Made available in DSpace on 2016-03-15T19:37:37Z (GMT). No. of bitstreams: 1 Fabio Nascimento Brandao.pdf: 1813031 bytes, checksum: 835f8ef8976c5ae514697bc8287324be (MD5) Previous issue date: 2011-08-10 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Biometrics is a study field which objective is the person recognition using physical and behavioural traits. These traits can be acquired by specific sensors or by cameras. The biometric systems are used for criminal identification, private areas access or document access. The iris recognition is reliable because the iris does not change in the lifetime and rarely it can be changed accidentally or intentionally since the iris is protected inside the human eye. This master thesis verify the possibility of usage the Bag of Keypoints technique for image parts classification, using SURF points inside these parts to use as seeds of the region growing technique and segment iris for a posterior recognition. Obtained results are of the order of 7% segmentation error and, although less than the state of the art (of the order of 3%), these results indicate the possibility of a more detailed verification of this technique, that can be easily adapted to other uses. / A biometria é uma área de estudo que tem como objetivo a identificação de pessoas através de características físicas e comportamentais. Estas características podem ser capturadas por sensores específicos, ou por câmeras. Os sistemas biométricos são utilizados, por exemplo, na identificação criminal, concessão de acesso à áreas privadas ou documentos. O reconhecimento de uma pessoa através da íris é confiável pois a íris não é alterada com o passar dos anos e são raros os casos de alteração acidental ou intencional uma vez que a íris está protegida dentro do olho humano. Esta dissertação explora a possibilidade do uso da técnica de Bag of Keypoints para classificação de partes de imagens, utiliza pontos SURF dentro destas partes para fazer crescimento de região e segmentar a íris para uma etapa posterior de identificação. Os resultados obtidos são da ordem de 7% de erro em termos de segmentação e, embora inferiores ao estado da arte (da ordem de 3%), indicam a possibilidade de uma exploração mais detalhada desta técnica, que pode ainda ser facilmente adaptada para outros usos.

An algorithm which is able to consistently identify features in an image is a basic building block of many object recognition systems. Attaining sufficient consistency is challenging, because factors such as pose and lighting can dramatically change a feature’s appearance. Effective feature identification therefore requires both a reliable and accurate keypoint detector and a discriminative categoriser (or quantiser). The Dual Tree Complex Wavelet Transform (DTCWT) decomposes an image into oriented subbands at a range of scales. The resulting domain is arguably well suited for further image analysis tasks such as feature identification. This thesis develops feature identification in the complex wavelet domain, building on previous keypoint detection work and exploring the use of random forests for descriptor quantisation. Firstly, we extended earlier work on keypoint detection energy functions. Existing complex wavelet based detectors were observed to suffer from two defects: a tendency to produce keypoints on straight edges at particular orientations and sensitivity to small translations of the image. We introduced a new corner energy function based on the Same Level Product (SLP) transform. This function performed well compared to previous ones, combining competitive edge rejection and positional stability properties. Secondly, we investigated the effect of changing the resolution at which the energy function is sampled. We used the undecimated DTCWT to calculate energy maps at the same resolution as the original images. This revealed the presence of fine details which could not be accurately interpolated from an energy map at the standard resolution. As a result, doubling the resolution of the map along each axis significantly improved both the reliability and posi-tional accuracy of detections. However, calculating the map using interpolated coefficients resulted in artefacts introduced by inaccuracies in the interpolation. We therefore proposed a modification to the standard DTCWT structure which doubles its output resolution for a modest computational cost. Thirdly, we developed a random forest based quantiser which operates on complex wavelet polar matching descriptors, with optional rotational invariance. Trees were evaluated on the basis of how consistently they quantised features into the same bins, and several examples of each feature were obtained by means of tracking. We found that the trees produced the most consistent quantisations when they were trained with a second set of tracked keypoints. Detecting keypoints using the the higher resolution energy maps also resulted in more consistent quantiser outputs, indicating the importance of the choice of detector on quantiser performance. Finally, we introduced a fast implementation of the DTCWT, keypoint detection and descriptor extraction algorithms for OpenCL-capable GPUs. Several aspects were optimised to enable it to run more efficiently on modern hardware, allowing it to process HD footage in faster than real time. This particularly aided the development of the detector algorithms by permitting interactive exploration of their failure modes using a live camera feed.

Made available in DSpace on 2016-03-15T19:37:55Z (GMT). No. of bitstreams: 1 RICARDO RIBANI.pdf: 13475262 bytes, checksum: 1e8f0a623498d0aa2fda9f44449b7325 (MD5) Previous issue date: 2015-02-11 / Fundo Mackenzie de Pesquisa / With the growing amount of multimedia information, the recommender systems have become more present in digital systems. Together with the growth of the internet, more and more people have access to large multimedia collections and consequently the user is often in doubt situations when making a choice. In order to help the user to make their own choices, this research presents a study around the content-based recommender systems applied to art paintings. Here are included approaches on image retrieval algorithms, computer vision and artificial intelligence concepts such as techniques for pattern recognition. One of the goals of this research was the creation of a software for mobile phones, applied to an art paintings database. The application uses an interface developed for mobile phones, where the user can point the phone s camera to a painting and based on this painting the system generates a recommendation of another painting in the same database, considering some parameters such as style, genre or color. / Os sistemas de recomendações estão cada dia mais presentes no meio digital. Com a crescente quantidade de informações e a popularização da internet, cada vez mais as pessoas tem acesso a grandes acervos multimídia. Com isso, consequentemente o usuário se encontra muitas vezes em situações de dúvida ao fazer uma escolha. Com o objetivo de auxiliar o usuário a fazer suas escolhas, o presente trabalho apresenta um estudo em torno dos sistemas de recomendações baseados em conteúdo de imagens. Este estudo engloba uma abordagem a respeito de algoritmos de recuperação de imagens, além da aplicação de conceitos de visão computacional e inteligência artificial, como técnicas para reconhecimento de padrões. Além do estudo teórico, este trabalho teve como objetivo a criação de um sistema computacional aplicado a um banco de dados de imagens de obras de arte. Uma aplicação que utiliza uma interface desenvolvida para telefones celulares, no qual o usuário pode capturar a imagem de uma obra através da câmera do celular e baseado nessa obra o sistema gera uma recomendação de outra dentro do mesmo banco de dados, considerando parâmetros configuráveis como estilo, gênero ou cores.

Soutenue par les progrès récents et rapides des techniques d'acquisition 3D, la reconnaissance d'objets 3D a suscité de nombreux efforts de recherche durant ces dernières années. Cependant, il reste à résoudre dans ce domaine plusieurs problématiques liées à la grande quantité d'information, à l'invariance à l'échelle et à l'angle de vue, aux occlusions et à la robustesse au bruit.Dans ce contexte, notre objectif est de reconnaitre un objet 3D isolé donné dans une vue requête, à partir d'une base d'apprentissage contenant quelques vues de cet objet. Notre idée est de formuler une méthodologie locale qui combine des aspects d'approches existantes et apporte une amélioration sur la performance de la reconnaissance.Nous avons opté pour une méthode par points d'intérêt (PIs) fondée sur des mesures de la variation locale de la forme. Notre sélection de points saillants est basée sur la combinaison de deux espaces de classification de surfaces : l'espace SC (indice de forme- intensité de courbure), et l'espace HK (courbure moyenne-courbure gaussienne).Dans la phase de description de l'ensemble des points extraits, nous proposons une signature d'histogrammes, qui joint une information sur la relation entre la normale du point référence et les normales des points voisins, avec une information sur les valeurs de l'indice de forme de ce voisinage. Les expérimentations menées ont permis d'évaluer quantitativement la stabilité et la robustesse de ces nouveaux détecteurs et descripteurs.Finalement nous évaluons, sur plusieurs bases publiques d'objets 3D, le taux de reconnaissance atteint par notre méthode, qui montre des performances supérieures aux techniques existantes. / There has been strong research interest in 3D object recognition over the last decade, due to the promising reliability of the 3D acquisition techniques. 3D recognition, however, conveys several issues related to the amount of information, to scales and viewpoints variation, to occlusions and to noise.In this context, our objective is to recognize an isolated object given in a request view, from a training database containing some views of this object. Our idea is to propose a local method that combines some existent approaches in order to improve recognition performance.We opted for an interest points (IPs) method based on local shape variation measures. Our selection of salient points is done by the combination of two surface classification spaces: the SC space (Shape Index-Curvedness), and the HK space (Mean curvature- Gaussian curvature).In description phase of the extracted set of points, we propose a histogram based signature, in which we join information about the relationship between the reference point normal and normals of its neighbors, with information about the shape index values of this neighborhood. Performed experiments allowed us to evaluate quantitatively the stability and the robustness of the new proposed detectors and descriptors.Finally we evaluate, on several public 3D objects databases, the recognition rate attained by our method, which outperforms existing techniques on same databases.

This paper describes the various components and phases of the search and recognition of facial expressions of unknown persons. They are presented here as well as possible solutions and methods of addressing each phase of the project. My master’s thesis is designed to recognize facial expressions of unknown persons. For this thesis, I was lent industrial video camera, computer, and place in a laboratory. Furthermore, we introduce the color spaces and their use. From the lead representatives selects the most appropriate assistance for the use of Matlab and the proposed algorithm. After finding a suitable color space segments skin color in the image. The skin, however, surrounds the entire body and so need to be found, the separated parts of the image representing the color of skin, a face. Once you find a face is needed to find relevant points for the identification subsequent deformation to definition of facial expressions. We define here the actual muscle movements in different expressions.

Keypoint detection and description is the first step of homography and essential matrix estimation, which in turn is used in Visual Odometry and Visual SLAM. This work explores the effect (in terms of speed and accuracy) of using different deep learning architectures for such keypoints. The fully convolutional networks — with heads for both the detector and descriptor — are trained through an existing self-supervised method, where correspondences are obtained through known randomly sampled homographies. A new strategy for choosing negative correspondences for the descriptor loss is presented, which enables more flexibility in the architecture design. The new strategy turns out to be essential as it enables networks that outperform the learnt baseline at no cost in inference time. Varying the model size leads to a trade-off in speed and accuracy, and while all models outperform ORB in homography estimation, only the larger models approach SIFT’s performance; performing about 1-7% worse. Training for longer and with additional types of data might give the push needed to outperform SIFT. While the smallest models are 3× faster and use 50× fewer parameters than the learnt baseline, they still require 3× as much time as SIFT while performing about 10-30% worse. However, there is still room for improvement through optimization methods that go beyond architecture modification, e.g. quantization, which might make the method faster than SIFT. / Nyckelpunkts-detektion och deskriptor-skapande är det första steget av homografi och essentiell matris estimering, vilket i sin tur används inom Visuell Odometri och Visuell SLAM. Det här arbetet utforskar effekten (i form av snabbhet och exakthet) av användandet av olika djupinlärnings-arkitekturer för sådana nyckelpunkter. De hel-faltade nätverken – med huvuden för både detektorn och deskriptorn – tränas genom en existerande själv-handledd metod, där korrespondenser fås genom kända slumpmässigt valda homografier. En ny strategi för valet av negativa korrespondenser för deskriptorns träning presenteras, vilket möjliggör mer flexibilitet i designen av arkitektur. Den nya strategin visar sig vara väsentlig då den möjliggör nätverk som presterar bättre än den lärda baslinjen utan någon kostnad i inferenstid. Varieringen av modellstorleken leder till en kompromiss mellan snabbhet och exakthet, och medan alla modellerna presterar bättre än ORB i homografi-estimering, så är det endast de större modellerna som närmar sig SIFTs prestanda; där de presterar 1-7% sämre. Att träna längre och med ytterligare typer av data kanske ger tillräcklig förbättring för att prestera bättre än SIFT. Även fast de minsta modellerna är 3× snabbare och använder 50× färre parametrar än den lärda baslinjen, så kräver de fortfarande 3× så mycket tid som SIFT medan de presterar runt 10-30% sämre. Men det finns fortfarande utrymme för förbättring genom optimeringsmetoder som övergränsar ändringar av arkitekturen, som till exempel kvantisering, vilket skulle kunna göra metoden snabbare än SIFT.

La vidéosurveillance a pour objectif principal de protéger les personnes et les biens en détectant tout comportement anormal. Ceci ne serait possible sans la détection de mouvement dans l’image. Ce processus complexe se base le plus souvent sur une opération de soustraction de l’arrière-plan statique d’une scène sur l’image. Mais il se trouve qu’en vidéosurveillance, des caméras sont souvent en mouvement, engendrant ainsi, un changement significatif de l’arrière-plan; la soustraction de l’arrière-plan devient alors problématique. Nous proposons dans ce travail, une méthode de détection de mouvement et particulièrement de chutes qui s’affranchit de la soustraction de l’arrière-plan et exploite la rotation de la caméra dans la détection du mouvement en utilisant le calcul homographique. Nos résultats sur des données synthétiques et réelles démontrent la faisabilité de cette approche. / The main objective of video surveillance is to protect persons and property by detecting any abnormal behavior. This is not possible without detecting motion in the image. This process is often based on the concept of subtraction of the scene background. However in video tracking, the cameras are themselves often in motion, causing a significant change of the background. So, background subtraction techniques become problematic. We propose in this work a motion detection approach, with the example application of fall detection. This approach is free of background subtraction for a rotating surveillance camera. The method uses the camera rotation to detect motion by using homographic calculation. Our results on synthetic and real video sequences demonstrate the feasibility of this approach.

La vidéosurveillance a pour objectif principal de protéger les personnes et les biens en détectant tout comportement anormal. Ceci ne serait possible sans la détection de mouvement dans l’image. Ce processus complexe se base le plus souvent sur une opération de soustraction de l’arrière-plan statique d’une scène sur l’image. Mais il se trouve qu’en vidéosurveillance, des caméras sont souvent en mouvement, engendrant ainsi, un changement significatif de l’arrière-plan; la soustraction de l’arrière-plan devient alors problématique. Nous proposons dans ce travail, une méthode de détection de mouvement et particulièrement de chutes qui s’affranchit de la soustraction de l’arrière-plan et exploite la rotation de la caméra dans la détection du mouvement en utilisant le calcul homographique. Nos résultats sur des données synthétiques et réelles démontrent la faisabilité de cette approche. / The main objective of video surveillance is to protect persons and property by detecting any abnormal behavior. This is not possible without detecting motion in the image. This process is often based on the concept of subtraction of the scene background. However in video tracking, the cameras are themselves often in motion, causing a significant change of the background. So, background subtraction techniques become problematic. We propose in this work a motion detection approach, with the example application of fall detection. This approach is free of background subtraction for a rotating surveillance camera. The method uses the camera rotation to detect motion by using homographic calculation. Our results on synthetic and real video sequences demonstrate the feasibility of this approach.

The evolution of technology has led the research into new emerging wearable devices such as the Smart Glasses. This technology provides with new visualization techniques. Augmented Reality is an advanced technology that could significantly ease the execution of much complex operations. Augmented Reality is a combination of both Virtual and Actual Reality, making accessible to the user new tools to safeguard in the transfer of knowledge in several environments and for several processes. This thesis explores the development of an android based image recognition application. The feature point detectors and descriptors are used as they can deal great with the correspondence problems. The selection of best image recognition technique on the smart glasses is chosen based on the time taken to retrieve the results and the amount of power consumed in the process. As the smart glasses are equipped with the limited resources, the selected approach should use low computation on it by making the device operations uninterruptable. The effective and efficient method for detection and recognition of the safety signs from images is selected. The ubiquitous SIFT and SURF feature detectors consume more time and are computationally complex and require very high-level hardware components for processing. The binary descriptors are taken into account as they are light weight and can support low power devices in a much effective style. A comparative analysis is being done on the working of binary descriptors like BRIEF, ORB, AKAZE, FREAK, etc., on the smart glasses based on their performance and the requirements. ORB is the most efficient among the binary descriptors and has been more effective for the smart glasses in terms of time measurements and low power consumption.

Visual surveillance in the maritime domain has been explored for more than a decade. Although it has produced a number of working systems and resulted in a mature technology, surveillance has been restricted to the port facilities or areas close to the coastline assuming a fixed-camera scenario. This dissertation presents several contributions in the domain of maritime surveillance. First, a novel algorithm for open-sea visual maritime surveillance is introduced. We explore a challenging situation with a camera mounted on a buoy or other floating platform. The developed algorithm detects, localizes, and tracks ships in the field of view of the camera. Specifically, our method is uniquely designed to handle a rapidly moving camera. Its performance is robust in the presence of a random relatively-large camera motion. In the context of ship detection, a new horizon detection scheme for a complex maritime domain is also developed. Second, the performance of the ship detection algorithm is evaluated on a dataset of 55,000 images. Accuracy of detection of up to 88% of ships is achieved. Lastly, we consider the topic of detection of the vanishing line of the ocean surface plane as a way to estimate the horizon in difficult situations. This allows extension of the ship-detection algorithm to beyond open-sea scenarios.