To see the other types of publications on this topic, follow the link: Phone recognition.
Dissertations / Theses on the topic 'Phone recognition'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Phone recognition.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Olausson, Erik. "Face Recognition for Mobile Phone Applications." Thesis, Linköping University, Department of Science and Technology, 2008. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-11850.
Full textAbstract:
Att applicera ansiktsigenkänning direkt på en mobiltelefon är en utmanande uppgift, inte minst med tanke på den begränsade minnes- och processorkapaciteten samt den stora variationen med avseende på ansiktsuttryck, hållning och ljusförhållande i inmatade bilder.
Det är fortfarande långt kvar till ett färdigutvecklat, robust och helautomatiskt ansiktsigenkänningssystem för den här miljön. Men resultaten i det här arbetet visar att genom att plocka ut feature-värden från lokala regioner samt applicera en välgjord warpstrategi för att minska problemen med variationer i position och rotation av huvudet, är det möjligt att uppnå rimliga och användbara igenkänningsnivåer. Speciellt för ett halvautomatiskt system där användaren har sista ordet om vem personen på bilden faktiskt är.
Med ett galleri bestående av 85 personer och endast en referensbild per person nådde systemet en igenkänningsgrad på 60% på en svårklassificerad serie testbilder. Totalt 73% av gångerna var den rätta individen inom de fyra främsta gissningarna.
Att lägga till extra referensbilder till galleriet höjer igenkänningsgraden rejält, till nästan 75% för helt korrekta gissningar och till 83,5% för topp fyra. Detta visar att en strategi där inmatade bilder läggs till som referensbilder i galleriet efterhand som de identifieras skulle löna sig ordentligt och göra systemet bättre efter hand likt en inlärningsprocess.
Detta exjobb belönades med pris för "Bästa industrirelevanta bidrag" vid Svenska sällskapet för automatiserad bildanalys årliga konferens i Lund, 13-14 mars 2008.
Applying face recognition directly on a mobile phone is a challenging proposal due to the unrestrained nature of input images and limitations in memory and processor capabilities.
A robust, fully automatic recognition system for this environment is still a far way off. However, results show that using local feature extraction and a warping scheme to reduce pose variation problems, it is possible to capitalize on high error tolerance and reach reasonable recognition rates, especially for a semi-automatic classification system where the user has the final say.
With a gallery of 85 individuals and only one gallery image per individual available the system is able to recognize close to 60 % of the faces in a very challenging test set, while the correct individual is in the top four guesses 73% of the time.
Adding extra reference images boosts performance to nearly 75% correct recognition and 83.5% in the top four guesses. This suggests a strategy where extra reference images are added one by one after correct classification, mimicking an online learning strategy.
2
Qin, Yinghao. "The Smart Phone as a Mouse." The University of Waikato, 2006. http://hdl.handle.net/10289/2289.
Full textAbstract:
With the development of hardware, mobile phone has become a feature-rich handheld device. Built-in camera and Bluetooth technology are supported in most current mobile phones. A real-time image processing experiment was conducted with a SonyEricsson P910i smartphone and a desktop computer. This thesis describes the design and implementation of a system which uses a mobile phone as a PC mouse. The movement of the mobile phone can be detected by analyzing the images captured by the onboard camera and the mouse cursor in the PC can be controlled by the movement of the phone.
3
Ghosh, Anubhab. "Normalizing Flow based Hidden Markov Models for Phone Recognition." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-286594.
Full textAbstract:
The task of Phone recognition is a fundamental task in Speech recognition and often serves a critical role in bench-marking purposes. Researchers have used a variety of models used in the past to address this task, using both generative and discriminative learning approaches. Among them, generative approaches such as the use of Gaussian mixture model-based hidden Markov models are always favored because of their mathematical tractability. However, the use of generative models such as hidden Markov models and its hybrid varieties is no longer in fashion owing to a large inclination to discriminative learning approaches, which have been found to perform better. The only downside is that these approaches do not always ensure mathematical tractability or convergence guarantees as opposed to their generative counterparts. So, the research problem was to investigate whether there could be a process of augmenting the modeling capability of generative Models using a kind of neural network based architectures that could simultaneously prove mathematically tractable and expressive. Normalizing flows are a class of generative models that have been garnered a lot of attention recently in the field of density estimation and offer a method for exact likelihood computation and inference. In this project, a few varieties of Normalizing flow-based hidden Markov models were used for the task of Phone recognition on the TIMIT dataset. It was been found that these models and their mixture model varieties outperformed classical generative model varieties like Gaussian mixture models. A decision fusion approach using classical Gaussian and Normalizing flow-based mixtures showed competitive results compared to discriminative learning approaches. Further analysis based on classes of speech phones was carried out to compare the generative models used. Additionally, a study of the robustness of these algorithms to noisy speech conditions was also carried out.Uppgiften för fonemigenkänning är en grundläggande uppgift i taligenkänning och tjänar ofta en kritisk roll i benchmarkingändamål. Forskare har använt en mängd olika modeller som använts tidigare för att hantera denna uppgift genom att använda både generativa och diskriminerande inlärningssätt. Bland dem är generativa tillvägagångssätt som användning av Gaussian-blandnings modellbaserade dolda Markov-modeller alltid föredragna på grund av deras matematiska spårbarhet. Men användningen av generativa modeller som dolda Markov-modeller och dess hybridvarianter är inte längre på mode på grund av en stor lutning till diskriminerande inlärningsmetoder, som har visat sig fungera bättre. Den enda nackdelen är att dessa tillvägagångssätt inte alltid säkerställer matematisk spårbarhet eller konvergensgarantier i motsats till deras generativa motsvarigheter. Således var forskningsproblemet att undersöka om det kan finnas en process för att förstärka modelleringsförmågan hos generativa modeller med hjälp av ett slags neurala nätverksbaserade arkitekturer som samtidigt kunde visa sig matematiskt spårbart och uttrycksfullt. Normaliseringsflöden är en klass generativa modeller som nyligen har fått mycket uppmärksamhet inom området för densitetsberäkning och erbjuder en metod för exakt sannolikhetsberäkning och slutsats. I detta projekt användes några få varianter av Normaliserande flödesbaserade dolda Markov-modeller för uppgiften att fonemigenkänna i TIMIT-datasatsen. Det visade sig att dessa modeller och deras blandningsmodellvarianter överträffade klassiska generativa modellvarianter som Gaussiska blandningsmodeller. Ett beslutssmältningsstrategi med klassiska Gaussiska och Normaliserande flödesbaserade blandningar visade konkurrenskraftiga resultat jämfört med diskriminerande inlärningsmetoder. Ytterligare analys baserat på klasser av talsignaler utfördes för att jämföra de generativa modellerna som användes. Dessutom genomfördes en studie av robustheten hos dessa algoritmer till bullriga talförhållanden.
4
Stearns, Cameron P. cstearns. "A SYSTEM FOR CELL PHONE ANTI-THEFT THROUGH GAIT RECOGNITION." DigitalCommons@CalPoly, 2014. https://digitalcommons.calpoly.edu/theses/1216.
Full textAbstract:
Studies show that smartphone thefts are a significant problem in the United States. [30] With many upcoming proposals to decrease the theft-rate of such devices, investigating new techniques for preventing smartphone theft is an important area of research. The prevalence of new biometric identification techniques for smartphones has led some researchers to propose biometric anti-theft measures for such devices, similar to the current fingerprint authentication system for iOS. Gait identification, a relatively recent field of study, seems to be a good fit for anti-theft because of the non-intrusive nature of passive pattern recognition in walking. In this paper, we reproduce and extend a modern gait recognition technique proposed in Cell Phone-Based Biometrics by testing the technique outside of the laboratory on real users under everyday conditions. We propose how this technique can be applied to create an anti-theft system, and we discuss future developments that will be necessary before such research is ready to be implemented in a release-quality product. Because previous studies have also centered around the ability to differentiate between individual users from a group, we will examine the accuracy of identifying whether or not a specific user is currently using a system. The system proposed in this paper shows results as high as 91% for cross-fold accuracy for some users; however, the predictive accuracy for a single day’s results ranged from 0.8% accuracy to 92.9% accuracy, showing an unreliability that makes such a system unlikely to be useful under the pressure of real-world conditions.
5
Chou, Christine S. (Christine Susan). "Language identification through parallel phone recognition dc by Christine S. Chou." Thesis, Massachusetts Institute of Technology, 1994. http://hdl.handle.net/1721.1/34056.
Full text
6
Mohapatra, Prateeti. "Deriving Novel Posterior Feature Spaces For Conditional Random Field - Based Phone Recognition." The Ohio State University, 2009. http://rave.ohiolink.edu/etdc/view?acc_num=osu1236784133.
Full text
7
Mohammed, Abdulmalik. "Obstacle detection and emergency exit sign recognition for autonomous navigation using camera phone." Thesis, University of Manchester, 2017. https://www.research.manchester.ac.uk/portal/en/theses/obstacle-detection-and-emergency-exit-sign-recognition-for-autonomous-navigation-using-camera-phone(e0224d89-e743-47a4-8c68-52f718457098).html.
Full textAbstract:
In this research work, we develop an obstacle detection and emergency exit sign recognition system on a mobile phone by extending the feature from accelerated segment test detector with Harris corner filter. The first step often required for many vision based applications is the detection of objects of interest in an image. Hence, in this research work, we introduce emergency exit sign detection method using colour histogram. The hue and saturation component of an HSV colour model are processed into features to build a 2D colour histogram. We backproject a 2D colour histogram to detect emergency exit sign from a captured image as the first task required before performing emergency exit sign recognition. The result of classification shows that the 2D histogram is fast and can discriminate between objects and background with accuracy. One of the challenges confronting object recognition methods is the type of image feature to compute. In this work therefore, we present two feature detectors and descriptor methods based on the feature from accelerated segment test detector with Harris corner filter. The first method is called Upright FAST-Harris and binary detector (U-FaHB), while the second method Scale Interpolated FAST-Harris and Binary (SIFaHB). In both methods, feature points are extracted using the accelerated segment test detectors and Harris filter to return the strongest corner points as features. However, in the case of SIFaHB, the extraction of feature points is done across the image plane and along the scale-space. The modular design of these detectors allows for the integration of descriptors of any kind. Therefore, we combine these detectors with binary test descriptor like BRIEF to compute feature regions. These detectors and the combined descriptor are evaluated using different images observed under various geometric and photometric transformations and the performance is compared with other detectors and descriptors. The results obtained show that our proposed feature detector and descriptor method is fast and performs better compared with other methods like SIFT, SURF, ORB, BRISK, CenSurE. Based on the potential of U-FaHB detector and descriptor, we extended it for use in optical flow computation, which we termed the Nearest-flow method. This method has the potential of computing flow vectors for use in obstacle detection. Just like any other new methods, we evaluated the Nearest flow method using real and synthetic image sequences. We compare the performance of the Nearest-flow with other methods like the Lucas and Kanade, Farneback and SIFT-flow. The results obtained show that our Nearest-flow method is faster to compute and performs better on real scene images compared with the other methods. In the final part of this research, we demonstrate the application potential of our proposed methods by developing an obstacle detection and exit sign recognition system on a camera phone and the result obtained shows that the methods have the potential to solve this vision based object detection and recognition problem.
8
Zhang, Zelun. "User mobility detection using foot force sensors and mobile phone GPS." Thesis, Queen Mary, University of London, 2014. http://qmro.qmul.ac.uk/xmlui/handle/123456789/9116.
Full textAbstract:
A user (or human) mobility context is defined as a type of user context that describes a type of whole body posture (e.g., standing versus sitting) and/or a type of travel or transportation mode (e.g., walking, cycling, travel by bus, etc). Such a context can be derived from low-level sensor data and spatial contexts, including location coordinates, 3D-orientation, direction (with respect to magnetic north), velocity and acceleration. Different value-added services can be adapted to users’ mobility contexts such as assessing how eco-friendly our travel is, and adapting travel information services such as maps to different transportation modes. Current sensor-based methods for user mobility detection have several key limitations: narrow range of recognition, coarse user mobility recognition capability, and low recognition accuracy. In this thesis, a new Foot-Force and GPS (FF+GPS) sensor method is proposed to overcome these challenges that leverages a set of wearable FF sensors in combination with mobile phone GPS. The novelty of this approach is that it provides a more comprehensive recognition capability in terms of reliably recognising various fine-grained human postures and transportation modes. In addition, by comparing the new FF+GPS method with both an accelerometer (ACC) method (62% accuracy) and an ACC+GPS based method (70% accuracy) as baseline methods, it obtains a higher accuracy (90%) with less computational complexity, when tested on a dataset obtained from ten individuals. In addition, the new FF+GPS method has been further extended and evaluated. More specifically, the trade-off between the computation and resources needed to support lower versus higher number of features and sensors has been investigated. The improved FF+GPS method reduced the number of classification features from 31 to 12, reduced the number of FF sensors from 8 to 4, and reduced the use of GPS in mobility activity recognition.
9
Wong, Kim-Yung Eddie. "Automatic spoken language identification utilizing acoustic and phonetic speech information." Thesis, Queensland University of Technology, 2004. https://eprints.qut.edu.au/37259/1/Kim-Yung_Wong_Thesis.pdf.
Full textAbstract:
Automatic spoken Language Identi¯cation (LID) is the process of identifying the language spoken within an utterance. The challenge that this task presents is that no prior information is available indicating the content of the utterance or the identity of the speaker. The trend of globalization and the pervasive popularity of the Internet will amplify the need for the capabilities spoken language identi¯ca- tion systems provide. A prominent application arises in call centers dealing with speakers speaking di®erent languages. Another important application is to index or search huge speech data archives and corpora that contain multiple languages. The aim of this research is to develop techniques targeted at producing a fast and more accurate automatic spoken LID system compared to the previous National Institute of Standards and Technology (NIST) Language Recognition Evaluation. Acoustic and phonetic speech information are targeted as the most suitable fea- tures for representing the characteristics of a language. To model the acoustic speech features a Gaussian Mixture Model based approach is employed. Pho- netic speech information is extracted using existing speech recognition technol- ogy. Various techniques to improve LID accuracy are also studied. One approach examined is the employment of Vocal Tract Length Normalization to reduce the speech variation caused by di®erent speakers. A linear data fusion technique is adopted to combine the various aspects of information extracted from speech. As a result of this research, a LID system was implemented and presented for evaluation in the 2003 Language Recognition Evaluation conducted by the NIST.
10
Martin, Terrence Lance. "Towards improved speech recognition for resource poor languages." Thesis, Queensland University of Technology, 2006. https://eprints.qut.edu.au/35771/1/Terrence_Martin_Thesis.pdf.
Full textAbstract:
In recent times, the improved levels of accuracy obtained by Automatic Speech Recognition (ASR) technology has made it viable for use in a number of commercial products. Unfortunately, these types of applications are limited to only a few of the world’s languages, primarily because ASR development is reliant on the availability of large amounts of language specific resources. This motivates the need for techniques which reduce this language-specific, resource dependency. Ideally, these approaches should generalise across languages, thereby providing scope for rapid creation of ASR capabilities for resource poor languages. Cross Lingual ASR emerges as a means for addressing this need. Underpinning this approach is the observation that sound production is largely influenced by the physiological construction of the vocal tract, and accordingly, is human, and not language specific. As a result, a common inventory of sounds exists across languages; a property which is exploitable, as sounds from a resource poor, target language can be recognised using models trained on resource rich, source languages. One of the initial impediments to the commercial uptake of ASR technology was its fragility in more challenging environments, such as conversational telephone speech. Subsequent improvements in these environments has gained consumer confidence. Pragmatically, if cross lingual techniques are to considered a viable alternative when resources are limited, they need to perform under the same types of conditions. Accordingly, this thesis evaluates cross lingual techniques using two speech environments; clean read speech and conversational telephone speech. Languages used in evaluations are German, Mandarin, Japanese and Spanish. Results highlight that previously proposed approaches provide respectable results for simpler environments such as read speech, but degrade significantly when in the more taxing conversational environment. Two separate approaches for addressing this degradation are proposed. The first is based on deriving better target language lexical representation, in terms of the source language model set. The second, and ultimately more successful approach, focuses on improving the classification accuracy of context-dependent (CD) models, by catering for the adverse influence of languages specific phonotactic properties. Whilst the primary research goal in this thesis is directed towards improving cross lingual techniques, the catalyst for investigating its use was based on expressed interest from several organisations for an Indonesian ASR capability. In Indonesia alone, there are over 200 million speakers of some Malay variant, provides further impetus and commercial justification for speech related research on this language. Unfortunately, at the beginning of the candidature, limited research had been conducted on the Indonesian language in the field of speech science, and virtually no resources existed. This thesis details the investigative and development work dedicated towards obtaining an ASR system with a 10000 word recognition vocabulary for the Indonesian language.
11
Rönnqvist, Patrik. "Surveillance Applications : Image Recognition on the Internet of Things." Thesis, Mittuniversitetet, Institutionen för informationsteknologi och medier, 2013. http://urn.kb.se/resolve?urn=urn:nbn:se:miun:diva-18557.
Full textAbstract:
This is a B.Sc. thesis within the Computer Science programme at the Mid Sweden University. The purpose of this project has been to investigate the possibility of using image based surveillance in smart applications on the Internet-of-Things. The goals involved investigating relevant technologies and designing, implementing and evaluating an application that can perform image recognition. A number of image recognition techniques have been investigated and the use of color histograms has been chosen for its simplicity and low resource requirement. The main source of study material has been the Internet. The solution has been developed in the Java programming language, for use on the Android operating system and using the MediaSense platform for communication. It consists of a camera application that produces image data and a monitor application that performs image recognition and handles user interaction. To evaluate the solution a number of tests have been performed and its pros and cons have been identified. The results show that the solution can differentiate between simple colored stick figures in a controlled environment. Variables such as lighting and the background are significant. The application can reliably send images from the camera to the monitor at a rate of one image every four seconds. The possibility of using streaming video instead of images has been investigated but found to be difficult under the given circumstances. It has been concluded that while the solution cannot differentiate between actual people it has shown that image based surveillance is possible on the IoT and the goals of this project have been satisfied. The results were expected and hold little newsworthiness. Suggested future work involves improvements to the MediaSense platform and infrastructure for processing and storing data.MediaSense
12
Nguyen, Trung Ky. "Génération d'histoires à partir de données de téléphone intelligentes : une approche de script Dealing with Imbalanced data sets for Human Activity Recognition using Mobile Phone sensors." Thesis, Université Grenoble Alpes (ComUE), 2019. http://www.theses.fr/2019GREAS030.
Full textAbstract:
Le script est une structure qui décrit une séquence stéréotypée d’événements ou d’actions survenant dans notre vie quotidienne. Les histoires utilisent des scripts , avec une ou plusieurs déviations intéressantes, qui nous permettent de mieux saisir les situations quotidiennes rapportées et les faits saillants du récit. Ainsi, la notion de script est très utile dans de nombreuses applications d’intelligence ambiante telles que la surveillance de la santé et les services d’urgence. Ces dernières années, l’avancement des technologies de détection et des systèmes intégrés permettent aux systèmes de santé de collecter en permanence les activités des êtres humains, en intégrant des capteurs dans des dispositifs portables (par exemple smart-phone ou smart-watch). La reconnaissance de l’activité humaine (HAR) a ainsi connue un essor important grâce notamment à des approches d’apprentissage automatique telles que le réseau neuronal ou le réseau bayésien. Ces avancées ouvre des perspectives qui vont au delà de la simple reconnaissance d’activités. Ce manuscrit défend la thèse selon laquelle ces données de capteurs portables peuvent être utilisées pour générer des récits articulés autour de scripts en utilisant l’apprentissage automatique. Il ne s’agit pas d’une tâche triviale en raison du grand écart sémantique entre les informations brutes de capteurs et les abstractions de haut niveau présente dans les récits. A notre connaissance, il n’existe toujours pas d’approche pour générer une histoire à partir de données de capteurs en utilisant l’apprentissage automatique, même si de nombreuses approches d’apprentissage automatique (réseaux de neurones convolutifs, réseaux de neurones profonds) ont été proposées pour la reconnaissance de l’activité humaine au cours des dernières années. Afin d’atteindre notre objectif, nous proposons premièrement dans cette thèse un nouveau cadre qui traite le problème des données non uniformément distribuées (problème du biais induit par des classes majoritaires par rapport aux classes minoritaires) basé sur un apprentissage actif associé à une technique de sur-échantillonnage afin d’améliorer la macro-exactitude de classification des modèles d’apprentissage classiques comme la perception multi-couche. Deuxièmement, nous présentons un nouveau système permettant de générer automatiquement des scripts à partir de données d’activité humaine à l’aide de l’apprentissage profond. Enfin, nous proposons une approche pour l’apprentissage de scripts à partir de textes en langage naturel capable d’exploiter l’information syntaxique et sémantique sur le contexte textuel des événements. Cette approche permet l’apprentissage de l’ordonnancement d’événements à partir d’histoires décrivant des situations typiques de vie quotidienne. Les performances des méthodes proposées sont systématiquement discutées sur une base expérimentaleScript is a structure describes an appropriate sequence of events or actions in our daily life. A story, is invoked a script with one or more interesting deviations, which allows us to deeper understand about what were happened in routine behaviour of our daily life. Therefore, it is essential in many ambient intelligence applications such as healthmonitoring and emergency services. Fortunately, in recent years, with the advancement of sensing technologies and embedded systems, which make health-care system possible to collect activities of human beings continuously, by integrating sensors into wearable devices (e.g., smart-phone, smart-watch, etc.). Hence, human activity recognition (HAR) has become a hot topic interest of research over the past decades. In order to do HAR, most researches used machine learning approaches such as Neural network, Bayesian network, etc. Therefore, the ultimate goal of our thesis is to generate such kind of stories or scripts from activity data of wearable sensors using machine learning approach. However, to best of our knowledge, it is not a trivial task due to very limitation of information of wearable sensors activity data. Hence, there is still no approach to generate script/story using machine learning, even though many machine learning approaches were proposed for HAR in recent years (e.g., convolutional neural network, deep neural network, etc.) to enhance the activity recognition accuracy. In order to achieve our goal, first of all in this thesis we proposed a novel framework, which solved for the problem of imbalanced data, based on active learning combined with oversampling technique so as to enhance the recognition accuracy of conventional machine learning models i.e., Multilayer Perceptron. Secondly, we introduce a novel scheme to automatically generate scripts from wearable sensor human activity data using deep learning models, and evaluate the generated method performance. Finally, we proposed a neural event embedding approach that is able to benefit from semantic and syntactic information about the textual context of events. The approach is able to learn the stereotypical order of events from sets of narrative describing typical situations of everyday life
13
Fransson, Linda, and Therese Jeansson. "Biometric methods and mobile access control." Thesis, Blekinge Tekniska Högskola, Avdelningen för programvarusystem, 2004. http://urn.kb.se/resolve?urn=urn:nbn:se:bth-5023.
Full textAbstract:
Our purpose with this thesis was to find biometric methods that can be used in access control of mobile access. The access control has two parts. Firstly, to validate the identity of the caller and, secondly, to ensure the validated user is not changed during the session that follows. Any solution to the access control problem is not available today, which means that anyone can get access to the mobile phone and the Internet. Therefore we have researched after a solution that can solve this problem but also on how to secure that no one else can take over an already validated session. We began to search for biometric methods that are available today to find them that would be best suited together with a mobile phone. After we had read information about them we did choose three methods for further investigation. These methods were Fingerprint Recognition, Iris Scan and Speaker Verification. Iris Scan is the method that is best suited to solve the authentication problem. The reasons for this are many. One of them is the uniqueness and stability of the iris, not even identical twins or the pair of the same individual has the same iris minutiae. The iris is also very protected behind eyelids, cornea and the aqueous humor and therefore difficult to damage. When it comes to the method itself, is it one of the most secure methods available today. One of the reasons for this is that the equal error rate is better than one in a million. However, this rate can be even better. It all depends on the Hamming Distance, which is a value that show how different the saved and temporarily template are, and what it is set to. To solve our session authentication, which was to make sure that no one else could take over a connected mobile phone, a sensor plate is the answer. This sensor will be able to sense for touch, heat and pulse. These three sensor measurements will together secure a validated session since the mobile phone will disconnect if the sensor looses its sensor data. There are, however, technological and other challenges to be solved before our proposed solutions will become viable. We address some of these issues in our thesis.
14
Matějka, Pavel. "FONOTAKTICKÉ A AKUSTICKÉ ROZPOZNÁVÁNÍ JAZYKŮ." Doctoral thesis, Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií, 2009. http://www.nusl.cz/ntk/nusl-233466.
Full textAbstract:
Práce pojednává o fonotaktickém a akustickém přístupu pro automatické rozpoznávání jazyka. První část práce pojednává o fonotaktickém přístupu založeném na výskytu fonémových sekvenci v řeči. Nejdříve je prezentován popis vývoje fonémového rozpoznávače jako techniky pro přepis řeči do sekvence smysluplných symbolů. Hlavní důraz je kladen na dobré natrénování fonémového rozpoznávače a kombinaci výsledků z několika fonémových rozpoznávačů trénovaných na různých jazycích (Paralelní fonémové rozpoznávání následované jazykovými modely (PPRLM)). Práce také pojednává o nové technice anti-modely v PPRLM a studuje použití fonémových grafů místo nejlepšího přepisu. Na závěr práce jsou porovnány dva přístupy modelování výstupu fonémového rozpoznávače -- standardní n-gramové jazykové modely a binární rozhodovací stromy. Hlavní přínos v akustickém přístupu je diskriminativní modelování cílových modelů jazyků a první experimenty s kombinací diskriminativního trénování a na příznacích, kde byl odstraněn vliv kanálu. Práce dále zkoumá různé druhy technik fúzi akustického a fonotaktického přístupu. Všechny experimenty jsou provedeny na standardních datech z NIST evaluaci konané v letech 2003, 2005 a 2007, takže jsou přímo porovnatelné s výsledky ostatních skupin zabývajících se automatickým rozpoznáváním jazyka. S fúzí uvedených technik jsme posunuli state-of-the-art výsledky a dosáhli vynikajících výsledků ve dvou NIST evaluacích.
15
Kalantari, Shahram. "Improving spoken term detection using complementary information." Thesis, Queensland University of Technology, 2015. https://eprints.qut.edu.au/90074/1/Shahram_Kalantari_Thesis.pdf.
Full textAbstract:
This research has made contributions to the area of spoken term detection (STD), defined as the process of finding all occurrences of a specified search term in a large collection of speech segments. The use of visual information in the form of lip movements of the speaker in addition to audio and the use of topic of the speech segments, and the expected frequency of words in the target speech domain, are proposed. By using these complementary information, improvement in the performance of STD has been achieved which enables efficient search of key words in large collection of multimedia documents.
16
Gande, Santhrushna. "Developing Java Programs on Android Mobile Phones Using Speech Recognition." CSUSB ScholarWorks, 2015. https://scholarworks.lib.csusb.edu/etd/232.
Full textAbstract:
Nowadays Android operating system based mobile phones and tablets are widely used and had millions of users around the world. The popularity of this operating system is due to its multi-tasking, ease of access and diverse device options. “Java Programming Speech Recognition Application” is an Android application used for handicapped individuals who are not able or have difficultation to type on a keyboard. This application allows the user to write a compute program (in Java Language) by dictating the words and without using a keyboard. The user needs to speak out the commands and symbols required for his/her program. The program has been designed to pick up the Java constant keywords (such as ‘boolean’, ‘break’, ‘if’ and ‘else’), similar to the word received by the speech recognizer system in the application. The “Java Programming Speech Recognition Application” contains external plug-ins such as programming editor and a speech recognizer to record and write the program. These plug-ins come in the form of libraries and pre-coded folders which have to be attached to the main program by the developer.
17
Da, Silva Sandro Cahanda Marinho. "Remote surveillance and face tracking with mobile phones (smart eyes)." Thesis, University of the Western Cape, 2005. http://etd.uwc.ac.za/index.php?module=etd&.
Full textAbstract:
This thesis addresses analysis, evaluation and simulation of low complexity face detection algorithms and tracking that could be used on mobile phones. Network access control using face recognition increases the user-friendliness in human-computer interaction. In order to realize a real time system implemented on handheld devices with low computing power, low complexity algorithms for face detection and face tracking are implemented. Skin color detection algorithms and face matching have low implementation complexity suitable for authentication of cellular network services. Novel approaches for reducing the complexities of these algorithms and fast implementation are introduced in this thesis. This includes a fast algorithm for face detection in video sequences, using a skin color model in the HSV (Hue-Saturation-Value) color space. It is combined with a Gaussian model of the H and S statistics and adaptive thresholds. These algorithms permit segmentation and detection of multiple faces in thumbnail images. Furthermore we evaluate and compare our results with those of a method implemented in the Chromatic Color space (YCbCr). We also test our test data on face detection method using Convolutional Neural Network architecture to study the suitability of using other approaches besides skin color as the basic feature for face detection. Finally, face tracking is done in 2D color video streams using HSV as the histogram color space. The program is used to compute 3D trajectories for a remote surveillance system.
18
Berchtold-Buschle, Martin [Verfasser], and Lars [Akademischer Betreuer] Wolf. "A Modular Classifier Concept for Activity Recognition on Mobile Phones / Martin Berchtold-Buschle ; Betreuer: Lars Wolf." Braunschweig : Technische Universität Braunschweig, 2011. http://d-nb.info/1175823899/34.
Full text
19
Ghaziasgar, Mehrdad. "The use of mobile phones as service-delivery devices in sign language machine translation system." Thesis, University of the Western Cape, 2010. http://etd.uwc.ac.za/index.php?module=etd&action=viewtitle&id=gen8Srv25Nme4_7216_1299134611.
Full textAbstract:
This thesis investigates the use of mobile phones as service-delivery devices in a sign language machine translation system. Four sign language visualization methods were evaluated on mobile phones. Three of the methods were synthetic sign language visualization methods. Three factors were considered: the intelligibility of sign language, as rendered by the method
the power consumption
and the bandwidth usage associated with each method. The average intelligibility rate was 65%, with some methods achieving intelligibility rates of up to 92%. The average le size was 162 KB and, on average, the power consumption increased to 180% of the idle state, across all methods. This research forms part of the Integration of Signed and Verbal Communication: South African Sign Language Recognition and Animation (SASL) project at the University of the Western Cape and serves as an integration platform for the group's research. In order to perform this research a machine translation system that uses mobile phones as service-delivery devices was developed as well as a 3D Avatar for mobile phones. It was concluded that mobile phones are suitable service-delivery platforms for sign language machine translation systems.
20
Hain, Horst-Udo. "Phonetische Transkription für ein multilinguales Sprachsynthesesystem." Doctoral thesis, Saechsische Landesbibliothek- Staats- und Universitaetsbibliothek Dresden, 2012. http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa-81777.
Full textAbstract:
Die vorliegende Arbeit beschäftigt sich mit einem datengetriebenen Verfahren zur Graphem-Phonem-Konvertierung für ein Sprachsynthesesystem. Die Aufgabe besteht darin, die Aussprache für beliebige Wörter zu bestimmen, auch für solche Wörter, die nicht im Lexikon des Systems enthalten sind. Die Architektur an sich ist sprachenunabhängig, von der Sprache abhängig sind lediglich die Wissensquellen, die zur Laufzeit des Systems geladen werden. Die Erstellung von Wissensquellen für weitere Sprachen soll weitgehend automatisch und ohne Einsatz von Expertenwissen möglich sein. Expertenwissen kann verwendet werden, um die Ergebnisse zu verbessern, darf aber keine Voraussetzung sein. Für die Bestimmung der Transkription werden zwei neuronale Netze verwendet. Das erste Netz generiert aus der Buchstabenfolge des Wortes die zu realisierenden Laute einschließlich der Silbengrenzen, und das zweite bestimmt im Anschluß daran die Position der Wortbetonung. Diese Trennung hat den Vorteil, daß man für die Bestimmung des Wortakzentes das Wissen über die gesamte Lautfolge einbeziehen kann. Andere Verfahren, die die Transkription in einem Schritt bestimmen, haben das Problem, bereits zu Beginn des Wortes über den Akzent entscheiden zu müssen, obwohl die Aussprache des Wortes noch gar nicht feststeht. Zudem bietet die Trennung die Möglichkeit, zwei speziell auf die Anforderung zugeschnittene Netze zu trainieren. Die Besonderheit der hier verwendeten neuronalen Netze ist die Einführung einer Skalierungsschicht zwischen der eigentlichen Eingabe und der versteckten Schicht. Eingabe und Skalierungsschicht werden über eine Diagonalmatrix verbunden, wobei auf die Gewichte dieser Verbindung ein Weight Decay (Gewichtezerfall) angewendet wird. Damit erreicht man eine Bewertung der Eingabeinformation während des Trainings. Eingabeknoten mit einem großen Informationsgehalt werden verstärkt, während weniger interessante Knoten abgeschwächt werden. Das kann sogar soweit gehen, daß einzelne Knoten vollständig abgetrennt werden. Der Zweck dieser Verbindung ist, den Einfluß des Rauschens in den Trainingsdaten zu reduzieren. Durch das Ausblenden der unwichtigen Eingabewerte ist das Netz besser in der Lage, sich auf die wichtigen Daten zu konzentrieren. Das beschleunigt das Training und verbessert die erzielten Ergebnisse. In Verbindung mit einem schrittweisen Ausdünnen der Gewichte (Pruning) werden zudem störende oder unwichtige Verbindungen innerhalb der Netzwerkarchitektur gelöscht. Damit wird die Generalisierungsfähigkeit noch einmal erhöht. Die Aufbereitung der Lexika zur Generierung der Trainingsmuster für die neuronalen Netze wird ebenfalls automatisch durchgeführt. Dafür wird mit Hilfe der dynamischen Zeitanpassung (DTW) der optimale Pfad in einer Ebene gesucht, die auf der einen Koordinate durch die Buchstaben des Wortes und auf der anderen Koordinate durch die Lautfolge aufgespannt wird. Somit erhält man eine Zuordnung der Laute zu den Buchstaben. Aus diesen Zuordnungen werden die Muster für das Training der Netze generiert. Um die Transkriptionsergebnisse weiter zu verbessern, wurde ein hybrides Verfahren unter Verwendung der Lexika und der Netze entwickelt. Unbekannte Wörter werden zuerst in Bestandteile aus dem Lexikon zerlegt und die Lautfolgen dieser Teilwörter zur Gesamttranskription zusammengesetzt. Dabei werden Lücken zwischen den Teilwörtern durch die neuronalen Netze aufgefüllt. Dies ist allerdings nicht ohne weiteres möglich, da es zu Fehlern an den Schnittstellen zwischen den Teiltranskriptionen kommen kann. Dieses Problem wird mit Hilfe des Lexikons gelöst, das für die Generierung der Trainingsmuster aufbereitet wurde. Hier ist eine eindeutige Zuordnung der Laute zu den sie generierenden Buchstaben enthalten. Somit können die Laute an den Schnittstellen neu bewertet und Transkriptionsfehler vermieden werden. Die Verlagsausgabe dieser Dissertation erschien 2005 im w.e.b.-Universitätsverlag Dresden (ISBN 3-937672-76-1)The topic of this thesis is a system which is able to perform a grapheme-to-phoneme conversion for several languages without changes in its architecture. This is achieved by separation of the language dependent knowledge bases from the run-time system. Main focus is an automated adaptation to new languages by generation of new knowledge bases without manual effort with a minimal requirement for additional information. The only source is a lexicon containing all the words together with their appropriate phonetic transcription. Additional knowledge can be used to improve or accelerate the adaptation process, but it must not be a prerequisite. Another requirement is a fully automatic process without manual interference or post-editing. This allows for the adaptation to a new language without even having a command of that language. The only precondition is the pronunciation dictionary which should be enough for the data-driven approach to learn a new language. The automatic adaptation process is divided into two parts. In the first step the lexicon is pre-processed to determine which grapheme sequence belongs to which phoneme. This is the basis for the generation of the training patterns for the data-driven learning algorithm. In the second part mapping rules are derived automatically which are finally used to create the phonetic transcription of any word, even if it not contained in the dictionary. Task is to have a generalisation process that can handle all words in a text that has to be read out by a text-to-speech system
21
Hirsch, Gérard. "Équations de relation floue et mesures d'incertain en reconnaissance de formes." Nancy 1, 1987. http://www.theses.fr/1987NAN10030.
Full textAbstract:
Il est appelé que le sylogisme indirect n'est pas parfait quelque soit l'opérateur de composition floue. Un opérateur de maximalisation (ou de minimalisation) est déterminé pour la composition sup-T norme (ou INF-T conorme). Après la reprise des résultats des mesures d'incertain il est donné une application numérique au problème de classification des phonèmes
22
Berri, Rafael Alceste. "Sistema de visão computacional para detecção do uso de telefones celulares ao dirigir." Universidade do Estado de Santa Catarina, 2014. http://tede.udesc.br/handle/handle/2038.
Full textAbstract:
Made available in DSpace on 2016-12-12T20:22:52Z (GMT). No. of bitstreams: 1
RAFAEL ALCESTE BERRI.pdf: 28428368 bytes, checksum: 667b9facc9809bfd5e0847e15279b0e6 (MD5)
Previous issue date: 2014-02-21Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
In this work, three proposals of systems have been developed using a frontal camera to monitor the driver and enabling to identificate if a cell phone is being used while driving the vehicle. It is estimated that 80% of crashes and 65% of near collisions involved drivers who were inattentive in traffic for three seconds before the event. Five videos in real environment were generated to test the systems. The pattern recognition system (RP) uses adaptive skin segmentation, feature extraction, and machine learning to detect cell phone usage on each frame. The cell phone detection happens when, in periods of 3 seconds, 60% (threshold) of frames or more are identified as a cell phone use, individually. The average accuracy on videos achieved was 87.25% with Multilayer Perceptron (MLP), Gaussian activation function, and two neurons of the intermediate layer. The movement detection system (DM) uses optical flow, filtering the most relevant movements of the scene, and three successive frames for detecting the movements to take the phone to the ear and take it off. The DM proposal was not demonstrated as being an effective solution for detecting cell phone use, reaching an accuracy of 52.86%. The third solution is a hybrid system. It uses the RP system for classification and the DM for choosing the RP parameters. The parameters chosen for RP are the threshold and the classification system. The definition of these two parameters occurs at the end of each period, based on movement detected by the DM. Experimentally it was established that, when the movement induces to use cell phone, it is proper to use the threshold of 60%, and the classifier as MLP/Gaussian with seven neurons of the intermediate layer; otherwise, it is used threshold 85%, and MLP/Gaussian with two neurons of the intermediate layer for classification. The hybrid solution is the most robust system with average accuracy of 91.68% in real environment.
Neste trabalho, são desenvolvidas três propostas de sistemas que permitem identificar o uso de celular, durante o ato de dirigir um veículo, utilizando imagens capturadas de uma câmera posicionada em frente ao motorista. Estima-se que 80% das colisões e 65% das quase colisões envolveram motoristas que não estavam prestando a devida atenção ao trânsito por três segundos antes do evento. Cinco vídeos em ambiente real foram gerados com o intuito de testar os sistemas. A proposta de reconhecimento de padrões (RP) emprega segmentação de pele adaptativa, extração de características e aprendizado de máquina (classificador) na detecção do celular em cada quadro processado. A detecção do uso do celular ocorre quando, em períodos de 3 segundos, ao menos em 60% dos quadros (corte) são identificados com celular. A acurácia média nos vídeos alcançou 87, 25% ao utilizar Perceptron Multi-camadas (MLP) com função de ativação gaussiana e dois neurônios na camada intermediária como classificador. A proposta de detecção de movimento (DM) utiliza o fluxo ótico, filtragem dos movimentos mais relevantes da cena e três quadros consecutivos para detectar os momentos de levar o celular ao ouvido e o retirá-lo. A aplicação do DM, como solução para detectar o uso do celular, não se demostrou eficaz atingindo uma acurácia de 52, 86%. A terceira proposta, uma solução híbrida, utiliza o sistema RP como classificador e o de DM como seu parametrizador. Os parâmetros escolhidos para o sistema de RP são o corte e o sistema classificador. A definição desses dois parâmetros ocorre ao final de cada período, baseada na movimentação detectada pela DM. Com experimentações definiu-se que, caso a movimentação induza ao uso do celular, é adequado o uso do corte de 60% e o classificador MLP/Gaussiana com sete neurônios na camada intermediária, caso contrário, utiliza-se o corte de 85% e classificador MLP/Gaussiana com dois neurônios na mesma camada. A versão híbrida é a solução desenvolvida mais robusta, atingindo a melhor acurácia média de 91, 68% em ambiente real.
23
Meng, Chao-Hong, and 孟昭宏. "Phone Recognition using Structural Support Vector Machine." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/45917295138142120792.
Full text
24
Chen, Ko-Chih, and 陳克智. "License Plate Detection and Recognition of Smart-Phone." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/01428773053617602065.
Full textAbstract:
碩士國立中央大學
資訊工程學系碩士在職專班
99
This paper presents an approach for license plate recognition using a camera-equipped smartphone. The proposed method provides a reliable and accurate technique to solve the problem of license plate recognition caused by the skew and shadow on the license plates. There are four stages in the proposed approach: license plate location, license plate rectification, character segmentation and character recognition. In the first stage, we locate the license plate by accumulating edge points, and then analyze the edge points and accumulation associated with vertical and horizontal dimensions of the image. As to the second stage, license plate rectification, we adopt local threshold to cope with the problem of shadow on the plates first. Next step involved analyzing black and white pixels in order to decide whether to invert the image or not. The researcher tries to engage the characteristics like length-width ratio, size, and position of the bounding box in the text region to eliminate the non-text portions. To solve the rotation, skew, and scale problems of the slanted license plates in the image, we use an affine transformation to estimate the skew angle. Edge points vertical direction accumulating and trough are used to segment characters section in the third stage. We normalize the characters size to 40 × 90. Finally, criterion of normalized cross-correlation is used in the last stage for character recognition. In behalf of shortening the process time for identification, the procedure of character reorganization is improved. We shrink the samples to one-fourth the size to conduct the first identification process. Then, three highest-coefficient samples are chosen to match the original input pattern. From these three samples, the highest-coefficient one is selected as the final result.
25
Hu, Yu-Ling, and 胡玉玲. "Menu Recognition Meal Ordering System Using Smart Phone." Thesis, 2012. http://ndltd.ncl.edu.tw/handle/43918929200078065304.
Full textAbstract:
碩士龍華科技大學
電子工程系碩士班
100
Meal ordering system has been widely used in restaurants. The system we present is based on the android smart phone, which takes the picture of the meal order form, recognizes the selected menu items, and sends the ordering information to the restaurant counter and kitchen via Wi-Fi for food and check processing. The system is composed of four parts. At first, image of the meal order form is acquired by the smart phone for processing. The image processing software is developed with OpenCV which includes eliminating the influence of the illumination, binarizing the image, correcting the skew image, and analyzing the information of the customer menu items. The second part is the development of an android App which provides a user friendly interface for meal ordering. The third part is the integration of android App and image processing software. The last part of the system is the implementation of a windows application program for meal ordering information management.
26
TSENG, CHIUNG-HSIANG, and 曾炯祥. "WIFI Mesh Face Recognition System With Phone Notification." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/fqk4j4.
Full textAbstract:
碩士崑山科技大學
電機工程研究所
106
This paper proposes that WIFI Mesh wireless face recognition system must be built WIFI Mesh network can be divided into Gateway and Internal Node two roles can be used to replace the wired video system needs to pull the shortcomings. Because this system is connected WIFI inside the Mesh network, Gateway is the only node that communicates with the outside world. It needs a set of fixed IPs that can be connected to the Internet. The Gateway itself also has a set of Camera and other Internal Node There are also a group of Camera. Users can get Gateway and Internal Nodes dynamic video data by connecting to the Gateway through the Internet. Between Gateway and Internal Nodes, WIFI is used to transmit video without any additional video cables. Internal Nodes only need to provide power, You can join the Mesh network, and the image is transmitted to the Gateway side by way of a jump point so that the user can view the image of the Internal Nodes when connected to the Gateway. Wi-Fi Mesh network face recognition with phone notification Android phone or Iphone, when the WIFI Mesh network is completed as long as the connection into the WIFI Mesh Getway, each Wifi Mesh node has a camera to intercept operational knowledge, and will know the face The result is transmitted to the Server host via the Wifi Mesh network, and then sent by the Server to recognize the face information push broadcast message to the mobile terminal. In connection with the WIFI Mesh network, the smart phone APP is used for information delivery, and the Internet , Wisdom, ease of security and remote control purposes.
27
Hsu, Shuo-Pin, and 許碩斌. "An Initial Study on Minimum Phone Error Discriminative Training for Continuous Phone Recognition System." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/05601427143243772117.
Full textAbstract:
碩士國立清華大學
資訊工程學系
94
Maximum Likelihood Estimation (MLE) is a traditional method for training acoustic models for speech recognition. This method does not consider discriminative relation between acoustic models, so some models are apt to obscure each other. In order to raise the differentiation degree between models, discriminative training criteria are proposed. Seeing that Minimum Phone Error (MPE) criterion has great progress reported in the literature, we apply MPE to continuous phone speech recognition system in this thesis. The procedure is to adopt MLE to train acoustic models first, and then use MPE to refine the models again. According to the experimental result, MPE can reduce phone error rate further. In general, MPE adopts phone lattice to express all possible sentences. In order to improve the efficiency, we use N-Best list to construct a phone lattice which is called N-Best Synthesized Lattice. Besides, in order to distinguish obscure phones and remove repeated words that appear in very close time, we use another kind of phone lattice called sausage that can improve the results of MPE.
28
PAN, YU-WEI, and 潘郁薇. "Using Smart Phone Sensor Data for Human Activity Recognition." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/4s6yjm.
Full textAbstract:
碩士國立中正大學
資訊工程研究所
104
Abstract In order to run a successful application, the most important is to understand users’ intentions and desires, then, combine with user habits. By capturing these information, they can provide better service and enhance marketing strategy to achieve this goal. Human activity recognition is an application that can help people to explore the useful data of users’ information. Data Mining is a process to find model from a amount of data, use data to build the model which is use to simulate the real world, then, use these models to describe the patterns and relations from the data. It can provide useful information when making decision or help to making predictions. Nowadays, applications for mobile devices become more widely. Smart phones supply a lot of user’s sensor data such as location sensor data, temperature sensor data, humidity sensor data and acceleration sensor data. Those sensor data build different point of view for data mining applications. In this study, the goal is expect that every day activities are recognized from data collected using smartphones accelerometer sensors and location information. We collected sensor data from users as they performed daily activities such as walking, running, riding and relaxing(static). We use data mining tools for data preprocessing and classification by analyzing and storing data so that we can recognize human activity. Keywords：Data Mining、Sensor data、Activity recognition
29
Gao, Shih-Ciao, and 高士喬. "Binding Book Music Recognition Based on Mobile Phone Image." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/39231764503849115178.
Full textAbstract:
碩士國立中央大學
電機工程研究所
98
This thesis presents a system for music recognition on non-flat surface music score, whether commercial or self-made music score by image processing. In this system, the image is captured by a mobile phone and sent to PC through Bluetooth protocol. And then the image distortion correction and music recognition are applied. Two modes are built in this system, namely random and assignment recognition mode. All and one part of music notes are recognized in random and assignment recognition mode, respectively. The recognition results are matched with the database for the song correction. Finally, the recognized music is converted to a MIDI file and played on the PC or mobile phone. The recognition system comprises two part processes, namely geometric distortion correction and music recognition. In geometric distortion correction, the corner and boundary detection and vertical and horizontal correction are applied to correct the image warping which is due to bookbinding. The meter and scale of music are recognized in music recognition part. The steps of music recognition are music staves detection, stuff lines detection and the music notes recognition. Finally, the music theory is applied to check the recognized results again. The experiment shows the recognition rate is about 95% in more than 30 songs, and the recognition time for each song is about 2.5 seconds.
30
Chu, Kuang-Chen, and 瞿光宸. "Banknote Recognition System for Smart Phone Based on SIFT Algorithm." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/8txxk8.
Full textAbstract:
碩士淡江大學
機械與機電工程學系碩士班
103
The smart phone banknote recognition system is developed in this study to provide a recognizing banknote solution for the visual impaired by using smart phone. The recognition system is built on the Android platform which is the most popular for smart phones. In this system a friendly operating interface for the visual impaired is designed. By pressing a start button on touch screen, the system begins taking photo, processing banknote recognition and gives the final result. In this study, the Scale Invariant Feature Transform is used to create feature points from the banknote image taken by smart phone. Then 128 dimensional vectors of feature points are used to compare with pre-built database of New Taiwan Dollars banknote which is also created by Scale Invariant Feature Transform. The recognition result with the value of the banknote is responded in voice to the visual impaired. It allows the visually impaired to easily know banknote denominations to solve the difficulty on banknote recognitions and improve convenience in their daily life.
31
Su, Ching-Yuan, and 蘇靖淵. "License Plate Recognition System with Gyroscope Sensor in Smart Phone." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/86200735522565340904.
Full textAbstract:
碩士國立勤益科技大學
資訊工程系
101
In this thesis, we propose a method and implement a license plate recognition system, LPRS, on a smart-phone. The recognition algorithm of LPRS is based on a similar random forest method. We plan several tasks and finish them by computer in order to make a thin, fast, and accurate system. There are static and continuous images when the LPRS is detecting by camera of smart-phone. However, the images from handheld device are influenced because the user has tremor. That would be effect directly for choosing suitable algorithm, but the proposed method is solved this problem by using return tri-axial information from gyroscope sensor of smart phone. Therefore, the images from camera capture are provided with rotation-invariant. The stages of LPRS are included image preprocessing, license plate detection and capture (LPDC), license plate area elements segmentation (LPAES), license plate area elements features computing (LPAEFC), and license plate elements recognition (LPER). The main goal of image preprocessing stage is reducing information of image to make them simply. Let the system become more efficient, and on the other, it will get location of license plate more accurate. We set up a region of interest (ROI) and an effective range of angle of detection. According to the setting, user can use LPRS intuitively and get result quickly. After setting and image preprocessing, it is scanned ROI and found the location of license plate through setting thresholds, upper and lower limit of color changing. However, the result depends. There is a rectangle for circling location of license plate. Sometime the rectangle size is too strange. The reason is that the LPRS is got a bad image of binarization, we can use it to check that is a wrong result or not. We call this processing conditional binarization. Then, the location of license plate will be re-found and got right result. If this processing is working, the LPAES and LPER stage will be improved. At the LPAES stage, we use the contrast between detecting element and background and cut them at their edges. The final stage, recognition, our goal is not only classify characters, but also classify text and non-text (noisy). According to experimental results, the propose method and system are provided with a very good recognition ability.
32
Yang, Hao-Chung, and 楊皓中. "Minimum Phone Error Training for Code-Mixed Bilingual Speech Recognition." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/94650767618554925939.
Full text
33
CHEN, CHIA-LIANG, and 陳佳良. "Artificial Intelligence and Mobile Phone Sensing based User Activity Recognition." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/5jj9t7.
Full textAbstract:
碩士東吳大學
巨量資料管理學院碩士學位學程
107
According to the 2105 Annual Report published by Health Promotion Administration of Ministry of Health and Welfare in 2017, the lack of improper diet and exercise can easily lead to obesity and further the development of chronic diseases. Eight of the country’s top ten causes of death are closely related to obesity. In 2015, there were nearly 50% of men and about 30% of women were overweight in over age 18 group. By exercising, physical fitness can be promoted to prevent the occurrence of chronic diseases, thereby reducing dependence on medical resources and avoiding waste of medical resources. In addition to the health problems to be overcome, the other serious problem is the problem of the aged society. Internationally, the proportion of people over the age of 65 to the total population of 7% is called an aging society, 14% is called a aged society, and when it reaches 20%, it is called a super-aged society. According to the National Development Council, our country has become an aging society in 1993. It is expected to enter the aged society in 2018 and enter the super-aged society in 2026. The health care of the aged people has become a problem that must be taken seriously. With the development of Micro Electro Mechanical Systems (MEMS), many wearable devices, such as smart wristbands, smart watches, and smart phones, have built-in sensors that detect the state of the body. For instance, identifying the action type and record the duration of exercise. In home care for aged people, wearable devices equip with motion sensors can proceed sleep detection, fall detection, and gait analysis can effectively get the critical information of the aged people at home. In addition to providing assistance in the situation of a critical event happened, it can also be discovered through long-term gait analysis whether the physical condition is abnormal. The three-axis accelerometer (hereinafter referred to as the accelerometer) and the three-axis gyroscope (hereinafter referred to as the gyroscope) are almost standard equipment for smart watches, smart wristbands, and smart phones. The main advantage of the accelerometers is that it can measure the static or slow moving sensor's pendulum acceleration value, but the disadvantage is that it can't measure the rotation angle along the gravity axis, and the gyroscope makes up for this shortcoming, and is not affected by gravity. The angular velocity can be measured directly, and this advantage can capture the subtle rotation of the human body. Combined with the accelerometer and the gyroscope for activity recognition, a relatively high recognition rate can be achieved. This study used two data sets collected from accelerometers and gyroscopes. One has two kind of activities collected from a single personal iphone 5c smartphone from the Kaggle website and the other has six kind of activities collected from smartphone (Samsung Galaxy S II) from thirty people from UCI machine learning repository website. The experiment adopts the supervised learning algorithm in machine learning, which is Logistic Regression, Decision Tree, Random forest and Support Vector Machine (SVM) respectively. The experimental results of kaggle data set show that decision tree algorithm is the best model under specific acceptance of accuracy and minimum model training time. But if the accuracy is the only goal of pursue, no matter the model training time is, the support vector machine algorithm will be the best model to be adopted. In addition, the experimental results of UCI data set, the decision tree algorithm achieves the highest accuracy.
34
Yang, Hsiang-Lin, and 楊翔麟. "Examining Product Identity of Mobile Phone by Form Feature Recognition." Thesis, 2007. http://ndltd.ncl.edu.tw/handle/37sjzm.
Full textAbstract:
碩士國立臺北科技大學
創新設計研究所
95
This research, taking Sony Ericsson mobile phones as examples, aims at examining if the product identification developed by the maker is consistent with consumers’ perception of the product styles and form features. The research begins with the collection and review of related documents to build an “evolution timeline” of the Sony Ericsson mobile phones for tracing the development of SE mobile phones in terms of features. Several SE mobile phones are then selected as representative models, and interviews with local sellers of Sony Ericsson flagship store and communication products are conducted to narrow the representative models down to four mobile phones for use as survey samples. The survey shows the samples to 50 general consumers, testing their perceptions of the four SE phones in order to identify the features or design elements that have successfully grasped the viewers’ attention. On the other hand, the four SE samples are also used in interviews with mobile phone designers so as to rank the design elements or features by their contribution to product identification. The differences between the results of the survey on general consumers and those of the survey on mobile phone designers are them compared and analyzed to test the consistency between Sony Ericsson production identification and Sony Ericsson product features. Results of the study suggest: （1） Both the consumer’s and the designers’ attentions focus on the most significant feature is “function keypad” and follows by “earphone” and “number keys.” Hence “function keypad” is accordingly identified as the main identification element of Sony Ericsson mobile phone.（2） Three factors that significantly influence consumers’ perception of product features are Profession, educational background in design, and experience of owning Sony Ericsson mobile phones. Profession appears to produce significant influences on perception of earphone, educational background in design on number keys, and experience of owning Sony Ericsson mobile phones on function keypad.
35
Chen, wen-Chih, and 陳文志. "Development of a Camera-Phone-Based Drug Barcode Recognition Support System." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/45248645836262935822.
Full textAbstract:
碩士國立陽明大學
衛生資訊與決策研究所
94
FDA was issuing a new rule to require certain human drug and biological product labels to have line bar codes. This will help reduce the number of medication errors in hospitals and other health care settings. Traditionally, the bar code readers used laser to scan bar code. This needed extra cost for the equipment to read the bar code. On the other way, using smartphone integrated camera to recognize visual tags is currently hot topic. Smartphone is a ubiquitous computing tool. With the phone and the camera, it give us a good opportunity to bring up an support system for care givers and the patients read and learn drug information just on their phone. This paper descripts a software support system for care givers and patients to identify the barcode use a camera phone. It is an economical and effective way if we can create software to help people to identify the barcode and show the information they need just on their phone. The result shows that the correct percentage of each bar code number is 86.6%. The correct percentage of full bar code is 95%. We had considered the real situation in the hospital, sometimes the barcode maybe draw by the pen. We draw on the barcode and try to recognition it. The result shows that the correct percentage of this kind of bar code is 92%. As a result, we thanks our algorithm is very feasibility.
36
Tsai, Kang-Chun, and 蔡康俊. "Posture Recognition with Mobile Phone G-Sensor and Artificial Neural Networks." Thesis, 2012. http://ndltd.ncl.edu.tw/handle/48837312789901100055.
Full textAbstract:
碩士淡江大學
資訊工程學系碩士班
100
Mobile phone is a popular device in the world, every people almost have a mobile phone in their hand. The smart phone applications combine many hardware devices, like GPS and G-Sensor. The smart phone can do more than your computer, moreover, the smart phone has mobility, hardware device and mobile network which is inexistent in any computer. Many researches are based on mobile phone applications in recent years. In this research, we propose a method that can recognize four human posture states, i.e., sit, stand, walk and run. When the user clicks the start button, the system will catch accelerometer data into the recognition module. In this module, data will be processed with moving average and enter artificial neural networks. When the state change, the system record posture state and duration into the database. In history module, the user can find each state duration and total calorie from user interface for a certain. We hope this system can be ease of use and really helps the user to achieve moderate exercise.
37
Ting-ChiaLee and 李定家. "Design of Liveness Detection and Identity Recognition System for Mobile Phone." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/n84v5s.
Full text
38
Ming-Li, Shen, and 沈明莉. "Corporate Homepage Design and Customer Recognition — A Comparison among Mobile Phone Companies." Thesis, 2004. http://ndltd.ncl.edu.tw/handle/51864143692517042816.
Full textAbstract:
碩士樹德科技大學
金融保險研究所
92
Internet connection provides a low cost, convenient, and instantaneous response to remote access along with multimedia access and rich contents. Via Internet connection, mobile phone companies can promote products, provide support to their customers, collect marketing information, and collect bills. This article compares Internet homepage design, flow of information, consistency, corporate image and easiness of access among of six mobile phone companies (Chunghwa Telecom Co., TransAsia Telecommunications, Mobitai Communications, FarEast Tone, KG Telecommunications, and Taiwan Cellular Corp) in Taiwan.
39
Yan-TingYang and 楊晏婷. "Phone Set Construction based on Articulatory Features for Code-Switching Speech Recognition." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/70297388027061823351.
Full text
40
Liu, Huang-Yu, and 劉黃裕. "iPill: Highly Efficient and Fully Automatic Pill Recognition System on Android Smart Phone." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/gnq3mx.
Full textAbstract:
碩士國立臺灣科技大學
電機工程系
103
Senior people may not be able to recognize the type of pill, and thus eating by mistake. Another possibility is the wrong frequency of pill taking in a day. The above are undoubtable critical issues in healthcare. Consequently, we develop an Android application to avoid the above scenarios. This thesis presents an efficient pill recognition method, which is useful in the application of healthcare. Specifically, an effective color segmentation called Pixel Matching Segmentation (PMS) with local adaptive thresholding (LAT) is utilized to segment the pill region, proposed a high accuracy shadow removal method with building codebook model in single image, where the geometric and rotation invariant imprint features are extracted with the aid of reference background. Since the imprint on the pill may not be clear in the image because of luminace variant, parametric oriented histogram equalization (POHE) is deployed to efficiently enhance the image, and the concentric circle masks is proposed to extract the imprint features. Subsequently, the libsvm is applied to train the model for pill recognition and classification. The proposed method is implemented on an Android mobile phone for testing and evaluation purposes. The proposed method yields a recognition accuracy of 99.07% of common 100 types of pill in Taiwan. Experimental results suggest that the proposed method can be an effective and convenient way for the application of pill recognition.
41
Wang, Jane-Wen, and 王禎文. "The Study of Consumer Brand Recognition and Segmentation in Smart Cell Phone Market." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/38728789980672865747.
Full textAbstract:
碩士國立中興大學
行銷學系所
98
In the past, most researches in smart phones were focus on application of specific industries or based on the functions of smart phones to segment the market. In this study, the main purpose was to analysis consumer brand awareness of smart phones and to study the differentiation of market segmentation, through the Technology Acceptance Model and consumer behavior in perceived value and risk. This research surveyed smart phone users through internet. There were 1979 usable questionnaires collected. After data collection, analysis conducted by descriptive statistics, factor analysis, cluster analysis, and multivariate analysis to verify the hypothesis of this study. The results showed that brand awareness should start from product ease of use and the effective use of smart phone. Also, smart phone market can divided into "risk-sensitive product-oriented", "product easy to use sensitive guidance", "and value-sensitive product-oriented" three groups to develop different marketing communications strategy. Finally, the empirical results of this study provided marketing suggestions for the smart phone marketers in Taiwan.
42
Fan, Yang-Wu, and 范揚武. "Application of Pattern Recognition and Position Control to Smart Phone Automatic Test System." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/63860285557227775462.
Full textAbstract:
碩士國立臺灣海洋大學
通訊與導航工程學系
101
With the progress of human civilization, as well as lifestyle changes, intelligent robots are gaining influence in human daily life. Robots can provide many services and helps to human, such as providing entertainment, life safety, health and other aspects of the service. Different types of robots have been developed in recent years. For a variety of needs, the development of robot system combines theoretical basis of many professional knowledge, such as path planning, visual image processing technology, body positioning, obstacle avoidance techniques, and arm control. Scientists try to use different algorithms on different applications. Intelligent robots consist of mechanics, electronics, automation, control, and communications technologies. Many researchers have tried to convert expertise into the robot systems so that the human-robot interaction in daily life can be more harmonious, and robot will have ability to finish various tasks. This paper presents an application of a 4 degrees-of-freedom articulated robot to a smart phone automatic test system. Intelligent scheme based on fuzzy logic theory, pattern recognition and HSL image process are proposed to control a robot arm for position control. The Denavit–Hartenberg model (D-H) is first used to analyze the robot arm movement. Then the forward kinematics is used to identify the relationships for each joint of the robot arm. The robot uses hierarchical fuzzy control to solve the problem regarding inverse kinematics for the industrial articulated robot. The human-machine interface is handled by the Labview 2010, and then uses the Matlab codes to the controller. With the webcam, coordinate is provided to the fuzzy controller, the robot arm can be moved to the desired position. In image processing, we found that the interference of light intensity is very troublesome. Therefore, we transform the RGB color space to the HSL color space which can significantly reduce the impact of light. Then, use match pattern to recognize the smart phone buttons. Vision Builder for Automated Inspection allows one to easily configure and benchmark a sequence of visual inspection steps as well as deploy the visual inspection system for automated inspection. The experimental results show that the proposed control scheme can drive the robot arm to press the desired buttons of the tested smart phone successfully.
43
Huang, Xiao Hong, and 黃曉鴻. "An initial study on continuous mandarin speech recognition based on context-dependent phone models." Thesis, 1994. http://ndltd.ncl.edu.tw/handle/38739351602092129570.
Full text
44
Tsai, Yi-ju, and 蔡奕如. "Application of Real-time Character Recognition and Fuzzy Control for Smart Phone Automatic Operation." Thesis, 2014. http://ndltd.ncl.edu.tw/handle/65923488537742790473.
Full textAbstract:
碩士國立臺灣海洋大學
通訊與導航工程學系
102
Advance robotic technologies bring convenience to human life not only in the industrial field but also the education and entertainment fields. Researchers have developed numerous types of robots to meet different demands so that the human-robot interaction can be more mature, robot arm can carry more powerful functions. This study presents an application of automatic recognizing words from PC screen and smart phone by a robot arm. After these words which represent commands are received by the robot, the robot will do the corresponding action to test smart phone. We use two cameras for experiment in this thesis. One of the webcams is utilized to capture the commands on the screen; the other one is utilized to recognize the words on the screen of the tested smart phone. The method of image process is based on RGB-Green Plane and Hue-Saturation-Luminance (HSL) color space to reduce the influence of light. Fuzzy theory is used in position control. Optical Character Recognition (OCR) technique is used for the character recognition, and the recognition results are then checked by the dictionary process to increase the recognition accuracy. The human-machine interface is handled by the Labview 2010, and then uses the Matlab codes to the controller. With the camera, coordinates are provided to the fuzzy controller, the robot arm can be moved to the desired position. Experimental results show that recognition accuracy is 92.4% for images on the computer screen and 99% by a dictionary process. The proposed control scheme can make the robot arm to perform different assigned functions successfully.
45
Lu, Tung-Hsuan, and 呂東烜. "Inter-word Tri-phone Model Search and Analysis in Large Vocabulary Continuous Speech Recognition." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/20495740563553788532.
Full text
46
Wu, Hsiao-Chien, and 吳孝謙. "Design of a Fall Detection System by Smart Phone with Recognition of Falling Direction." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/69858002825455650109.
Full textAbstract:
碩士輔仁大學
電機工程學系碩士班
101
In this paper we use the accelerometer of a smart phone to design and implement a fall monitor. We not only analyze the change of acceleration but also analyze three typical human actions. Then we compare the actions of going upstairs, going downstairs, standing up, sitting down, running and jumping, with the characteristics of a fall. These are weightlessness, impact and overturning of the body. Because the waist is the center of gravity in the human body, our system is used more effectively when we place the smart phone at the waist. Our system is based on an open source system platform and the accelerometer in the smart phone. Because of being is based on the smart phone, our system can be used outdoors.
47
Chen, Jia-Yu, and 陳佳妤. "Minimum Phone Error Training of Acoustic Models and Features for Large Vocabulary Mandarin Speech Recognition." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/63448829355525193378.
Full textAbstract:
碩士國立臺灣大學
電機工程學研究所
94
Traditional speech recognition uses maximum likelihood estimation to train parameters of HMM. Such method can make correct transcript have largest posterior probability; however it can’t separate confused models effectively. Discriminative training can take correct transcript and recognized result into consideration at the same time, trying to separate confused models in high dimensional space. Based on minimum phone error (MPE) and feature-space minimum phone error (fMPE), the thesis will introduce discriminative training’s background knowledge, basic theory and experimental results. The thesis has four parts: The first part is the basic theory, including risk estimation and auxiliary function. Risk estimation starts from minimum Bayesian risk, introducing widely explored model training methods, including maximum likelihood estimation, maximum mutual information estimation, overall risk criterion estimation, and minimum phone error. The objective functions can be regarded as extension of Bayesian risk. In addition, the thesis will review strong-sense and weak-sense auxiliary functions and smoothing function. Strong-sense and weak-sense auxiliary functions can be used to find the optimal solution. When using weak-sense auxiliary function to find solutions, adding smoothing function can improve convergence speed. The second part is the experimental architecture, including NTNU broadcast news corpus, lexicon and language model. The recognizer uses left-to-right, frame-synchronous tree copy search to implement LVCSR. The thesis uses maximum likelihood training results of mel frequency cepstrum coefficients and features processed by heteroscedastic linear discriminant analysis as baseline. The third part is minimum phone error. The method uses minimum phone error as direct objective function. From the update equation we can see that the newly trained model parameters are closer to correctly-recognized features (belong to numerator lattices) and move far away from wrongly-recognized features (belong to denominator lattices). The I-smoothing technique introduces model’s prior to optimize estimation. Besides, the thesis will introduce the approximation of phone error-how to use lattice to approximate all recognized results and how to use forward-backward algorithms to calculate average accuracy. The experimental results show that this method can reduce 3% character error rate in the corpus. The fourth part is the feature-space minimum phone error. The method projects features into high-dimension space and generate an offset vector added to original feature and leads to discrimination. The transform matrix is trained by minimum phone error followed by gradient descent to do update. There are direct differential and indirect differential. Indirect differential can reflect the model change on features so that feature training and model training can be done iteratively. Offset feature-space minimum phone error is different in the high dimension feature. The method can save 1/4 computation and achieve similar improvement. My thesis proposed dimension-weighted offset feature-space minimum phone error which treats different dimensions with different weights. Experimental results show that theses methods have 3% character error rate reduction. Dimension-weighted offset feature-space minimum phone error has larger improvements and more robust in training.
48
Xu, Long Lun, and 許隆倫. "Continuous mandarin speech recognition based on right context dependent phone models with respect to speakers." Thesis, 1995. http://ndltd.ncl.edu.tw/handle/05275259162797381485.
Full text
49
SHIH, HUA-YUAN, and 施驊原. "Mobile Phone Identity Recognition System Using Touch Screen Based on Fuzzy Logic and Decision Tree." Thesis, 2017. http://ndltd.ncl.edu.tw/handle/973veb.
Full text
50
Hsu, Jui-Te, and 徐瑞德. "The Influence on Interface Recognition with Different Buttuon Form ─ Take the Mobile Phone as a Case Study." Thesis, 2007. http://ndltd.ncl.edu.tw/handle/btpv79.
Full textAbstract:
碩士大同大學
工業設計學系(所)
95
The size of today’s electronic products tends to be small, and those products have multiple functions as well. Under the condition of limited space in the panel with increasing functions, many buttons with different functions are forced to be set in a control panel with small space. It causes the crowded space of the panel. In this situation, how to recognize which buttons belong to control button with the same function for the user become the focus to the designer in the design of panel. Therefore, button panel in the mobile phone will be used as the sample in this study to discuss how to increase the effect of recognition when the different functions are set in the same operating panel, and to reach a good effect of transmission. The result of this study shows that (1) there are more than two functions in a panel when they are set in a control panel, the bringing of factors with different designs can increase their discrepancy and reach the effect of fast recognition. However, when the designer needs to use the different design factors in a control panel with the same function to emphasize on the discrepancy between this panel and that panel, he needs to collocate it with other design factors in a panel with the same functions by using the overlap. By doing so, he can reach a good effect of grouping. (2) If the button design is designed by the same design factors, then he should make the difference in the arrangement of the button to express such difference in the different panels. (3) In the panel with two kinds of functions, the design can be differentiated by using law of proximity and law of similarity in the grouping. Besides, the arrangement or the size of relative functions can be set similarly to give the effect of grouping. (4) The awareness about operation from the user should be taken into account in the panel design.