues de la caméra fisheye et du télémètre laser. La détection de la route est initialisée `a partir du calcul de l’image invariante aux conditions d’illumination basée sur l’espace log-chromatique. Un seuillage sur l’histogramme normalisé est appliqué pour classifier les pixels de la route. Ensuite, la cohérence de la détection de la route est vérifiée en utilisant les mesures du télémètre laser. La segmentation de la route est enfin affinée en exploitant deux détections de la route successives. Pour cela, une carte de distance est calculée dans l’espace couleur HSI (Hue,Saturation, Intensity). La méthode est expérimentée sur des données réelles. Une méthode de détection d’obstacles basée sur les données de la caméra fisheye, du télémètre laser, d’un GPS et d’une cartographie routière est ensuite proposée. On s’intéresse notamment aux objets mobiles apparaissant flous dans l’image fisheye. Les régions d’intérêts de l’image sont extraites `a partir de la méthode de détection de la route proposée précédemment. Puis, la détection dans l’image du marquage de la ligne centrale de la route est mise en correspondance avec un modelé de route reconstruit `a partir des données GPS et cartographiques. Pour cela, la transformation IPM (Inverse Perspective Mapping) est appliquée à l’image. Les régions contenant potentiellement des obstacles sont alors extraites puis confirmées à l’aide du télémètre laser.L’approche est testée sur des données réelles et comparée `a deux méthodes de la littérature. Enfin, la dernière problématique étudiée est le suivi temporel des obstacles détectés `a l’aide de l’utilisation conjointe des données de la caméra fisheye et du télémètre laser. Pour cela, les resultats de détection d’obstacles précédemment obtenus sont exploit ´es ainsi qu’une approche de croissance de région. La méthode proposée est également testée sur des données réelles
Road scene understanding is one of key research topics of intelligent vehicles. This thesis focuses on detection and tracking of obstacles by multisensors data fusion and analysis. The considered system is composed of a lidar, a fisheye camera and aglobal positioning system (GPS). Several steps of the perception scheme are studied: extrinsic calibration between fisheye camera and lidar, road detection and obstacles detection and tracking. Firstly, a new method for extinsic calibration between fisheye camera and lidar is proposed. For intrinsic modeling of the fisheye camera, three models of the literatureare studied and compared. For extrinsic calibration between the two sensors, the normal to the lidar plane is firstly estimated based on the determination of ń known ż points. The extrinsic parameters are then computed using a least square approachbased on geometrical constraints, the lidar plane normal and the lidar measurements. The second part of this thesis is dedicated to road detection exploiting both fisheye camera and lidar data. The road is firstly coarse detected considering the illumination invariant image. Then the normalised histogram based classification is validated using the lidar data. The road segmentation is finally refined exploiting two successive roaddetection results and distance map computed in HSI color space. The third step focuses on obstacles detection, especially in case of motion blur. The proposed method combines previously detected road, map, GPS and lidar information.Regions of interest are extracted from previously road detection. Then road central lines are extracted from the image and matched with road shape model extracted from 2DŋSIG map. Lidar measurements are used to validated the results.The final step is object tracking still using fisheye camera and lidar. The proposed method is based on previously detected obstacles and a region growth approach. All the methods proposed in this thesis are tested, evaluated and compared to stateŋofŋtheŋart approaches using real data acquired with the IRTESŋSET laboratory experimental platform

Modern cars are often equipped with sensors like radar, infrared cameras and stereo cameras that collect information about its surroundings. By using a stereo camera, it is possible to receive information about the distance to points in front of the car. This information can be used to estimate the height of the predicted path of the car. An application which does this is the stereo based Road surface preview (RSP) algorithm. By using the output from the RSP algorithm it is possible to use active suspension control, which controls the vertical movement of the wheels relative to the chassis. This application primarily makes the driving experience more comfortable, but also extends the durability of the vehicle. The idea behind this Master’s thesis is to create an evaluation tool for the RSP algorithm, which can be used at arbitrary roads.  The thesis describes the proposed evaluation tool, where focus has been to make an accurate comparison of camera data received from the RSP algorithm and laser data used as ground truth in this thesis. Since the tool shall be used at the company proposing this thesis, focus has also been on making the tool user friendly. The report discusses the proposed methods, possible sources to errors and improvements. The evaluation tool considered in this thesis shows good results for the available test data, which made it possible to include an investigation of a possible improvement of the RSP algorithm.

Modern day cars are often equipped with a variety of sensors that collect information about the car and its surroundings. The stereo camera is an example of a sensor that in addition to regular images also provides distances to points in its environment. This information can, for example, be used for detecting approaching obstacles and warn the driver if a collision is imminent or even automatically brake the vehicle. Objects that constitute a potential danger are usually located on the road in front of the vehicle which makes the road surface a suitable reference level from which to measure the object's heights. This Master's thesis describes how an estimate of the road surface can be found to in order to make these height measurements. The thesis describes how the large amount of data generated by the stereo camera can be scaled down to a more effective representation in the form of an elevation map. The report discusses a method for relating data from different instances in time using information from the vehicle's motion sensors and shows how this method can be used for temporal filtering of the elevation map. For estimating the road surface two different methods are compared, one that uses a RANSAC-approach to iterate for a good surface model fit and one that uses conditional random fields for modeling the probability of different parts of the elevation map to be part of the road. A way to detect curb lines and how to use them to improve the road surface estimate is shown. Both methods for road classification show good results with a few differences that are discussed towards the end of the report. An example of how the road surface estimate can be used to detect obstacles is also included.

Deep learning for safe autonomous transport is rapidly emerging. Fast and robust perception for autonomous vehicles will be crucial for future navigation in urban areas with high traffic and human interplay. Previous work focuses on extracting full image depth maps, or finding specific road features such as lanes. However, in urban environments lanes are not always present, and sensors such as LiDAR with 3D point clouds provide a quite sparse depth perception of road with demanding algorithmic approaches. In this thesis we derive a novel convolutional neural network that we call AutoNet. It is designed as an encoder-decoder network for pixel-wise depth estimation of an urban drivable free-space road, using only a monocular camera, and handled as a supervised regression problem. AutoNet is also constructed as a classification network to solely classify and segment the drivable free-space in real- time with monocular vision, handled as a supervised classification problem, which shows to be a simpler and more robust solution than the regression approach. We also implement the state of the art neural network ENet for comparison, which is designed for fast real-time semantic segmentation and fast inference speed. The evaluation shows that AutoNet outperforms ENet for every performance metrics, but shows to be slower in terms of frame rate. However, optimization techniques are proposed for future work, on how to advance the frame rate of the network while still maintaining the robustness and performance. All the training and evaluation is done on the Cityscapes dataset. New ground truth labels for road depth perception are created for training with a novel approach of fusing pre-computed depth maps with semantic labels. Data collection with a Scania vehicle is conducted, mounted with a monocular camera to test the final derived models. The proposed AutoNet shows promising state of the art performance in regards to road depth estimation as well as road classification.
Deep learning för säkra autonoma transportsystem framträder mer och mer inom forskning och utveckling. Snabb och robust uppfattning om miljön för autonoma fordon kommer att vara avgörande för framtida navigering inom stadsområden med stor trafiksampel. I denna avhandling härleder vi en ny form av ett neuralt nätverk som vi kallar AutoNet. Där nätverket är designat som en autoencoder för pixelvis djupskattning av den fria körbara vägytan för stadsområden, där nätverket endast använder sig av en monokulär kamera och dess bilder. Det föreslagna nätverket för djupskattning hanteras som ett regressions problem. AutoNet är även konstruerad som ett klassificeringsnätverk som endast ska klassificera och segmentera den körbara vägytan i realtid med monokulärt seende. Där detta är hanterat som ett övervakande klassificerings problem, som även visar sig vara en mer simpel och mer robust lösning för att hitta vägyta i stadsområden. Vi implementerar även ett av de främsta neurala nätverken ENet för jämförelse. ENet är utformat för snabb semantisk segmentering i realtid, med hög prediktions- hastighet. Evalueringen av nätverken visar att AutoNet utklassar ENet i varje prestandamätning för noggrannhet, men visar sig vara långsammare med avseende på antal bilder per sekund. Olika optimeringslösningar föreslås för framtida arbete, för hur man ökar nätverk-modelens bildhastighet samtidigt som man behåller robustheten.All träning och utvärdering görs på Cityscapes dataset. Ny data för träning samt evaluering för djupskattningen för väg skapas med ett nytt tillvägagångssätt, genom att kombinera förberäknade djupkartor med semantiska etiketter för väg. Datainsamling med ett Scania-fordon utförs även, monterad med en monoculär kamera för att testa den slutgiltiga härleda modellen. Det föreslagna nätverket AutoNet visar sig vara en lovande topp-presterande modell i fråga om djupuppskattning för väg samt vägklassificering för stadsområden.

Advanced Driver Assistance System (ADAS) was widely learned nowadays. As crucial parts of ADAS, lane markings detection, as well as other objects detection, have become more popular than before. However, most methods implemented in such areas cannot perfectly balance the performance of accuracy versus efficiency, and the mainstream methods (e.g. Machine Learning) suffer from several limitations which can hardly break the wall between partial autonomous and fully autonomous driving. This thesis proposed a real-time lane marking detection framework for ADAS, which included 4-extreme points set descriptor and a rule-based cascade classifier. By analyzing the behavior of lane markings on the road surface, a characteristic of markings was discovered, i.e., standard markings can sustain their shape in the perpendicular plane of the driving direction. By employing this feature, a 4-extreme points set descriptor was applied to describe the shape of each marking first. Specifically, after processing Maximally Stable Extremal Region (MSER) and Hough transforms on a 2-D image, several contours of interest are obtained. A bounding box, with borders parallel to the image coordinate, intersected with each contour at 4 points in the edge, which was named 4-extreme points set. Afterward, to verify consistency of each contour and standard marking, some rules abstracted from construction manual are employed such as Area Filter, Colour Filter, Relative Location Filter, Convex Filter, etc. To reduce the errors caused by changes in driving direction, an enhanced module was then introduced. By tracking the vanishing point as well as other key points of the road net, a method for 3-D reconstruction, with respect to the optical axis between vanishing point and camera center, is possible. The principle of such algorithm was exhibited, and a description about how to obtain the depth information from this model was also provided. Among all of these processes, a key-point based classification method is the main contribution of this paper because of its function in eliminating the deformation of the object caused by inverse perspective mapping. Several experiments were conducted in highway and urban roads in Ottawa. The detection rate of the markings by the proposed algorithm reached an average accuracy rate of 96.77% while F1 Score (harmonic mean of precision and recall) also attained a rate of 90.57%. In summary, the proposed method exhibited a state-of-the-art performance and represents a significant advancement of understanding.

This work is an interdisciplinary exploration of negotiations between the nations that make up Canada. It explores the disparity that remains between Aboriginals and non Aboriginals in Canadian North America at a systemic level. It will show that the postcolonial era is rampant with colonial doctrine and that these principles and policies maintain a dogmatic system that can not allow for the continued existence of Aboriginals as separate and distinct peoples. I will show my understanding and interpretation of an old Indigenous system and suggest ways in which aspects of this ancient system may be valuable in creating a coordination of world views that can allow for both factions to exist and prosper. I will specifically address how the differing world views that exist between Aboriginal and non-Aboriginal Canadians—and the inequality between these two groups of peoples—has been and remains infused in the negotiation process that these governments attempt to complete. The final aspect of this work will be a theatrical production piece that allows (in some small way) the traditional Indigenous approach to ‘law’ to be given equal weight as the Supreme Court in Delgamuukw suggests.

An einbahnige Außerortsstraßen (Landstraßen) werden sehr hohe Anforderungen gestellt. Einerseits sollen sie eine hohe Verkehrssicherheit aufweisen und andererseits gemäß ihrer Verkehrsbedeutung eine angemessene Verkehrsqualität und Leistungsfähigkeit bereitstellen. Im Landstraßennetz liegt die Erfüllung dieser Anforderungen jedoch vielerorts deutlich unter den Erwartungen. Unfälle auf Landstraßen sind im Vergleich zu Autobahnen oder Innerortsstraßen durch eine besonders hohe Schwere gekennzeichnet. Sehr hohe und oft an den Streckenverlauf nicht hinreichend angepasste Geschwindigkeiten, Fehleinschätzungen von Geschwindigkeiten bzw. Geschwindigkeitsdifferenzen entgegenkommender und vorausfahrender Fahrzeuge sowie die im Standardfall für Überholvorgänge notwendige Nutzung des Gegenfahrstreifens sind die Hauptursachen für die schweren Unfälle auf Landstraßen. Das Ziel der vorliegenden Arbeit war die Untersuchung der Wirksamkeit von kurz- und mittelfristig umsetzbaren Maßnahmen zur Verbesserung der Verkehrssicherheit auf unfallauffälligen einbahnigen Außerortsstraßen. Die Untersuchungen dafür erfolgten im Rahmen des Großversuches „Außerortsstraßensicherheit“ (AOSI). Im Ergebnis sollten Einsatzempfehlungen für die Planungs- und Entwurfspraxis und für zukünftige Regelwerke für den Landstraßenentwurf abgeleitet werden. Für die Durchsetzung der zulässigen Höchstgeschwindigkeit wurden auf fünf ausgewählten Bundesstraßenabschnitten linienhaft angeordnete ortsfeste Geschwindigkeitsüberwachungsanlagen (OGÜ) eingerichtet. Auf fünf weiteren Bundesstraßenabschnitten wurde das Überholen nur auf den dafür angelegten Überholfahrstreifen (ÜFS) zugelassen. Auf den verbliebenen zweistreifigen Zwischenabschnitten wurden Überholverbote (ÜV) erlassen. Nach dem von der AOSI-Projektgruppe vorgegebenen Messregime und den so ermittelten Rohdaten wurde eine eigene Untersuchungsmethodik entwickelt und die vorliegende Datenlage entsprechend aufbereitet. Darauf gestützt erfolgte die Auswertung und Interpretation sowie die Ableitung der Empfehlungen. Durch die linienhafte ortsfeste Geschwindigkeitsüberwachung konnte eine deutliche Verbesserung der Verkehrssicherheit erzielt werden und damit das Unfallrisiko auf den Untersuchungsstrecken teils erheblich reduziert werden. Es konnte gezeigt werden, dass die Grundlage dieser Entwicklung die Durchsetzung der zulässigen Höchstgeschwindigkeit war, die vor der Überwachung um bis zu 20 km/h überschritten wurde. Diese Entwicklung war nicht nur auf eine Reduktion der Fahrunfälle zurückzuführen. Durch die Verringerung der Geschwindigkeitsunterschiede zwischen den Fahrzeugen sank der Überholdruck, was besonders auf den Untersuchungsstrecken mit einer gestreckten Linienführung auch einen Rückgang der Unfälle im Längsverkehr bewirkte. Nach der Eingewöhnungszeit stellte sich das erwartete Fahrverhalten entlang der Untersuchungsstrecken ein, das nicht mehr durch deutliche Verzögerungs- und Beschleunigungsvorgänge im Bereich der Geschwindigkeitsüberwachungsanlagen geprägt war. Förderlich für diese Entwicklung war trotz anfänglicher Skepsis die breite Akzeptanz der Kraftfahrer zu den OGÜ-Anlagen als sinnvolle Maßnahme zur Verbesserung der Verkehrssicherheit. Erreicht wurde dies mit einer verständlichen Öffentlichkeitsarbeit zu den Hintergründen und Zielen der Maßnahme. Die Kombination von abschnittsweisen sicheren Überholmöglichkeiten (ÜFS) und Überholverboten in den einbahnig zweistreifigen Zwischenabschnitten (ÜV) ist eine geeignete, mittelfristig umsetzbare Maßnahme zur Verbesserung der Verkehrssicherheit. Die Wirkung der Maßnahme ist vorrangig auf die Unfälle im Längsverkehr gerichtet und sollte daher auf Strecken erwogen werden, bei denen das Unfallgeschehen vor allem durch einen hohen Anteil von Überholunfällen gekennzeichnet ist. Der Vorteil liegt im Abbau des Überholdrucks unabhängig von den verfügbaren Zeitlücken im Gegenverkehr. Dafür haben sich auch kurze Überholfahrstreifen (l ≥ 600 m) bewährt, die im Bestandsnetz eher realisierbar sind. Die untersuchte Maßnahmenkombination aus einem sicheren Überholangebot und einem Überholverbot in den einbahnig zweistreifigen Abschnitten erfährt zudem eine hohe Akzeptanz bei den Verkehrsteilnehmern. Durch die Ergebnisse dieser Arbeit liegen Empfehlungen vor, um die Verkehrssicherheit auf unfallauffälligen Abschnitten einbahniger Außerortsstraßen kurz- und mittelfristig zu verbessern. Dies gilt vor allem dann, wenn das Unfallgeschehen vorrangig auf eine zu hohe und unangepasste Geschwindigkeit oder auf Fehler bei Überholvorgängen zurückzuführen ist. Die Ergebnisse liefern darüber hinaus belastbare Grundlagen für das neue Regelwerk für den Landstraßenentwurf.

De nos jours, les systèmes de vision sont de plus en plus utilisés dans le contexte routier. Ils permettent ainsi d'assurer la sécurité et faciliter la mobilité. Ces systèmes de vision sont généralement affectés par la dégradation des conditions météorologiques en présence de brouillard ou de pluie forte, phénomènes limitant la visibilité et réduisant ainsi la qualité des images. Afin d'optimiser les performances des systèmes de vision, il est nécessaire de disposer d'un système de détection fiable de ces conditions météorologiques défavorables.Il existe des capteurs météorologiques dédiés à la mesure physique, mais ils sont coûteux. Ce problème peut être résolu en utilisant les caméras qui sont déjà installées sur les routes. Ces dernières peuvent remplir simultanément deux fonctions : l'acquisition d'images pour les applications de surveillance et la mesure physique des conditions météorologiques au lieu des capteurs dédiés. Suite au grand succès des réseaux de neurones convolutifs (CNN) dans la classification et la reconnaissance d'images, nous avons utilisé une méthode d'apprentissage profond pour étudier le problème de la classification météorologique. L'objectif de notre étude est de chercher dans un premier temps à mettre au point un classifieur du temps, qui permet de discriminer entre temps « normal », brouillard et pluie. Dans un deuxième temps, une fois la classe connue, nous cherchons à développer un modèle de mesure de la distance de visibilité météorologique du brouillard. Rappelons que l'utilisation des CNN exige l'utilisation de bases de données d'apprentissage et de test. Pour cela, deux bases de données ont été utilisées, "Cerema-AWP database" (https://ceremadlcfmds.wixsite.com/cerema-databases), et la base "Cerema-AWH database", en cours d'acquisition depuis 2017 sur le site de la Fageole sur l'autoroute A75. Chaque image des deux bases est labellisée automatiquement grâce aux données météorologiques relevées sur le site permettant de caractériser diverses gammes de pluie et de brouillard. La base Cerema-AWH, qui a été mise en place dans le cadre de nos travaux, contient cinq sous-bases : conditions normales de jour, brouillard fort, brouillard faible, pluie forte et pluie faible. Les intensités de pluie varient de 0 mm/h à 70 mm/h et les visibilités météorologiques de brouillard varient entre 50m et 1800m. Parmi les réseaux de neurones connus et qui ont montré leur performance dans le domaine de la reconnaissance et la classification, nous pouvons citer LeNet, ResNet-152, Inception-v4 et DenseNet-121. Nous avons appliqué ces réseaux dans notre système de classification des conditions météorologiques dégradées. En premier lieu, une étude justificative de l'usage des réseaux de neurones convolutifs est effectuée. Elle étudie la nature de la donnée d'entrée et les hyperparamètres optimaux qu'il faut utiliser pour aboutir aux meilleurs résultats. Ensuite, une analyse des différentes composantes d'un réseau de neurones est menée en construisant une architecture instrumentale de réseau de neurones. La classification des conditions météorologiques avec les réseaux de neurones profonds a atteint un score de 83% pour une classification de cinq classes et 99% pour une classification de trois classes.Ensuite, une analyse sur les données d'entrée et de sortie a été faite permettant d'étudier l'impact du changement de scènes et celui du nombre de données d'entrée et du nombre de classes météorologiques sur le résultat de classification.Enfin, une méthode de transfert de bases de données a été appliquée. Cette méthode permet d'étudier la portabilité du système de classification des conditions météorologiques d'un site à un autre. Un score de classification de 63% a été obtenu en faisant un transfert entre une base publique et la base Cerema-AWH. (...)
Nowadays, vision systems are becoming more and more used in the road context. They ensure safety and facilitate mobility. These vision systems are generally affected by the degradation of weather conditions, like heavy fog or strong rain, phenomena limiting the visibility and thus reducing the quality of the images. In order to optimize the performance of the vision systems, it is necessary to have a reliable detection system for these adverse weather conditions.There are meteorological sensors dedicated to physical measurement, but they are expensive. Since cameras are already installed on the road, they can simultaneously perform two functions: image acquisition for surveillance applications and physical measurement of weather conditions instead of dedicated sensors. Following the great success of convolutional neural networks (CNN) in classification and image recognition, we used a deep learning method to study the problem of meteorological classification. The objective of our study is to first seek to develop a classifier of time, which discriminates between "normal" conditions, fog and rain. In a second step, once the class is known, we seek to develop a model for measuring meteorological visibility.The use of CNN requires the use of train and test databases. For this, two databases were used, "Cerema-AWP database" (https://ceremadlcfmds.wixsite.com/cerema-databases), and the "Cerema-AWH database", which has been acquired since 2017 on the Fageole site on the highway A75. Each image of the two bases is labeled automatically thanks to meteorological data collected on the site to characterize various levels of precipitation for rain and fog.The Cerema-AWH base, which was set up as part of our work, contains 5 sub-bases: normal day conditions, heavy fog, light fog, heavy rain and light rain. Rainfall intensities range from 0 mm/h to 70mm/h and fog weather visibilities range from 50m to 1800m. Among the known neural networks that have demonstrated their performance in the field of recognition and classification, we can cite LeNet, ResNet-152, Inception-v4 and DenseNet-121. We have applied these networks in our adverse weather classification system. We start by the study of the use of convolutional neural networks. The nature of the input data and the optimal hyper-parameters that must be used to achieve the best results. An analysis of the different components of a neural network is done by constructing an instrumental neural network architecture. The conclusions drawn from this analysis show that we must use deep neural networks. This type of network is able to classify five meteorological classes of Cerema-AWH base with a classification score of 83% and three meteorological classes with a score of 99%Then, an analysis of the input and output data was made to study the impact of scenes change, the input's data and the meteorological classes number on the classification result.Finally, a database transfer method is developed. We study the portability from one site to another of our adverse weather conditions classification system. A classification score of 63% by making a transfer between a public database and Cerema-AWH database is obtained.After the classification, the second step of our study is to measure the meteorological visibility of the fog. For this, we use a neural network that generates continuous values. Two fog variants were tested: light and heavy fog combined and heavy fog (road fog) only. The evaluation of the result is done using a correlation coefficient R² between the real values and the predicted values. We compare this coefficient with the correlation coefficient between the two sensors used to measure the weather visibility on site. Among the results obtained and more specifically for road fog, the correlation coefficient reaches a value of 0.74 which is close to the physical sensors value (0.76)

Diploma thesis deals with the possibilities of using video records in analysis road accidents. There are described static and dynamic camera systems. There are also explained the principles of functioning digital cameras and other related terms and associated principles. In this thesis there are compared several types of dynamic cameras, desribed analysis an evaluation procedure. Part of this thesis is to describe the legal issues of recording.

This thesis describes the design and subsequent implementation of a unit for traffic monitoring using Raspberry PI. First section provides a quick overview of assistance systems, which use a road lane detection techniques. Next there is a description of two diferent methods for road lane detection. Follow the description of monitoring scene. Then the work describe the practical part including the design and realization of supporting electronics, selecting of each components, including the modifying of cameras, mechanical design and creating of unit. Another section is about selection and installation of appropriate software components necessary for running of the unit and the selection of development tools for creating user application. After description of graphical user interafce, there is a description of road lanes detection algorithm. At the end of the thesis is summarized a reliability of unit in real traffic situation. At the appendix there are technical drawings, describing the unit.

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

La fragmentation des paysages et des habitats induite par les infrastructures linéaires de transport terrestres est une des principales causes de la perte de biodiversité actuelle. Parmi ces infrastructures, la route est un acteur majeur de fragmentation, d’autant plus qu’elle possède des effets propres dus au trafic circulant qui induit des collisions véhicule-faune et une pollution des paysages. Afin de diminuer ces effets négatifs, des mesures de réduction sont mises en place, notamment des passages à faune permettant de faire traverser la faune de part et d’autre des voies. La route crée aussi de nouveaux habitats potentiels pour les espèces de la petite faune dans des paysages anthropisés et fragmentés. Dans ce mémoire sont démontrées (1) la potentialité d’habitat de différents éléments routiers ; (2) la possibilité de prédire les collisions véhicule-faune afin de positionner au mieux les mesures de réduction ; (3) l’importance de la méthodologie dans l’évaluation de l’efficacité des passages à faune ; et (4) la possibilité d’améliorer les passages à faune existants. Ces résultats permettront d’améliorer les stratégies de défragmentation des paysages
Habitats and landscape fragmentation, caused by linear land transports infrastructures, is one of the major cause for the current loss of biodiversity. Among those infrastructures, road is a major cause of fragmentation, especially as it possess specific traffic-linked effects, which induces wildlife-vehicles collisions and landscape pollution. In order to decrease those negative effects, mitigation measures are taken, among which wildlife crossings, enabling wildlife to cross the road. Road also creates new potential habitats for small wildlife species in anthropogenic and fragmented landscapes. In this essay are shown (1) the potential as habitat of different road-linked elements; (2) the possibility to anticipate wildlife-vehicles collisions in order to improve the position of mitigation measures; (3) the importance of methodology in the evaluation of wildlife crossings effectiveness; and (4) the possibility to improve existing wildlife crossings. Those results will allow improving landscape defragmentation strategies

My dissertation critiques Western culture’s vertical command of “growing up” to adult completion (rational, heterosexual, married, wealthy, professionally successful) as a reductionist itinerary of human movement leading to subjective sedimentations. Rather, my project proposes ways of “moving rhizomatically” by which it advances a notion of a machinic identity that moves continuously, contingently, and waywardly along less vertical, less excruciating and more horizontal, life-affirmative trails. To this end, my thesis proposes a “rhizomatic semiosis” as extrapolated from the philosophy of Gilles Deleuze and Félix Guattari to put forward a notion of language and, by implication, subjectivity, as dynamic and metamorphic. Rather than trying to figure out who the child is or what it experiences consciously, my project wishes to embrace an elusiveness at the heart of subjectivity to argue for continued identity creation beyond the apparently confining parameters of adulthood. This dissertation, then, is about the need to re-examine our ways of growing beyond the lines of teleological progression. By turning to Deleuze’s child, an intangible one that “makes desperate attempts to carry out a performance that the psychoanalyst totally misconstrues” (A Thousand Plateaus 13), I wish to shift focus away from the hierarchical, binary, and ideal model of “growing up” and toward a notion of movement that makes way for plural identities in their becoming. This endeavour reveals itself in particular in the work of John Wray, Todd Field, Peter Cameron, Sara Prichard, Michael Cunningham, and Cormac McCarthy, whose work has received little or no attention at all—a lacuna in research that exists perhaps due to these artists’ innovative approach to a minor literature that promotes the notion of a machinic self and questions the dominant modes of Western culture’s literature for, around, and of children.

Road detection is a crucial concern in Autonomous Navigation and Driving Assistance. Despite the multiple existing algorithms to detect the road, the literature does not offer a single effective algorithm for all situations. A global more robust set-up would count on multiple distinct algorithms running in parallel, or even from multiple cameras. Then, all these algorithms’ outputs should be merged or combined to produce a more robust and informed detection of the road lane, so that it works in more situations than each algorithm by itself. This dissertation integrated in the ATLAS-CAR2 project, developed at the University of Aveiro, proposes a ROS-based architecture to manage and combine multiple sources of lane detection algorithms ranging from the algorithms that return the spatial localization of the road lane lines and those whose results are the navigable zone represented as a polygon. The architecture is fully scalable and has proved to be a valuable tool to test and parametrise individual algorithms. The combination of the algorithms’ results used in this work uses a confidence based merging of individual detections.
A deteção de estradas é uma questão crucial na Navegação Autónoma e na Assistência à Condução. Apesar de os múltiplos algoritmos existentes para detetar a estrada, a literatura não oferece um único algoritmo eficaz para todas as situações. Uma configuração global mais robusta incorporaria vários algoritmos distintos e executados em paralelo, ou mesmo baseado em múltiplas câmaras. Então, todos os resultados destes algoritmos devem ser fundidos ou combinados para produzir uma deteção mais robusta e informada da via da estrada, para que funcione em mais situações do que cada algoritmo funcionando individualmente. Esta dissertação integrada no projeto ATLASCAR2, desenvolvido na Universidade de Aveiro, propõe uma arquitetura baseada em ROS para gerir e combinar múltiplas fontes de algoritmos de deteção de vias da estrada, desde algoritmos que devolvem a localização espacial da faixa de rodagem até àqueles cujos resultados são a zona navegável representada como um polı́gono. A arquitetura é totalmente escalável e provou ser uma ferramenta valiosa para testar e parametrizar algoritmos individuais. A combinação dos resultados dos algoritmos utilizados neste trabalho utiliza uma combinação de deteções individuais baseada na confiança.
Mestrado em Engenharia Mecânica

Over the past century, populations of North American grassland songbirds have declined sharply as a consequence of habitat destruction. Alberta’s mixed-grass prairie constitutes Canada’s largest remaining tract of native grassland. However, this region has recently undergone a rapid expansion of conventional oil and natural gas development, and few studies have documented its effects on songbird nesting success. During the 2012-2014 breeding seasons, I monitored 813 nests of grassland songbirds located at sites that varied with respect to presence/absence, distance from, and types of oil and gas infrastructure (pump jacks, screw pumps, compressor stations) and gravel roads. Nest survival was significantly lower at infrastructure sites relative to controls for both Savannah sparrow and vesper sparrow. Additionally, vesper sparrow nest density was greater within 100 m of structures. These findings suggest that habitat disturbance caused by infrastructure may result in increased frequencies of nest predation at multiple spatial scales.
February 2016

碩士
國立虎尾科技大學
光電與材料科技研究所
101
The present thesis aims to provide a road safety warning as well as recording and monitoring device that can be used both for daytime visual inspection and nighttime safety warning enhancement so that the weaknesses inherent in the technical structure of the existing monitoring devices are improved. Thus, the road safety warning as well as recording and monitoring device of the present thesis is designed to warn road users about road safety conditions and to monitor and record road conditions regardless of different weather conditions: ample or insufficient daylight and poor night vision. The device also uses monitoring and recording technologies to clearly record any road conditions so as to raise alertness among road users and avoid possible dispute between relevant parties since the road conditions are recorded whenever an accident occurs. This is the major issue the present study aims to tackle. The present study is not simply a research project on paper, but an applicable product that addresses the weaknesses derived from the existing technical structure and can be directly used for road safety warning, monitoring and recording tasks.

This thesis is focused on explaining the behaviour of population of Aesculapian Snake (Zamenis longissimus) in Poohří region related to a busy road which crosses its area of distribution. To avoid the busy road Aesculapian Snake uses the roads culverts to cross the road safely. Snakes used culverts the most often on July, when their activity culminates. One of the most used culverts was culvert Nr. 2, probably because of its proximity to man-made hatch. Snakes started their activity at 8:00 a.m. and finished at 7:00 p.m. In this study their activity culminates between 4:00 p.m. and 6:00 p.m. and by temperature between 21 - 25°C. None of adults of Aesculapian Snake was detected killed on the road. There were found only juvenile snakes of this species which weren't acquainted yet with local threats.

碩士
朝陽科技大學
資訊工程系碩士班
95
The intelligent security surveillance system is actively pushed by the governments in Taiwan and other countries in the last few years. With the high technology, it provides the traditional security surveillance system with higher application. Thus a car license plate recognition system based on the digital image processing technique recently becomes a popular research topic. The car license plate recognition system includes three parts: License-Plate Location, Character Extraction, and License-Plate Recognition. In this paper, we derive a new method of the automatic car license plate recognition system in a safely monitored environment. To License-Plate Location, we analyze and describe the features of a car license plate through independent component analysis. We can obtain the features of the car license plate through the independent component analysis filters. The result shows that the average accuracy can be up to 94.3% in different weather and situations. It is higher than the traditional edge-based method which is about 80.6%. We proposed the new dynamic threshold methods for character extraction. It can mostly finish the character extraction in the conditions of shadow variety and contamination of the license plate character. To License-Plate Recognition, we proposed a method to rapidly classify the character based on the structure of character. We can real-time and accurately achieve the character recognition with template matching method. It shows that in the road of high variety environment, the average accuracy of the License-Plate Recognition is 86.3%.

碩士
國立交通大學
電控工程研究所
100
Visual-based vehicle detection techniques applied to Intelligent Transportation System (ITS) to improve the efficiency and precision of analyzing heavy video information have been studied for years. In this study, we take advantage of AdaBoost algorithm’s accurate detection rate cooperate with Gaussian Mixture Model (GMM) and calibration method we proposed to do classification. We transform the raw color image into a gray level image and resize it for acceleration. After resizing, the image is send to Gaussian mixture model (GMM) approach to obtain reliable background images, and then use background subtraction technique to extract foreground objects. In this stage, we use calibration for transferring pixel coordinates to world coordinates and cooperate with foreground subtracted by GMM to get and record foreground’s information, such as width and height for further classification. Here, we use concept of overlapping area for tracking and recording foreground’s information. For classification’s criteria, the AdaBoost vehicle detector uses slicing window to detect vehicles and verifies whether the foreground subtracted by GMM [22] [23] is vehicle. Through AdaBoost vehicle detector, we can separate vehicles from pedestrians, motorcycles, and objects. Next, we utilize width and height ratio, edge complexity, and speed get from calibration to separate pedestrians and motorcycles from objects. Because we found that pedestrians and objects have different edge complexity range. The objective of this study is to classify vehicles, pedestrians, and objects which include AdaBoost vehicle detector and calibrating the foreground’s width, height, and speed. Therefore, the system can warns us of the dangerous situation to protect both drivers and pedestrians. The experimental results proved the proposed system achieved a good performance of classifying vehicles, pedestrians, and objects. The implemented system also extracted useful traffic information that can be used for further processing, like classifying vehicles, ex: sedan, truck, bus, and classifying pedestrians and motorcycles by adding auxiliary features.

碩士
聖約翰科技大學
電機工程系碩士班
107
This thesis attempts to develop a system with RGB and FIR (Far Infrared) cameras and road object recognition function. It consists of a set of RGB cameras and FIR cameras and other peripherals, and uses the Linux platform-based computing system platform for road object recognition. First of all, this thesis uses Intel i7-9700K processor platform as the core of the system, RGB lens adopts GOPRO HERO7, and FIR lens adopts far infrared lens produced by American FLIR. The RGB camera and the FIR camera are used to capture the road image to be recognized, and the road object is transmitted to the image processing unit by the system platform, and the RGB image is transmitted through the deep learning model trained by the YOLOV3 algorithm of the Intel OpenVINO identification system. FIR images are instantly identified. The road object detection and identification system developed in this thesis can improve the unrecognizable state of RGB in harsh weather or poor light environment, etc. Through thermal imaging, avoiding the situation of strong light or poor light interference, the FIR thermal image is used as a compensation scheme for assisting LiDAR, Radar and RGB lens to increase the driving safety level of the autonomous vehicle. Finally, in order to verify the road object recognition system developed in this thesis, the road shooting environment is actually used for object recognition testing, including general daytime, nighttime, and rainy days that are prone to accidents, in and out of tunnels, and backlighting of opposite lanes. The results show that the road object recognition system of this thesis has a very good improvement effect.

碩士
國立中央大學
資訊工程學系
105
The development of road transportation makes it easier for people to drive their vehicles. Unfortunately, it also causes the increasing number of traffic accidents. Advanced driver assistance systems, which in many ways is based on the future trajectory prediction of the vehicle, are developed to alert the driver for potential dangers. Road conditions, including road geometry and the distance between lead vehicle and host vehicle, are important factors in improving accuracy of future trajectory prediction. The road conditions can be obtained by the camera affixed in the vehicle. In this thesis, we propose a image processing system, which includes a curve detection algorithm (CDA) and a distance conversion (DC) algorithm, to obtain these road conditions from the image. First, CDA detects the lane stripes and calculates the vanishing point. The road trend can then be identified according to the location of the vanishing point. DC is used to convert the distance between the lead and host vehicles in the image to the real distance. Through analyses and experiments, it is shown that the proposed system achieves a higher precision than the baseline algorithms.

I investigated the value of under-road tunnels as a conservation strategy to protect a long-toed salamander population, in south-west Alberta, whose overwintering sites and breeding habitat (Linnet Lake) are separated by a road. I conducted a mark-recapture study from 2008-2009, capturing salamanders using roadside fences and pitfall traps. Four tunnels were monitored in 2009 using traps and cameras. A 2008 estimate indicated that the population declined by 60% since 1994, however, road mortality was dramatically reduced following installation of fences and tunnels. Camera and trap data documented 130 salamanders navigating tunnels in 2009. I found little evidence of juvenile recruitment from Linnet Lake, likely because of predation by lake chub. Experiments showed that lake chub consumed salamander larvae, and fish presence altered larval behaviour. Continued monitoring is needed to determine if reduced road mortality translates into population gains, and whether fish predation threatens the persistence of the long-toed salamander population.
Ecology

Road expansion and increased traffic likely exacerbates barriers to amphibian migration and dispersal. Within British Columbia’s south Okanagan valley there is particular concern that the COSEWIC-listed blotched tiger salamander (Ambystoma mavortium melanostictum) and Great Basin spadefoot (Spea intermontana) are vulnerable to road effects in their annual movements from upland overwintering habitat to lowland breeding areas. My study utilizes a before after control impact approach to assess amphibian movement and population threats across this highway-bisected landscape. Throughout the spring and summer of 2010-2012, fifty two kilometers of roadways (31 km of highway, 21 km of paved backroad) were repeatedly surveyed from the Canada-USA border to north of Oliver, BC; surveys were carried out utilising vehicles and on foot. Along Highway 97, a three kilometer four-lane highway expansion project was constructed through 2010 and open to traffic use in 2011. Adjacent to a floodplain, survey effort was focused throughout this transect for informed roadkill mitigation structure placement and ongoing ecopassage effectiveness monitoring. Automated camera trap monitoring of culverts within highly concentrated amphibian road hotspots during spring and summer 2011 (three culverts) and 2012 (two culverts) resulted in over eight hundred amphibian culvert events observed. Two sample Wilcoxon tests revealed differences between years in amphibian occurrence between 2010 and 2012 (W = 4679.5, p= 0.02), and mortalities among transect areas, with the largest differences between years within the Osoyoos passing lanes transect. Amphibian mortalities within the passing lanes transect were significantly reduced with the implementation of mitigation structures (x̅2010= 13.2 ± 32.5, x̅2011= 4.7 ± 12.8, x̅2012= 2.3 ± 7.3; 2010 vs. 2012: W= 1535.5, p< 0.001). Roadkill mitigation structures proved effective in observed amphibian occurrence of the entire passing lanes stretch as well as at distances 100 m and 200 m from observed culverts. Double fenced areas resulted in a 94% reduction in amphibian road occurrence. Five species of amphibians were observed over the three survey years (4051 road incidences over 657 survey hours): Pacific chorus frog (Pseudacris regilla), Western toad (Anaxyrus boreas), long-toed salamander (Ambystoma macrodactylum) plus blotched tiger salamander and Great Basin spadefoot. This study aims to provide a better understanding of amphibian hotspots on roadways and ecopassage use within the south Okanagan. It may act as a catalyst to further wildlife-vehicle interaction studies with improved mitigation solutions for amphibian roadway fatalities.