Σκοπός της παρούσας εργασίας είναι η μελέτη, η ανάπτυξη καθώς και η μερική εφαρμογή κάποιων μεθόδων για την ανίχνευση θέσης κάποιου οχήματος. Ιδιαίτερη βάση δόθηκε στη μελέτη και την ανάλυση της οπτικής ροής που θεωρείται βασική συγκριτικά με τις υπόλοιπες μεθόδους.Τέλος αναλύεται και μια μέθοδος κατάτμησης εικόνων. / The goal of this thesis is to study, develop and implement some methods of car detection. Particular emphasis is given to the analysis of optical flow, which is considered to be critical compared to other methods. Finally an analysis of a method for image segmentation is being developed.

En modern grafikprocessor är oerhört kraftfull och har en prestanda som potentiellt sett är många gånger högre än för en modern mikroprocessor. I takt med att grafikprocessorn blivit alltmer programmerbar har det blivit möjligt att använda den för beräkningstunga tillämpningar utanför dess normala användningsområde. Inom det här arbetet utreds vilka möjligheter och begränsningar som uppstår vid användandet av grafikprocessorer för generell programmering. Arbetet inriktas främst mot signal- och bildbehandlingstillämpningar men mycket av principerna är tillämpliga även inom andra områden. Ett ramverk för bildbehandling implementeras och några algoritmer inom bildanalys realiseras och utvärderas, bland annat stereoseende och beräkning av optiskt flöde. Resultaten visar på att vissa tillämpningar kan uppvisa en avsevärd prestandaökning i en grafikprocessor jämfört med i en mikroprocessor men att andra tillämpningar kan vara ineffektiva eller mycket svåra att implementera. / The modern graphical processing unit, GPU, is an extremely powerful unit, potentially many times more powerful than a modern microprocessor. Due to its increasing programmability it has recently become possible to use it in computation intensive applications outside its normal usage. This work investigates the possibilities and limitations of general purpose programming on GPUs. The work mainly concentrates on signal and image processing although much of the principles are applicable to other areas as well. A framework for image processing on GPUs is implemented and a few computer vision algorithms are implemented and evaluated, among them stereo vision and optical flow. The results show that some applications can gain a substantial speedup when implemented correctly in the GPU but others can be inefficent or extremly hard to implement.

There has been a rapid progress of the graphics processor the last years, much because of the demands from computer games on speed and image quality. Because of the graphics processor’s special architecture it is much faster at solving parallel problems than the normal processor. Due to its increasing programmability it is possible to use it for other tasks than it was originally designed for. Even though graphics processors have been programmable for some time, it has been quite difficult to learn how to use them. CUDA enables the programmer to use C-code, with a few extensions, to program NVIDIA’s graphics processor and completely skip the traditional programming models. This thesis investigates if the graphics processor can be used for calculations without knowledge of how the hardware mechanisms work. An image processing algorithm calculating the optical flow has been implemented. The result shows that it is rather easy to implement programs using CUDA, but some knowledge of how the graphics processor works is required to achieve high performance.

O presente trabalho reúne elementos para a análise dos principais problemas e dificuldades inerentes à implementação de um dispositivo para a aferição de velocímetros de veículos automotores, de forma a atender a legislação prevista no código nacional de Trânsito, determinada pela Portaria n.º 115 de 1998, que define erros máximos permitidos para velocímetros de ± 5km/h para velocidades de até 100km/h e de ± 5% para velocidades acima de 100km/h. Inicialmente são apresentados e comparados os principais métodos de medição de velocidade abordados na literatura. A partir desta análise, é implementado um método inovador decorrente do efeito de escorrimento ou blur, onde, através do estudo das características e regularidades contidas em uma única imagem borrada, será determinada a velocidade da superfície em análise. A partir de uma bancada que simula o solo em movimento e, utilizando-se uma câmera CCD (Charge Coupled Device) e um frame grabber, fez-se a aquisição de imagens da superfície asfáltica em movimento. A informação de velocidade pôde então ser determinada, através da análise das regularidades contidas na imagem dinâmica devido ao efeito de escorrimento ou efeito blur. Obteve-se resultados suficientes para a aferição de velocímetros, com erros máximos abaixo de 5%. A técnica desenvolvida e avaliada na prática através de uma bancada que simula o asfalto em movimento, demonstrou uma precisão de 0,8% numa faixa de velocidades de 0 a 20km/h , de 1,5% numa faixa de velocidades de 20 a 60km/h e de 2,5% numa faixa de velocidades de 60km/h a 80km/h. Finalmente foram investigados os fatores preponderantes que limitaram os erros nesta ordem de grandeza. / This work gathers elements for the study and analysis of the main problems and difficulties inherents to the implementation of a device to the gauging of a speedometer of automotive vehicles, in order to meet the foreseen legislation described by the governmental decree nº 115 from 1998, which defines the maximum allowed error for speedometers of ± 5km/h for speeds till 100km/h and ± 5% for speeds above 100km/h. From this study, a new method, based on the blur effect, is proposed. In the method, the speed of the target surface is determined by analyzing the characteristics and regularities contained in a single blurred image. Starting from a device that simulates the soil movement and, using a CCD (Charge Coupled Device) camera and a frame grabber, the acquisition of images of the moving asphalt surface was done. The information of speed then could be determined, through the study of the regularities contained in the dynamic image due to the blur effect. The necessary results for the gauging have been achieved with success, with precision below 5%. The developed and evaluated technique in practice through a device that simulates the asphalt in movement has demonstrated a precision of 0,8% in a range of speeds from 0 to 20km/h, 1,5% in a range of speeds from 20km/h to 60km/h and 2,5% in a range of speeds from 60km/h to 80km/h. Finally, it was investigated the preponderant factors which have limited the errors in this order of greatness.

Orientador: Siome Klein Goldenstein / Dissertação (mestrado) - Universidade Estadual de Campinas, Instituto de Computação / Made available in DSpace on 2018-08-13T11:06:36Z (GMT). No. of bitstreams: 1 Hauagge_DanielCabrini_M.pdf: 29273846 bytes, checksum: d697300202f6081a1441b4748b9a711c (MD5) Previous issue date: 2008 / Resumo: Apresentamos nesta dissertação quatro algoritmos voltados para a classicação automatizada de frutas. A subtração de fundo baseada na distância de Mahalanobis. O rastreamento das frutas em uma esteira usando a subtração de fundo, casamento de padrões e fluxo óptico. A reconstrução tridimensional da fruta a partir de imagens dela na esteira, onde recuperamos a posição da câmera com relação a fruta usando fluxo óptico e uma estimativa grosseira do movimento da fruta. A forma da fruta é obtida a partir das silhuetas reprojetadas no espaço tridimensional usando duas abordagens diferentes. Finalmente, a localização do pedúnculo e cálice a partir do eixo de simetria da reconstrução tridimensional. Realizamos testes com os quatro algoritmos. Obtivemos bons resultados com os dois primeiros. Para a reconstrução tridimensional verificamos bons resultados para algumas etapas do processo (fluxo óptico, estimativa inicial e otimização não-linear do movimento de câmera). Resultados fracos foram obtidos para a reprojeção das silhuetas usando os dois métodos. Analisamos as causas dos erros e propomos métodos que poderiam ser usados para melhorá-los. Os resultados da localização do pedúnculo e cálice foram insatisfatórios mas acreditamos que melhorariam se obtivéssemos uma reconstrução mais precisa. Também criamos um sistema de captura que reproduz as condições dentro de um sistema comercial de classificação. Com este aparato construímos quatro grandes bases de dados com aproximadamente 3000 frutas, 35 imagens de cada uma, contendo quatro variedades de maçã. Outras 6 bases menores foram criadas. / Abstract: We present in this dissertation four algorithms targeted at the automated classification of fruits. Background subtraction based on Mahalanobis distance. Fruit tracking on a conveyor belt using background subtraction, pattern matching and optical flow. The 3D reconstruction of the fruit from its images on the conveyor belt, where we recover the camera position, with respect to the fruit, using optical flow and a rough estimate of fruit motion. The fruit's shape is recovered from the silhouette re-projected into 3D space using two different approaches. Finally, the location of the stem and calyx based on the symmetry axis of the 3D reconstruction. We also present the results of tests conducted with the four algorithms. We obtained good results with the first two. For the three-dimensional reconstruction we obtained good results with some of the intermediary steps (optical flow, initial estimate and nonlinear re_nement of camera motion). Poor results were obtained for the re-projection of the silhouette's, using two approaches. We analyze the causes of these difficulties and suggest approaches that could improve them. The localization of stem and calyx was compromised by the poor 3D reconstruction so we believe that it will improve once we address the problems with the reconstruction algorithm. We created an image capturing system that reproduces the conditions inside a commercial grading machine. With this device we acquired four big data sets with approximately 3000 apples, 35 images of each, comprising four varieties. Another 6 smaller data-sets were also created. / Mestrado / Visão Computacional / Mestre em Ciência da Computação

Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-06-06T12:10:59Z No. of bitstreams: 1 luizmauriliodasilvamaciel.pdf: 11082986 bytes, checksum: ce05919ae5e2bd37882838427a25f4a3 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-06-06T12:17:02Z (GMT) No. of bitstreams: 1 luizmauriliodasilvamaciel.pdf: 11082986 bytes, checksum: ce05919ae5e2bd37882838427a25f4a3 (MD5) / Made available in DSpace on 2017-06-06T12:17:02Z (GMT). No. of bitstreams: 1 luizmauriliodasilvamaciel.pdf: 11082986 bytes, checksum: ce05919ae5e2bd37882838427a25f4a3 (MD5) Previous issue date: 2014-02-27 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Identificar movimento em vídeos é uma tarefa fundamental a fim de analisar a sua infor-mação semântica. Uma das principais ferramentas para a identificação de movimento é o fluxo óptico, o qual estima a projeção da velocidade 3D dos objetos sobre o plano da câmera. Neste trabalho é proposto um método diferencial de fluxo óptico baseado na equação da onda. O fluxo óptico é calculado através da minimização de funcional de energia composto por dois termos: um termo de constância de brilho e um termo de energia da onda. O fluxo é então determinado através da resolução iterativa de um sistema de equações lineares. O desacoplamento entre os pixels na solução garante convergência rápida e torna o método adequado para a paralelização. No entanto, nossa abordagem não converge para todos os pontos de imagem, sendo apresentadas as suas condições de convergência. O fluxo proposto é aplicado no problema de reconhecimento de ação através da criação de um descritor global de vídeo baseado em histogramas de fluxo óptico (HOF). Apesar da sua esparsidade, o método proposto supera as abordagens clássicas. Também são avaliadas medidas de erro de fluxo óptico para algumas sequências de imagens conhecidas. Os erros encontrados são similares para o nosso método e as abordagens clássicas de fluxo óptico. / Identification of motion in videos is a fundamental task to analyse their semantic informa-tion. One of the main tools for motion identification is the optical flow, which estimates the projection of the 3D velocity of the objects onto the plane of the camera. In this work, we propose a differential optical flow method based on the wave equation. The optical flow is computed by minimizing a functional energy composed by two terms: brightness constancy and energy of the wave. The flow is then determined by solving iteratively a system of linear equations. The decoupling of the pixels in the solution ensures quick convergence and makes the method suitable for parallelization. However, our approach does not converge for all the image points and we present its convergence conditions. We apply our optical flow in the action recognition problem by creating a global video descriptor based on histograms of optical flow (HOF). Despite its sparsity, our method outperforms the classical approaches. We also evaluate optical flow error measures for some known image sequences. The errors found are similar for our method and the classical optical flow approaches.

This master thesis investigates the difficulties of constructing a depth map using one low resolution grayscale camera mounted in the front of a car. The goal is to produce a depth map in real-time to assist other algorithms in the safety system of a car. This has been shown to be difficult using the evaluated combination of camera position and choice of algorithms. The main problem is to estimate an accurate optical flow. Another problem is to handle moving objects. The conclusion is that the implementations, mainly triangulation of corresponding points tracked using a Lucas Kanade tracker, provide information of too poor quality to be useful for the safety system of a car. / I detta examensarbete undersöks svårigheterna kring att skapa en djupbild från att endast använda en lågupplöst gråskalekamera monterad framtill i en bil. Målet är att producera en djupbild i realtid som kan nyttjas i andra delar av bilens säkerhetssystem. Detta har visat sig vara svårt att lösa med den undersökta kombinationen av kameraplacering och val av algoritmer. Det huvudsakliga problemet är att räkna ut ett noggrant optiskt flöde. Andra problem härrör från objekt som rör på sig. Slutsatsen är att implementationerna, mestadels triangulering av korresponderande punktpar som följts med hjälp av en Lucas Kanade-följare, ger resultat av för dålig kvalitet för att vara till nytta för bilens säkerhetssystem.

L’objectif de cette thèse est l’étude de l’évolution de champ scalaire transporté par un écoulement à partir d’une séquence d’images temporelles. L’estimation du champ de vitesse d’un écoulement turbulent est d’une importance majeure pour mieux comprendre le phénomène physique. Jusqu’à présent, le problème de la turbulence est généralement ignoré dans l’équation de mouvement des méthodes existantes. Les images contiennent une information discrète correspondant à la taille du pixel. Selon le niveau de turbulence de l’écoulement, les résolutions des pixels et du temps peuvent devenir trop grandes pour négliger l’effet des petites échelles (sous-pixel) sur le champ de vitesse. Nous proposons pour cela, une équation de mouvement définie par l’équation de transport de concentration filtrée pour laquelle un modèle classique de viscosité turbulente sous-maille est introduit afin de tenir compte de cet effet. Nous utilisons pour formuler le problème, une approche Markovienne. Une méthode de multirésolution par décomposition pyramidale, sans transformation d’image intermédiaire au cours du processus, est proposée. Cela permet de diminuer le nombre d’opérations sur les images. La méthode d’optimisation utilisée, couplée avec une approche multigrille, permet d’obtenir le champ de vitesse réel optimal. Notre approche est testée sur des séquences d’images synthétiques et réelles (expérience PIV et tempête de sable à partir d’image de télédétection) avec des nombres de Reynolds élevés. Les comparaisons avec des approches existantes sont très prometteuses. / The objective of this thesis is to study the evolution of scalar field carried by a flow from a temporal image sequence. The estimation of the velocity field of turbulent flow is of major importance for understanding the physical phenomenon. Up to now the problem of turbulence is generally ignored in the flow equation of existing methods. An information given by image is discrete at pixel size. Depending on the turbulent rate of the flow, pixel and time resolutions may become too large to neglect the effect of sub-pixel small-scales on the pixel velocity field. For this, we propose a flow equation defined by a filtered concentration transport equation where a classic turbulent sub-grid eddy viscosity model is introduced in order to account for this effect. To formulate the problem, we use a Markovian approach. An unwarping multiresolution by pyramidal decomposition is proposed which reduces the number of operations on images. The optimization coupled with a multigrid approach allows to estimate the optimal 2D real velocity field. Our approach is tested on synthetic andreal image sequences (PIV laboratory experiment and remote sensing data of dust storm event) with high Reynolds number. Comparisons with existing approaches are very promising.

Nous nous intéressons dans cette thèse au problème de l'estimation dense du mouvement dans des séquences d'images, également désigné sous le terme de flot optique. Les approches usuelles exploitent une paramétrisation locale ou une régularisation globale du champ de déplacement. Nous explorons plusieurs façons de combiner ces deux stratégies, pour surmonter leurs limitations respectives. Nous nous plaçons dans un premier temps dans un cadre variationnel global, et considérons un filtrage local du terme de données. Nous proposons un filtrage spatialement adaptatif, optimisé conjointement au mouvement, pour empêcher le sur-lissage induit par le filtrage spatialement constant. Dans une seconde partie, nous proposons un cadre générique d'agrégation pour l'estimation du flot optique. Sous sa forme générale, il consiste en une estimation locale de candidats de mouvements, suivie de leur combinaison à l'étape d'agrégation avec un modèle global. Ce schéma permet une estimation efficace des grands déplacements et des discontinuités de mouvement. Nous développons également une méthode générique de gestion des occultations. Notre méthode est validée par une analyse expérimentale conséquente sur des bases de données de référence en vision par ordinateur. Nous démontrons la supériorité de notre méthode par rapport à l'état de l'art sur les séquences présentant de grands déplacements. La dernière partie de la thèse est consacrée à l'adaptation des approches précédentes à des problématiques d'imagerie biologique. Les changements locaux importants d'intensité observés en imagerie de fluorescence sont estimés et compensé par une adaptation de notre schéma d'agrégation. Nous proposons également une méthode variationnelle avec filtrage local dédiée au cas de mouvements diffusifs de particules. / This thesis is concerned with dense motion estimation in image sequences, also known as optical flow. Usual approaches exploit either local parametrization or global regularization of the motion field. We explore several ways to combine these two strategies, to overcome their respective limitations. We first address the problem in a global variational framework, and consider local filtering of the data term. We design a spatially adaptive filtering optimized jointly with motion, to prevent over-smoothing induced by the spatially constant approach. In a second part, we propose a generic two-step aggregation framework for optical flow estimation. The most general form is a local computation of motion candidates, combined in the aggregation step through a global model. Large displacements and motion discontinuities are efficiently recovered with this scheme. We also develop a generic exemplar-based occlusion handling to deal with large displacements. Our method is validated with extensive experiments in computer vision benchmarks. We demonstrate the superiority of our method over state-of-the-art on sequences with large displacements. Finally, we adapt the previous methods to biological imaging issues. Estimation and compensation of large local intensity changes frequently occurring in fluorescence imaging are efficiently estimated and compensated with an adaptation of our aggregation framework. We also propose a variational method with local filtering dedicated to the case of diffusive motion of particles.

Doctor en Ciencias de la Ingeniería, Mención Fluidodinámica / En este trabajo se presenta el problema del transporte de calor y masa para un sistema compuesto de dos fases fluidas en un medio poroso, el cual puede ser relevante en el uso de CO2 como fluido de trabajo en reservorios geotermales. El medio poroso fue modelado usando una celda Hele-Shaw, la cual es ampliamente usada para la visualización del transporte de escalares. Los objetivos de este trabajo son (a) investigar teórica y numéricamente las leyes de escalamiento que gobiernan los procesos de mezcla en un medio poroso, y (b) visualizar los procesos de transporte usando celdas Hele-Shaw y avanzadas técnicas de segmentación de imágenes. El modelo derivado es una extensión de la ecuación de Polubarinova-Kochina, la cual es válida para regímenes de flujos altos en medios porosos. El transporte de calor y masa en un medio poroso es gobernado por el número de Nusselt Nu' y la tasa de disipación media escalar h"'i_ = `mix/L , donde `mix es la longitud de mezcla y L es la longitud horizontal de la celda . La principal contribución de este trabajo es la demostración de la existencia de los escalamientos hNu'i_ _ Ran(_) ' y h"'i_ _ Ram(_) ' , respectivamente, donde Ra' es el número de Rayleigh sujeto a la definición del escalar ' , el cual puede ser temperatura o concentración, y _ es el cociente entre el espaciamiento de la celda y su altura. Ambas cantidades se relacionan a través del modelo hNu'i_ = _(_2Ra', _)Ra' h"'i_ , donde _ contiene información de los efectos de difusión lateral de mezcla. Esta contribución extiende los resultados más recientes publicados en la literatura. Dada la importancia del parámetro adimensional _ en la escala de laboratorio, se realizó un detallado análisis lineal de la convección termal y la inestabilidad de Rayleigh-Taylor, incorporando efectos de tensión interfacial ya que el CO2 es un fluido parcialmente miscible con el agua. Los resultados obtenidos usando métodos asintóticos corrigen los análisis descritos en la literatura, así como también entregan nuevas evidencias de que el inicio de la inestabilidad de Rayleigh-Taylor depende completamente de los efectos interfaciales de la mezcla. Finalmente, para visualizar el transporte escalar, se realizaron experimentos de convección termal y mezcla por contraste de densidad. Se usaron técnicas de atenuación de luz y Schlieren sintético, además de métodos de segmentación de imágenes basados en principios variacionales, los cuales permitieron obtener interesantes resultados visuales del proceso de transporte. Se demostró que la aplicación del método optical flow permite reconstruir el mapa de temperaturas en celdas Hele-Shaw con una mejor resolución de imagen que el algoritmo PIV digital, obteniendo resultados acordes a lo esperando en sistemas geotermales sedimentarios. Además, se muestra experimentalmente que el uso de segmentación multifase es ideal para calcular propiedades físicas del proceso de mezcla, además de cantidades relacionadas con el transporte escalar, sin conocer a priori los valores de cantidades físicas tales como la densidad y velocidad de flujo