Dissertations / Theses on the topic 'Blink recognition'

The thesis deals with eye blink detection, which is part of complex topic of face detection and recognition. The work intents on digital image processing. There is analyse of the topic and description of image databases for testing. Two main chapters describe design of eye blink detection with digital image processing with IR technology and without IR technology.

This master thesis deals with the issues of the eye blink recognition from video. The main task is to analyse algorithms dealing with a detection of persons and make a program that could recognize the eye blink. Analysis of these algorithms and their problems are in the first part of this thesis. In the second part design and properties of my program are described. The realization of the program is based on the method of move detection using the accumulated difference frame, which helps to identify the eye areas. The eye blink detection algorithm tests a match between a tresholded pattern of the eye area taken from the actual frame and the frame before. The resolution whether the eye blink happened or not, is based on the level of the match. The algorithm is designed for watching a sitting man, which is slightly moving. The background can be a little dynamic as well. An average quality video with a moderator and dynamic backround was used as a tested subject.

In this thesis work, automatic recognition algorithms for digital modulated signals are surveyed. Feature extraction and classification algorithm stages are the main parts of a modulation recognition system. Performance of the modulation recognition system mainly depends on the prior knowledge of some of the signal parameters, selection of the key features and classification algorithm selection. Unfortunately, most of the features require some of the signal parameters such as carrier frequency, pulse shape, time of arrival, initial phase, symbol rate, signal to noise ratio, to be known or to be extracted. Thus, in this thesis, features which do not require prior knowledge of the signal parameters, such as the number of the peaks in the envelope histogram and the locations of these peaks, the number of peaks in the frequency histogram, higher order moments of the signal are considered. Particularly, symbol rate and signal to noise ratio estimation methods are surveyed. A method based on the cyclostationarity analysis is used for symbol rate estimation and a method based on the eigenvector decomposition is used for the estimation of signal to noise ratio. Also, estimated signal to noise ratio is used to improve the performance of the classification algorithm. Two methods are proposed for modulation recognition: 1) Decision tree based method 2) Bayesian based classification method A method to estimate the symbol rate and carrier frequency offset of minimum-shift keying (MSK) signal is also investigated.

Automatic speech recognition has become a standard feature on many consumer electronics and automotive products, and the accuracy of the decoded speech has improved dramatically over time. Often, designers of these products achieve accuracy by employing microphone arrays and beamforming algorithms to reduce interference. However, beamforming microphone arrays are too large for small form factor products such as smart watches. Yet these small form factor products, which have precious little space for tactile user input (i.e. knobs, buttons and touch screens), would benefit immensely from a user interface based on reliably accurate automatic speech recognition. This thesis proposes a solution for interference mitigation that employs blind source separation with a compact array of commercially available unidirectional microphone elements. Such an array provides adequate spatial diversity to enable blind source separation and would easily fit in a smart watch or similar small form factor product. The solution is characterized using publicly available speech audio clips recorded for the purpose of testing automatic speech recognition algorithms. The proposal is modelled in different interference environments and the efficacy of the solution is evaluated. Factors affecting the performance of the solution are identified and their influence quantified. An expectation is presented for the quality of separation as well as the resulting improvement in word error rate that can be achieved from decoding the separated speech estimate versus the mixture obtained from a single unidirectional microphone element. Finally, directions for future work are proposed, which have the potential to improve the performance of the solution thereby making it a commercially viable product.

The aim of this study is to investigate whether blind individuals display cognitive advantages over sighted individuals with regard to second language acquisition. Previous studies from neuropsychology have indicated that this is the case. It has been found that blind L1 speakers can compensate for loss of vision by developing better perceptual and cognitive skills compared to sighted individuals, skills that are highly relevant to language acquisition. These studies do not, however, investigate blind L2 speakers, for whom it is not clear whether these advantages are also found.  In all, 80 adults participated in the study: 40 L2 speakers of Swedish (11 early blind, 9 late blind, 20 sighted, AO<18) and a matching group and subgroups of L1 speakers. These speakers underwent tests on speech perception in noise, accentedness in an L2 and memory functions. The results revealed that L2 speakers are at a great disadvantage perceiving speech in noise compared to L1 speakers, and that there was no advantage associated with blindness. In the L1 speakers group, however, the results revealed that the early blind had advantages compared to the late blind and sighted in white noise, but that both blind groups were more negatively affected by babble noise than the sighted. The results in relation to accentedness in an L2 revealed that there were no advantages associated with blindness. The results further revealed there were no advantages associated with blindness on the episodic memory test. The results did, however, reveal that the early blind performed significantly better than the late blind and sighted on all phonological short-term memory tests and that both the early and late blind were significantly better than the sighted on recognition memory for new words, irrespective of language background. The conclusion is that blindness is associated with advantages in, for example, ability to learn new words and syntax, acquisition rate, ultimate L2 attainment, and language aptitude.

L'objectif de cette thèse est de proposer de nouveaux algorithmes pour surmonter les limitations actuelles et de relever les défis ouverts dans le traitement de l'imagerie spectroscopique par résonance magnétique (ISRM). L'ISRM est une modalité non invasive capable de fournir la distribution spatiale des composés biochimiques (métabolites) utilisés comme biomarqueurs de la maladie. Les informations fournies par l'ISRM peuvent être utilisées pour le diagnostic, le traitement et le suivi de plusieurs maladies telles que le cancer ou des troubles neurologiques. Cette modalité se montre utile en routine clinique notamment lorsqu'il est possible d'en extraire des informations précises et fiables. Malgré les nombreuses publications sur le sujet, l'interprétation des données d'ISRM est toujours un problème difficile en raison de différents facteurs tels que le faible rapport signal sur bruit des signaux, le chevauchement des raies spectrales ou la présence de signaux de nuisance. Cette thèse aborde le problème de l'interprétation des données d'ISRM et la caractérisation de la rechute des patients souffrant de tumeurs cérébrales. Ces objectifs sont abordés à travers une approche méthodologique intégrant des connaissances a priori sur les données d'ISRM avec une régularisation spatio-spectrale. Concernant le cadre applicatif, cette thèse contribue à l'intégration de l'ISRM dans le workflow de traitement en radiothérapie dans le cadre du projet européen SUMMER (Software for the Use of Multi-Modality images in External Radiotherapy) financé par la Commission européenne (FP7-PEOPLE-ITN)
The aim of this thesis is to propose new algorithms to overcome the current limitations and to address the open challenges in the processing of magnetic resonance spectroscopic imaging (MRSI) data. MRSI is a non-invasive modality able to provide the spatial distribution of relevant biochemical compounds (metabolites) commonly used as biomarkers of disease. Information provided by MRSI can be used as a valuable insight for the diagnosis, treatment and follow-up of several diseases such as cancer or neurological disorders. Obtaining accurate and reliable information from in vivo MRSI signals is a crucial requirement for the clinical utility of this technique. Despite the numerous publications on the topic, the interpretation of MRSI data is still a challenging problem due to different factors such as the low signal-to-noise ratio (SNR) of the signals, the overlap of spectral lines or the presence of nuisance components. This thesis addresses the problem of interpreting MRSI data and characterizing recurrence in tumor brain patients. These objectives are addressed through a methodological approach based on novel processing methods that incorporate prior knowledge on the MRSI data using a spatio-spectral regularization. As an application, the thesis addresses the integration of MRSI into the radiotherapy treatment workflow within the context of the European project SUMMER (Software for the Use of Multi-Modality images in External Radiotherapy) founded by the European Commission (FP7-PEOPLE-ITN framework)

Made available in DSpace on 2009-11-18T18:56:41Z (GMT). No. of bitstreams: 1 jufigale.pdf: 544424 bytes, checksum: cfef2e560c1f213101132dc90a70c55a (MD5) Previous issue date: 2009
This work is funded based on the uneasiness with the concept of State as a public machine for development. Of State as a public machine to deliberate valid practices for valid methods and to limit valid subjects in valid spaces. In midst of this specific context, this work dedicates itself to investigate the following research problem: the mistaken recognition of the blind subject in public spaces of representation. For this reason, it was addressed the following question: how the blind subject is recognized in public spaces of representation? To answer the question, it was necessary to contextualize how the blind subject is being recognized in various public spaces of representation. In the international scope, the human rights debate held between the National States was analyzed (BRAND, 2005; KOERNER, 2002; UN, 2006). In the national arena, constitutional rights, federal laws, public policies and institutions representing the blind subject were examined (CABRAL, 2008; SARAVIA, 2006). Finally, in a local context, the fundaments of the concept of citizen for the subject recognition were investigated (AGAMBEN, 2002; RORTY, 1999, DELEUZE AND GUATTARI, 1996). The methodology included reports of national and international representatives in the Lusophone Countries Meeting for Dissemination and Implementation of the Convention on the Rights of Persons with Disabilities and, mainly, interviews with blind subjects. The data was processed by content analysis and was discussed based on the following categories: representation spaces; representation modes; representation amplitude; representation premises. The results show, regarding such spaces of representation, the growing importance of thinking the rights of persons with disabilities ¿ group in which belongs the blind subject ¿ as of the international and national scenario. However, the blind subjects announced alternative local spaces for representation: church, internet, radio, etc. Regarding the representation modes, the role of law and standards has been advocated specially in the human rights field. The importance of the cooperation between the States and the civil society to ensure, in practice, the rights achieved was also emphasized. But other forms of representation, directly linked to each interviewee¿ history, was important. Regarding the representation amplitude, there were arguments in defense of a conception of human dignity and freedom to all inhabitants of the globe. The lusophone event highlighted the concern of the cultural peculiarities of those involved in the meeting. The blind interviewees argued for citizenship as construction of instruments for freedom and autonomy, but recognized that this is not a clear desire between the blind people in general, and even less in society as a whole. With respect to the representation premises, the fundaments for the recognition of the blind subject were based on the primacy of reason at the expense of personal experimentation. Experimentation that serves as the foundation of a new form of recognition of the blind subject in public spaces of representation, one more interested in singularities, impenetrable by reason, unmovable to another, and which are irreducible to each subject. The final considerations suggest that if the State has a reason to be, this is not another than to offer instruments to manifest as many as the existential possibilities of the subject. This is the concept of State for development.
Este trabalho se fundou a partir do desconforto com a concepção de Estado como máquina pública para o desenvolvimento. De Estado como máquina pública a deliberar práticas válidas por métodos válidos e a circunscrever sujeitos válidos em espaços válidos. Inserido nesse contexto, este trabalho se dedicou a investigar o seguinte problema de pesquisa: o indevido reconhecimento do sujeito cego em espaços públicos de representação. Para isso, foi endereçada a seguinte pergunta: de que forma o sujeito cego é reconhecido em espaços públicos de representação? Para responder a pergunta, foi preciso contextualizar como o sujeito cego vem sendo reconhecido nos diversos espaços públicos de representação. No espaço internacional, foi analisado o debate sobre os direitos humanos travado entre os Estados Nacionais (BRAND, 2005; KOERNER, 2002; ONU, 2006). No espaço nacional, foram examinados os direitos constitucionais, as leis federais, políticas públicas e instituições representativas do sujeito cego (CABRAL, 2008; SARAVIA, 2006). Por fim, no espaço local, foram investigados os fundamentos do conceito de cidadão para reconhecimento do sujeito (AGAMBEN, 2002; RORTY,1999; DELEUZE E GUATTARI, 1996). A metodologia adotada contemplou relatos de representantes nacionais e internacionais presentes no Encontro dos Países Lusófonos para Divulgação e Implementação da Convenção sobre os Direitos das Pessoas com Deficiência e, primordialmente, entrevistas com sujeitos cegos. Os dados foram tratados por análise de conteúdo e discutidos a partir das seguintes categorias: instâncias de representação; formas de representação; amplitude de representação; e premissas de representação. Os resultados apontam, quanto às instâncias de representação, a importância crescente do cenário internacional e nacional para pensar os direitos das pessoas com deficiência ¿ grupo dentro do qual se inserem os sujeitos cegos. Contudo, os sujeitos cegos anunciaram espaços locais alternativos para representação: igreja, internet, rádio, etc. Quanto às formas de representação, o papel da lei e das normas foi defendido especialmente no âmbito dos direitos humanos. Foi também ressaltada a importância da cooperação entre Estados e com a sociedade civil para garantir, na prática, os direitos conquistados. Mas outras formas de representação, diretamente vinculadas à história de cada entrevistado, se mostraram importantes. Em relação à amplitude de representação, pesaram argumentos em defesa de uma concepção de dignidade e liberdade humanas a todos os habitantes do globo. O evento dos países lusófonos apontou a preocupação com respeito às peculiaridades culturais dos envolvidos no encontro. Os cegos entrevistados defenderam cidadania como construção de instrumentos para a liberdade e autonomia, mas assumiram que este desejo não é claro entre os cegos de uma forma geral e, menos ainda, na sociedade como um todo. Com respeito às premissas de representação, os fundamentos de reconhecimento do sujeito cego se assumem calcados na primazia da razão em detrimento da experimentação pessoal. Experimentação esta que funda uma nova forma de reconhecimento do sujeito cego em espaços públicos de representação, mais interessada em singularidades, impenetráveis pela razão, intransferíveis a outro, e que fazem de cada sujeito irredutível. As considerações finais sugerem que se o Estado tem razão de ser, esta não é outra que oferecer instrumentos para que possam se manifestar tantas quantas forem as possibilidades de existência dos sujeitos. Esta é a concepção de Estado para o desenvolvimento.

碩士
中華大學
生物資訊學系(所)
97
In this study, we developed a wisdom system that can actively remind the computer user to let his/her eyes take a break. This system uses a PC camera to capture the image of the user and the background. It applies recognition technology to validate the user. At the same time, the system uses a built-in timer to record how long the user has worked on the computer. When the user is using the computer, the system not only detects the visual distance but also count the eye’s blinks by the facial feature. If the user doesn’t keep a proper visual distance from the monitor of the computer, his/her eyes’ blinks too often or too less, his/her has used the computer for a long time, illumination is too bright or too dark according to the background lighting behind the user, or the user uses computer not within a proper time, the system will pop up a warning message via words /voice to inform the user. Therefore, this system can initiatively be used to protect computer user’s eyes from harm. This system consists of three parts: the face detection, the face recognition, the visual distance detection and the blink detection technology. In the face detection part, a PC camera is used to capture the color image, and then the environment noise is removed by doing connected component labeling and morphology process to find human face. In the face recognition part, we use Principal Component Analysis technology to extract the feature of human face, and then use Euclidean distance technology to validate the user. In the visual distance detection part, Ellipse mask method is used for obtaining face information and detecting user visual distance from the monitor of computer. In the blink detection part, set the eyes region to count the blinks frequency to estimate the user weary degree.

碩士
國立雲林科技大學
電機工程系
102
Conventional face recognition authentication system is unable to discriminate real-world human faces from faces in the photo or in the video, and is easily faked by faces in the photo or in the video. So live face recognition authentication system with accurate live detection is getting fundamental and crucial. In order to accomplish real-time live face recognition authentication system, live detection based on blink detection and some extra simple live motion analysis is a cost-effective and high-practicability auxiliary method, especially the blink detection is the most intuitive and noninvasive. 　　This thesis proposes Parallel-AdaBoost blink detection to perform AdaBoost eyes detection stage and AdaBoost open-eyes detection stage simultaneously through multithreaded parallelism. That is, AdaBoost open-eyes detection works on the region of interest of eyes candidate detected by AdaBoost eyes detection simultaneously and iteratively. If the detection result of AdaBoost open-eyes detection is true, the eyes are open. If otherwise, the eyes are half-closed or completely-closed. Live blink motion is determined by two alternatively-happened detection results of AdaBoost open-eyes detection. 　　Experimental results show the detection accuracy and execution speed of proposed Parallel-AdaBoost blink detection are better than conventional ones. The proposed Parallel-AdaBoost blink detection can prevent face recognition authentication system from being broken by faces in the photo, and some extra simple live motion request and analysis can be integrated to further prevent face recognition authentication system from being cheated by faces in the video. 　　On the other hand, on Android embedded platform, this thesis implements and integrates the proposed Parallel-AdaBoost blink detection to accomplish “Portable 4-Factor Authentication Device Accelerated By Multi-Thread Pipeline Parallelism” for applications of mobile payment, mobile authentication, and mobile locker.

碩士
國立高雄應用科技大學
電子與資訊工程研究所碩士班
96
ABSTRACT With technological development, several of the extensive application for the visually impaired that has brought new hope, also by image recognition technology, began to help the visually impaired learning problems. At present the visually impaired and hearing sensitivity is mainly to learn, and Braille and audio equipment that is visually impaired learning of the media, but if there is no appropriate learning assistant device, it will cause the visually impaired and the general public a greater learning gap. In this paper, designed by e-learning assistant device (Electronic Learning Aids, ELA) platform, in order to help the visually impaired to hinder learning to overcome the obstacles, the use of ETOMS Electronics Corp. ET21X130D 160x120 Pixel Image Sensor (CMOS Image Sensor, CIS) As a real-time imaging, detected at any time studying images on the card surface targets, through graphic indicators, bearing the main further information other than the message, through ELAN Microelectronics Corp. ESLZ000 digital signal processors (Digital Signal Processor, DSP) have a compression The interception of information and image feature value, in order to get the message to do additional identification and judgment, the output generated messages and audio and visually impaired persons have interactive learning and then to the establishment of self-confidence and self-learning objectives. Keyword: the visually impaired, ELA, DSP, image recognition, CIS

碩士
國立交通大學
資訊科學學系
86
Forms are composed of fields and items. Among the fields and items of a data form, there exist some special dependence relationships. To improvethe efficiency of form processing, in addition to good form segmentationand recognition process, the understanding of the dependence relationshipsis also necessary. In this study, new approaches to segmentation of blankdata forms, recognition of form formats, and understanding of field anditem dependence relationships are proposed. In the segmentation of a blankdata form image, the fields are extracted by alternative vertical andhorizontal projections and a logical tree structure of the form is constructed.After segmentation, form encoding is performed by traversing the tree structure. A form can be transformed into an attributed string, and the attributed string is proven to be a unique code. It can be used to recognizeforms by attributed string matching. In the understanding of a blank data form,various types of dependence relationships are identified and methods for detection of the dependence relationships are proposed. In the proposed form processing system, a window-based interface for the above works is also provided. Some experimental results showing the feasibility of the proposed methods are also included.

碩士
銘傳大學
資訊傳播工程學系碩士班
108
The vehicle is equipped with mirrors inside and outside, which is the primary tool for driver to observe with the environment around the vehicle. However, the driving blind spots area generated by the vehicle, which has always been the leading cause of traffic accidents. The fisheye camera image could be displayed half-sphere-wide image. In case of the three fisheye cameras installed at different positions outside the vehicle, which might be efficiently catch-all driving visual blind spots area. So through the graphic recognition technology, object recognition for driving blind spots could be carried out to reduce traffic accidents that might be caused by driving blind spots. Due to the fisheye image is close to the periphery, the object in the image will have the higher the distortion. Under the limitation of the object definition, it is difficult to define the object in the blind spots area. This paper employs the convolutional neural network (CNN) method to improve the accuracy of object recognition in the blind zone. For the advantages of the convolution model, this paper employs VGG Net-based and ResNet-based multiple convolution models to achieve merge model research, which could make the identification accuracy to have more accurate for different types of objects appearing in the visual blind zone. Finally, the fisheye imaging experiment of the actual driving is probably used to display the reference more intuitively to achieve the warning function for drivers.

碩士
國立成功大學
電機工程學系碩博士班
97
Speech is the most primitive and efficient method for human communication. With the development of science and technology, we use computer to accomplish many complex operations and applications. Therefore, how to make computers to understand the human language is one of the most important parts in speech processing. As speech recognition is the first step for language understanding, it is our principal issue to recognize speech information more efficiently and accurately. In general, speech recognition systems can be classified into two types according to the environment of use. The first type is near filed handheld microphone and the other is far filed hands-free device. For the near filed handheld microphone, it can achieve better performance in recognition rate because of its low interference from environment; for the far filed hands-free device, the noise and reverberation will easily cause the decrease of recognition rate. Besides, the speech recognition algorithms can be classified into speaker-dependent and speaker-independent ones. Speaker-dependent speech recognition needs a training procedure to construct the corresponding acoustical model for each speaker. For the speaker-independent speech recognition, a generalized acoustical model is trained so that no further training is required for use. Considering the convenience for users, it is well accepted to integrate the far field hands-free devices with the speaker-independent speech recognition algorithm. The far field hands-free device usually collects the sounds through the microphone array. A microphone array can enhance the speech quality or make up for the lack of single-channel microphone. In this thesis, we use two microphones to collect the sounds, and extract target speech by blind source separation (BSS). The residue noise is then removed by the subspace enhancement method. Finally, the enhanced speech is recognized by a speech recognition system. In this thesis, we construct the speech recognition system via HTK toolkit developed by University of Cambridge. The experimental results show that the proposed system is suitable for several presented noisy environments, and it effectively improves the recognition rate by 20%. For the SNR evaluation, the proposed system can make enhanced speech SNR have 20 dB higher than original corrupted speech which was ranged from 0dB to 10dB.

碩士
國立高雄應用科技大學
電機工程系博碩士班
101
This research presents a mixed state estimation method to estimate the source number in test audio and determine whether applied blind source separation technology to be separated and recognized. This method solves the traditional voiceprint recognition does not identify all the species in mixed signal and the separation number which sets by user in blind source separation technology. In the part of identification, mixed state estimation method estimated the separation number in test audio and identified the separation source signal after using non-negative matrix factor 2D deconvolution(NMF2D) to separate the test audio. The research sets up 7 items of mixed recordings in connection with 6 kinds of frogs into source number estimation, source separated and voiceprint recognition. In the part of 6 kinds of frogs croak which obtained 84.98% recognition rate from 373 non-mixed signal syllables; In mixed recordings part which obtained 78.43% recognition rate from 269 syllables after separating mixed signals. It proved that the method can obtain better recognition results in the single and mixed signal recognition.

Particularly in recent years, prompted by the need to gain greater independent access to a wider range of information, many persons who are blind make extensive use of screen access technology, optical character recognition devices, refreshable Braille displays and electronic notetakers in a variety of contexts. These reading and writing media have proved to be so useful and effective, raising debates in the literature on whether there is a decline in the use of Braille, or whether Braille as a reading and writing medium would become obsolete. Following a discussion on the development of tactual reading and writing media as part of an historical background to blindness, as well as an evaluation of the various reading and writing media used in South Africa by persons who are blind, this study, using a quantitative approach with a survey design, aimed to determine the impact of the various reading and writing media on the education and employment of persons who are blind. Based on the findings of the study, what emerges forcefully with regard to the preference of a medium for reading or writing is that a greater number of persons who are blind prefer Braille and computers with speech output. Notwithstanding this, there is support for the need to provide instruction in the use of the various reading and writing media, highlighting the critical value and role of the various media. Additionally, while persons who are blind appear to be convinced that computers will not replace Braille, they were, however, divided on whether there is a decline in the use of Braille, and whether computers would replace audiotapes. Finally, conclusions, based mainly on the findings of the study are drawn, and recommendations, both for future research, and for an integrated reading and writing model, are made.
Educational Studies
D.Ed.(Special Needs Educstion)

Σκοπός της παρούσας διπλωματικής εργασίας είναι η δημιουργία ενός συστήματος μίξης ηχητικών σημάτων και προσπάθεια διαχωρισμού τους με βάση τις μεθόδους τυφλού διαχωρισμού σημάτων. Έχοντας ως δεδομένα τα αρχικά σήματα των πηγών γίνεται προσπάθεια, αρχικά μέσω της εφαρμογής της μεθόδου Ανάλυσης Ανεξάρτητων Συνιστωσών (ICA) για την περίπτωση της στιγμιαίας μίξης και στη συνέχεια μέσω της χρήσης αλγορίθμων που στηρίζονται στο μοντέλο παράλληλου παράγοντα (PARAFAC) για την περίπτωση της συνελικτικής μίξης, να προσδιοριστούν τα σήματα των πηγών από τα σήματα μίξης. Επιπλέον, τροποποιώντας τις παραμέτρους του συστήματος που μελετάμε σε κάθε περίπτωση, προσπαθούμε να πετύχουμε τη βέλτιστη απόδοση του διαχωρισμού.
The subject of this diploma thesis is the creation of a mixing system of speech signals and the attempt of their separation using the methods of blind source separation (BSS). Considering the original source signals known, we attempt, firstly by using independent component analysis for instantaneous mixtures and then by using PARAFAC model for convolutive mixtures, to extract the original source signals from the mixing signals. Moreover, by modifying the parameters of the system we make an effort to achieve the best performance of the separation.