Dissertations / Theses on the topic 'Histogrammes de Gradient orienté'

Cette thèse est dédiée à l'étude de méthodes de vision artificielle pour la détection et la reconnaissance d'objets structurés, plus précisément les véhicules automobiles. La première partie est vouée à la détection de véhicules sur des scènes routières à l'aide d'un système embarqué de vision monoculaire. La stratégie utilisée se fonde sur une cascade de classifieurs de type Adaboost qui permet la concaténation des fonctions de classification discriminantes et génératives. Nous avons proposé aussi des méthodes pour classifier les véhicules détectés. La deuxième partie est consacrée à la reconnaissance du type d'un véhicule (constructeur, modèle) à partir de sa vue de face. L'application principale visée est le contrôle d'accès dans des parkings ou péages d'autoroutes. Le système système de reconnaissance multi-classes utilise un descripteur visuel local, à base de pixels de contour orientés. La classification est obtenue à partir d'une méthode de votes, robuste aux occultations partielles.

This work presents improvements to Monte Carlo Localization (MCL) for a mobile robot using computer vision. Solutions to the localization problem aim to provide fine resolution on location approximation, and also be resistant to changes in the environment. One such environment change is the kidnapped/teleported robot problem, where a robot is suddenly transported to a new location and must re-localize. The standard method of "Augmented MCL" uses particle filtering combined with addition of random particles under certain conditions to solve the kidnapped robot problem. This solution is robust, but not always fast. This work combines Histogram of Oriented Gradients (HOG) computer vision with particle filtering to speed up the localization process. The major slowdown in Augmented MCL is the conditional addition of random particles, which depends on the ratio of a short term and long term average of particle weights. This ratio does not change quickly when a robot is kidnapped, leading the robot to believe it is in the wrong location for a period of time. This work replaces this average-based conditional with a comparison of the HOG image directly in front of the robot with a cached version. This resulted in a speedup ranging from from 25.3% to 80.7% (depending on parameters used) in localization time over the baseline Augmented MCL.

Ce travail de thèse a été réalisé dans le cadre de la coopération entre l’Université de Technologie de Compiègne (UTC) et le Centre Technique des Industries Mécaniques (CETIM). Nous présentons un système de détection de personnes pour l’aide à la conduite dans les engins de chantier. Une partie du travail a été dédiée à l’analyse du contexte de l’application, ce qui a permis de proposer un système de perception composé d’une caméra monoculaire fisheye et d’un Lidar. L’utilisation des caméras fisheye donne l’avantage d’un champ de vision très large avec en contrepartie, la nécessité de gérer les fortes distorsions dans l’étape de détection. A notre connaissance, il n’y a pas eu de recherches dédiées au problème de la détection de personnes dans les images fisheye. Pour cette raison, nous nous sommes concentrés sur l’étude et la quantification de l’impact des distorsions radiales sur l’apparence des personnes dans les images et nous avons proposé des approches adaptatives pour gérer ces spécificités. Nos propositions se sont inspirées de deux approches de l’état de l’art pour la détection des personnes : les histogrammes de gradient orientés (HOG) et le modèle des parties déformables (DPM). Tout d’abord, en enrichissant la base d’apprentissage avec des imagettes fisheye artificielles, nous avons pu montrer que les classificateurs peuvent prendre en compte les distorsions dans la phase d’apprentissage. Cependant, adapter les échantillons d’entrée, n’est pas la solution optimale pour traiter le problème de déformation de l’apparence des personnes dans les images. Nous avons alors décidé d’adapter l’approche de DPM pour prendre explicitement en compte le modèle de distorsions. Il est apparu que les modèles déformables peuvent être modifiés pour s’adapter aux fortes distorsions des images fisheye, mais ceci avec un coût de calculatoire supérieur. Dans cette thèse, nous présentons également une approche de fusion Lidar/camera fisheye. Une architecture de fusion séquentielle est utilisée et permet de réduire les fausses détections et le coût calculatoire de manière importante. Un jeu de données en environnement de chantier a été construit et différentes expériences ont été réalisées pour évaluer les performances du système. Les résultats sont prometteurs, à la fois en terme de vitesse de traitement et de performance de détection
This thesis has been carried out in the framework of the cooperation between the Compiègne University of Technology (UTC) and the Technical Centre for Mechanical Industries (CETIM). In this work, we present a vision-based multi-sensors people detection system for safety on heavy machines. A perception system composed of a monocular fisheye camera and a Lidar is proposed. The use of fisheye cameras provides an advantage of a wide field-of-view but yields the problem of handling the strong distortions in the detection stage.To the best of our knowledge, no research works have been dedicated to people detection in fisheye images. For that reason, we focus on investigating and quantifying the strong radial distortions impacts on people appearance and proposing adaptive approaches to handle that specificity. Our propositions are inspired by the two state-of-the-art people detection approaches : the Histogram of Oriented Gradient (HOG) and the Deformable Parts Model (DPM). First, by enriching the training data set, we prove that the classifier can take into account the distortions. However, fitting the training samples to the model, is not the best solution to handle the deformation of people appearance. We then decided to adapt the DPM approach to handle properly the problem. It turned out that the deformable models can be modified to be even better adapted to the strong distortions of the fisheye images. Still, such approach has adrawback of the high computation cost and complexity. In this thesis, we also present a framework that allows the fusion of the Lidar modality to enhance the vision-based people detection algorithm. A sequential Lidar-based fusion architecture is used, which addresses directly the problem of reducing the false detections and computation cost in vision-based-only system. A heavy machine dataset have been also built and different experiments have been carried out to evaluate the performances of the system. The results are promising, both in term of processing speed and performances

De nos jours les algorithmes de vision par ordinateur abondent dans les applications de vidéo-surveillance, de reconstruction 3D, de véhicules autonomes, d'imagerie médicale, etc… La détection et la mise en correspondance d'objets dans les images constitue une étape clé dans ces algorithmes.Les méthodes les plus communes pour la mise en correspondance d'objets ou d'images sont basées sur des descripteurs locaux, avec tout d'abord la détection de points d'intérêt, puis l'extraction de caractéristiques de voisinages des points d'intérêt, et enfin la construction des descripteurs d'image.Dans cette thèse, nous présentons des contributions au domaine de la mise en correspondance d'images par l'utilisation de demi filtres tournants. Nous suivons ici trois approches : la première présente un nouveau descripteur à faible débit et une stratégie de mise en correspondance intégrés à une plateforme vidéo. Deuxièmement, nous construisons un nouveau descripteur local en intégrant la réponse de demi filtres tournant dans un histogramme de gradient orienté (HOG) ; enfin nous proposons une nouvelle approche pour la construction d'un descripteur utilisant des statistiques du second ordre. Toutes ces trois approches apportent des résultats intéressants et prometteurs.Mots-clés : Demi filtres tournants, descripteur local d'image, mise en correspondance, histogramme de gradient orienté (HOG), Différence de gaussiennes
Nowadays computer vision algorithms can be found abundantly in applications relatedto video surveillance, 3D reconstruction, autonomous vehicles, medical imaging etc. Image/object matching and detection forms an integral step in many of these algorithms.The most common methods for Image/object matching and detection are based on localimage descriptors, where interest points in the image are initially detected, followed byextracting the image features from the neighbourhood of the interest point and finally,constructing the image descriptor. In this thesis, contributions to the field of the imagefeature matching using rotating half filters are presented. Here we follow three approaches:first, by presenting a new low bit-rate descriptor and a cascade matching strategy whichare integrated on a video platform. Secondly, we construct a new local image patch descriptorby embedding the response of rotating half filters in the Histogram of Orientedgradient (HoG) framework and finally by proposing a new approach for descriptor constructionby using second order image statistics. All the three approaches provides aninteresting and promising results by outperforming the state of art descriptors.Key-words: Rotating half filters, local image descriptor, image matching, Histogram of Orientated Gradients (HoG), Difference of Gaussian (DoG)

Performance of automatic speech recognition (ASR) systems utilizing only acoustic information degrades significantly in noisy environments such as a car cabins. Incorporating audio and visual information together can improve performance in these situations. This work proposes a lip detection and tracking algorithm to serve as a visual front end to an audio-visual automatic speech recognition (AVASR) system. Several color spaces are examined that are effective for segmenting lips from skin pixels. These color components and several features are used to characterize lips and to train cascaded lip detectors. Pre- and post-processing techniques are employed to maximize detector accuracy. The trained lip detector is incorporated into an adaptive mean-shift tracking algorithm for tracking lips in a car cabin environment. The resulting detector achieves 96.8% accuracy, and the tracker is shown to recover and adapt in scenarios where mean-shift alone fails.

This master thesis describes the design and development of the system for detection and recognition of whole coat of arms as well as each heraldic parts. In the thesis are presented methods of computer vision for segmentation and detection of an object and selected methods that are the most suitable. Most of the heraldic parts are segmented using a convolution neural networks and the rest using active contours. The Histogram of the gradient method was selected for coats of arms detection in an image. For training and functionality verification is used my own data set. The resulting system can serve as an auxiliary tool used in auxiliary sciences of history.

Le but de cette thèse est de développer des méthodes pratiques plus performantes pour la détection d'instances de classes d'objets de la vie quotidienne dans les images. Nous présentons une famille de détecteurs qui incorporent trois types d'indices visuelles performantes - histogrammes de gradients orientés (Histograms of Oriented Gradients, HOG), motifs locaux binaires (Local Binary Patterns, LBP) et motifs locaux ternaires (Local Ternary Patterns, LTP) - dans des méthodes de discrimination efficaces de type machine à vecteur de support latent (Latent SVM), sous deux régimes de réduction de dimension - moindres carrées partielles (Partial Least Squares, PLS) et sélection de variables par élagage de poids SVM (SVM Weight Truncation). Sur plusieurs jeux de données importantes, notamment ceux du PASCAL VOC2006 et VOC2007, INRIA Person et ETH Zurich, nous démontrons que nos méthodes améliorent l'état de l'art du domaine. Nos contributions principales sont : Nous étudions l'indice visuelle LTP pour la détection d'objets. Nous démontrons que sa performance est globalement mieux que celle des indices bien établies HOG et LBP parce qu'elle permet d'encoder à la fois la texture locale de l'objet et sa forme globale, tout en étant résistante aux variations d'éclairage. Grâce à ces atouts, LTP fonctionne aussi bien pour les classes qui sont caractérisées principalement par leurs structures que pour celles qui sont caractérisées par leurs textures. En plus, nous démontrons que les indices HOG, LBP et LTP sont bien complémentaires, de sorte qu'un jeux d'indices étendu qui intègre tous les trois améliore encore la performance. Les jeux d'indices visuelles performantes étant de dimension assez élevée, nous proposons deux méthodes de réduction de dimension afin d'améliorer leur vitesse et réduire leur utilisation de mémoire. La première, basée sur la projection moindres carrés partielles, diminue significativement le temps de formation des détecteurs linéaires, sans réduction de précision ni perte de vitesse d'exécution. La seconde, fondée sur la sélection de variables par l'élagage des poids du SVM, nous permet de réduire le nombre d'indices actives par un ordre de grandeur avec une réduction minime, voire même une petite augmentation, de la précision du détecteur. Malgré sa simplicité, cette méthode de sélection de variables surpasse toutes les autres approches que nous avons mis à l'essai.

Submitted by Johnny Rodrigues (johnnyrodrigues@ufcg.edu.br) on 2018-04-18T15:01:39Z No. of bitstreams: 1 MARCELL MANFRIN BARBACENA - DISSERTAÇÃO PPGCC 2014..pdf: 1468565 bytes, checksum: b94d20ffdace21ece654986ffd8fbb63 (MD5)
Made available in DSpace on 2018-04-18T15:01:39Z (GMT). No. of bitstreams: 1 MARCELL MANFRIN BARBACENA - DISSERTAÇÃO PPGCC 2014..pdf: 1468565 bytes, checksum: b94d20ffdace21ece654986ffd8fbb63 (MD5) Previous issue date: 2014
Impulsionadas pela crescente demanda por sistemas de segurança para proteção do indivíduo e da propriedade nos dias atuais, várias pesquisas têm sido desenvolvidas com foco na implantação de sistemas de vigilância por vídeo com ampla cobertura. Um dos problemas de pesquisa em aberto nas áreas de visão computacional e redes de computadores envolvem a escalabilidade desses sistemas, principalmente devido ao aumento do número de câmeras transmitindo vídeos em tempo real para monitoramento e processamento. Neste contexto, o objetivo geral deste trabalho é avaliar o impacto que a redução da taxa de transmissão dos fluxos de vídeos impõe na eficácia dos algoritmos de detecção de pessoas utilizados em sistemas inteligentes de videovigilância. Foram realizados experimentos utilizando vídeos em alta resolução no contexto de vigilância com tomadas externas e com um algoritmo de detecção de pessoas baseado em histogramas de gradientes orientados, nos quais se coletou, como medida de eficácia do algoritmo, a métrica de área sob a curva de precisão e revocação para, em sequência, serem aplicados os testes estatísticos de Friedman e de comparações múltiplas com um controle na aferição das hipóteses levantadas. Os resultados obtidos indicaram que é possível uma redução da taxa de transmissão em mais de 70% sem que haja redução da eficácia do algoritmo de detecção de pessoas.
Motivated by the growing demand for security systems to protect persons and properties in the nowadays, several researches have been developed focusing on the deployment of widearea video coverage surveillance systems. One open research problem in the areas of computer vision and computer networks involves the scalability of these systems, mainly due to the increasing number of cameras transmitting real-time video for monitoring and processing. In this context, the aim of this study was to evaluate the impact that transmission data-rate reduction of video streams imposes on the effectiveness of people detection algorithms used in intelligent video surveillance systems. With a proposed experimental design, experiments were performed using high-resolution wide-area external coverage video surveillance and using an algorithm for people detection based on histograms of oriented gradients. As a measure of effectiveness of the people detection algorithm, the metric of area under the precision-recall curve was collected and statistical tests of Friedman and multiple comparisons with a control were applied to evaluate the hypotheses. The results indicated that it is possible to reduce transmission rate by more than 70% without decrease in the effectiveness of the people detection algorithm.

碩士
國立臺北科技大學
電子工程系
106
A high precision and fast pedestrian detection system is always playing an important role in applications of driver assistant, surveilance systems. Recently, these kind of technology become more popular and widely used in our life. However, implementing a human detection system needs a reliable feature extraction algorithm to conquer interference from different kind of environments. As a result, our paper used Histogram of Oriented Gradient (HOG) algorithm to extract feature from computer vision images. Although HOG is so good at handling those issues, it still has a deadly disadvantage, it takes too much computation time. In order to achieve a fast and reliable pedestrian detection system, we used FPGA to implement HOG algorithm and simplified those complicated formula such as square root and arctangent operations. At last, we implement the proposed method on Altera Stratix IV platform with PCI Express interface and achieved 207 MHz operating frequency.

碩士
義守大學
資訊工程學系
103
With the popularity of Smartphone, the applications of QR Code become more diverse. To store information and advertising sale, we can often find QR Codes in a periodical or on the poster. General users can utilize mobile phone''s application to decode QR Code easily. Such mobile application requires to get close to and point at the QR Code to read data. This study uses HOGs (Histograms of Oriented Gradients) to perform QR code detection. It aims to improve the recognition rate of QR Code in an image. When extracting HOG feature from input image, each pixel will be divided into nine directions with intensities, We then use feature information to identify the presence of QR Code in an image. Since a QR Code consists of black and white small boxes, through the HOG feature, information can easily recognized. To achieve this function, we use AdaBoost method to efficiently classify the existence of QR codes.

碩士
國立臺灣大學
資訊工程學研究所
96
In this thesis we introduce an Augmented Histograms of Oriented Gradients (AHOG) feature for human detection from a non-static camera. This research tries to increase the discriminating power of original Histograms of Oriented Gradients (HOG) feature by adding human shape properties, such as contour distances, symmetry, gradient density, and shape approximation. The relations among AHOG features are characterized by the contour distances to the centroid of human. By observing on the biological structure of a human shape, we impose the symmetry property on every HOG feature and compute the similarity between feature itself and its symmetric pair so as to weigh HOG features. After that, the capability of describing human features is greatly improved when being compared with that of traditional one, especially when the moving humans are under consideration. Besides, we also augment the gradient density into AHOG to mitigate the influences caused by repetitive backgrounds. Moreover, we reject the false detections via an elliptical verifier learned when one tries to approximate a human shape. In the experiments, our proposed human detection method demonstrates highly reliable accuracy and provides the comparable performance to the state-of-the-art human detector on different databases.

碩士
國立高雄第一科技大學
電腦與通訊工程所
97
In this work, we propose an algorithm of combining Histograms of Oriented Gradients(HOG) with global feature for human detection from a non-static camera. We use AdaBoost algorithm to learn local characteristics of human based on HOG by giving massive training samples. The human detector is represented by a set of selected discriminative HOG features. Since local feature is easily affected by complex backgrounds and noise, the idea of this work is to incorporate the global feature for improving the detection accuracy. Here, we adopt the head contour as the global feature. The score for evaluating the existence of the head contour is through the Chamfer distance. Furthermore, the score distributions of the pedestrian and non-pedestrian are modeled by Gaussian and Anova distributions, respectively. The combination of the human detector based on local features and head contour is achieved through the adjustment of the hyperplane of support vector machine. In the experiments, we exhibit that our proposed human detection method not only has higher detection rate but also lower false positive rate in comparison with the state-of-the-art human detector.

Η παρούσα ειδική ερευνητική εργασία εκπονήθηκε στα πλαίσια του Διατμηματικού Προγράμματος Μεταπτυχιακών Σπουδών στην “Ηλεκτρονική και Επεξεργασία της Πληροφορίας”, στο Τμήμα Φυσικής του Πανεπιστημίου Πατρών. Αντικείμενο της παρούσας εργασίας είναι η “Υλοποίηση σε FPGA του περιγραφέα HOG για ανίχνευση ανθρώπων σε εικόνες και βίντεο”. Το πρώτο κεφάλαιο αποτελεί μια εισαγωγή στις βασικότερες έννοιες που χρησιμοποιούνται στην παρούσα εργασία. Περιγράφεται επίσης η αναπτυξιακή πλακέτα που χρησιμοποιήθηκε καθώς και τα επί μέρους στοιχεία που τη συνθέτουν. Τέλος γίνεται μια συνοπτική αναφορά σε εργασίες με παρόμοιο αντικείμενο, οι οποίες με επηρέασαν στο σχεδιασμό και την υλοποίηση του συστήματός μου. Στο δεύτερο κεφάλαιο αναλύεται ο περιγραφέας Ιστογραμμάτων Προσανατολισμού της Βάθμωσης ή όπως είναι ευρύτερα γνωστός Histograms of Oriented Gradient Descriptor. Παρουσιάζονται τα βήματα όπως περιγράφονται στην εργασία των Dalal&Triggs[4] και οι βέλτιστες τιμές των παραμέτρων του περιγραφέα. Στο τρίτο κεφάλαιο ακολουθώντας τα βήματα του δευτέρου κεφαλαίου, παρουσιάζεται η διαδικασία υλοποίησης του περιγραφέα στο Matlab. Εκτός της υλοποίησης έγινε και μια προεργασία για τη μεταφορά του σε γλώσσα περιγραφής υλικού. Η προεργασία αυτή περιλαμβάνει απλοποιήσεις και τροποποιήσεις με σκοπό να μειωθεί το υπολογιστικό κόστος. Τέλος παρουσιάζονται τα αποτελέσματα δοκιμών της απόδοσης του περιγραφέα για τις διάφορες απλοποιήσεις. Στο τέταρτο κεφάλαιο γίνεται μια μικρή αναφορά στους ταξινομητές. Περιγράφονται οι ταξινομητές που δοκιμάστηκαν στην παρούσα εργασία ως προς συγκεκριμένα χαρακτηριστικά τους καθώς και την υπολογιστική τους πολυπλοκότητα για την συγκεκριμένη εφαρμογή. Το πέμπτο και τελευταίο κεφάλαιο περιλαμβάνει την περιγραφή της υλοποίησης σε VHDL. Αναλύονται τα επί μέρους κυκλώματα και όπου κρίθηκε αναγκαίο χρησιμοποιήθηκαν σχήματα ή πίνακες. Σε κάποιες περιπτώσεις δίνονται και οι κυματομορφές των κυκλωμάτων.
This thesis took place within the frame work of the Interdeparmental Master’s Program in “Electronics and Information Processing”, at the Department of Physics of University of Patras. The objective of this work is the implementation in FPGA of the HOG descriptor for the detection of people, images and videos. The first chapter is an introduction about the basic concepts, which are used across the manuscript. (Additional descriptions concern the development board which was used as well as the individual parts that compose it.) In the end, there is a brief reference to past projects focusing on similar objectives, which influenced the design and the implementation of my system. The second chapter concerns the presentation and discussion of the Histograms of Oriented Gradient descriptor. The steps of the procedure and the best parameter values of the descriptor are presented in a similar way as they are described in the paper of Dalal and Triggs. In the third chapter, following the steps of the previous one, the focus shifts to the descriptor’s implementation procedure in Matlab. Besides the implementation, there is a preparation for the transference of the descriptor in a Hardware Description Language. This preparation includes simplifications and modifications aiming at the reduction of the computational cost. Finally, we see the tests’ results of the descriptor’s performance concerning the various simplifications. The fourth chapter is a partial reference to the classifiers. The description is about the classifiers that were used in the present work with respect to their features and their computational complexity of this particular application. The fifth and final chapter refers to the description of the implementation in VHDL. There is an analysis of the partial circuits and, when necessary, shapes and tables were used. In some cases, the waveforms of the circuits are being presented.