Thesis (MScEng (Mechanical and Mechatronic Engineering))--University of Stellenbosch, 2008. / A rapid measurement system is implemented using two digital video cameras, presenting a faster and less expensive solution to certain metrology problems. The cameras are calibrated from one stereo image-pair of a 3D calibration grid that allows an immediate assessment of the achievable metric accuracy of the system. Three different methods, using either laser tracking or structured light patterns, were developed and employed to solve the coordinate extraction and correspondence matching problems. Different image processing techniques were used to speed up the entire measurement process. All software development was accomplished using only freely distributed software packages. The system achieves calibration in less than a minute and accumulates point correspondences at 12 frames per second. Accuracies of greater than 0.4 mm are achieved for a 235 x 190 x 95 mm measurement volume using a single pair of images with 640 x 480 pixel resolution each.

In our aging society, the design and implementation of a high-performance autonomous distributed vision information system for autonomous physical services become ever more important. In line with this development, the proposed Intelligent Vision Agent System, IVAS, is able to automatically detect and identify a target for a specific task by surveying a human activities space. The main subject of this thesis is the optimal configuration of a sensor system meant to capture the target objects and their environment within certain required specifications. The thesis thus discusses how a discrete sensor causes a depth spatial quantisation uncertainty, which significantly contributes to the 3D depth reconstruction accuracy. For a sensor stereo pair, the quantisation uncertainty is represented by the intervals between the iso-disparity surfaces. A mathematical geometry model is then proposed to analyse the iso-disparity surfaces and optimise the sensors’ configurations according to the required constrains. The thesis also introduces the dithering algorithm which significantly reduces the depth reconstruction uncertainty. This algorithm assures high depth reconstruction accuracy from a few images captured by low-resolution sensors. To ensure the visibility needed for surveillance, tracking, and 3D reconstruction, the thesis introduces constraints of the target space, the stereo pair characteristics, and the depth reconstruction accuracy. The target space, the space in which human activity takes place, is modelled as a tetrahedron, and a field of view in spherical coordinates is proposed. The minimum number of stereo pairs necessary to cover the entire target space and the arrangement of the stereo pairs’ movement is optimised through integer linear programming. In order to better understand human behaviour and perception, the proposed adaptive measurement method makes use of a fuzzily defined variable, FDV. The FDV approach enables an estimation of a quality index based on qualitative and quantitative factors. The suggested method uses a neural network as a tool that contains a learning function that allows the integration of the human factor into a quantitative quality index. The thesis consists of two parts, where Part I gives a brief overview of the applied theory and research methods used, and Part II contains the five papers included in the thesis.

The subject of this dissertation is the development of three-dimensional (3D) surface profilers for semiconductor back-end inspection. The value of this study is: 1) to provide a new phase-to-height relationship for Fourier Transform Profilometry (FTP) that is universal as it allows alternate FTP system architectures for a micrometer scale object measurement, and 2) to provide a new method for full field substrate warpage and ball grid array (BGA) coplanarity inspection using machine vision. The desire to increase electronic device performance has resulted in denser and smaller IC packaging. As the dimensions of the devices decrease, the requirements for substrate flatness and surface quality become critical in avoiding device failure. For a high yield production, there is an increasing demand in the requirement for the dimensional verification of height, which requires 3D inspection. Based on the current demands from the semiconductor industry, this dissertation addresses the development of fast in-line surface profilers for large volume IC package inspection. Specifically, this dissertation studies two noncontact surface profilers. The first profiler is based on FTP for measuring the IC package front surface, the silicon die and the epoxy underfill profile. The second profiler is based on stereovision and it is intended for inspecting the BGA coplanarity and the substrate warpage. A geometrical shape based matching algorithm is also developed for finding point correspondences between IC package images. The FTP profiler provides a 1 σRMS error of about 4 μm for an IC package sample in an area of 14 mm x 6.5 mm with a 0.13 second data acquisition time. For evaluating the performance of the stereovision system, the linearity between our system and a confocal microscope is studied by measuring a particular IC sample with an area of 38 mm x 28.5 mm. The correlation coefficient is 0.965 and the 2σdifference in the two methods is 26.9 μm for the warpage measurement. For BGA coplanarity inspection the correlation coefficient is 0.952 and the 2difference is 31.2 μm. Data acquisition takes about 0.2 seconds for full field measurements.

As the requirement for more realistic 3D environments is pushed forward by the computer {graphics | movie | simulation | games} industry, attention turns away from the creation of purely synthetic, artist derived environments towards the use of real world captures from the 3D world in which we live. However, common 3D acquisition techniques, such as laser scanning and stereo capture, are realistically only 2.5D in nature - such that the backs and occluded portions of objects cannot be realised from a single uni-directional viewpoint. Although multi-directional capture has existed for sometime, this incurs additional temporal and computational cost with no existing guarantee that the resulting acquisition will be free of minor holes, missing surfaces and alike. Drawing inspiration from the study of human abilities in 3D visual completion, we consider the automated completion of these hidden or missing portions in 3D scenes originally acquired from 2.5D (or 3D) capture. We propose an approach based on the visual propagation of available scene knowledge from the known (visible) scene areas to these unknown (invisible) 3D regions (i.e. the completion of unknown volumes via visual propagation - the concept of volume completion). Our proposed approach uses a combination of global surface fitting, to derive an initial underlying geometric surface completion, together with a 3D extension of nonparametric texture synthesis in order to provide the propagation of localised structural 3D surface detail (i.e. surface relief). We further extend our technique both to the combined completion of 3D surface relief and colour and additionally to hierarchical surface completion that offers both improved structural results and computational efficiency gains over our initial non-hierarchical technique. To validate the success of these approaches we present the completion and extension of numerous 2.5D (and 3D) surface examples with relief ranging in natural, man-made, stochastic, regular and irregular forms. These results are evaluated both subjectively within our definition of plausible completion and quantitatively by statistical analysis in the geometric and colour domains.

Nous étudions dans cette thèse le problème de reconstruction 3D multi-vue à partir d’une séquence d’images au sol acquises dans des environnements urbains ainsi que la prise en compte d’a priori permettant la préservation de la structure sous-jacente de la géométrie 3D observée, ainsi que le passage à l’échelle de tels processus de reconstruction qui est intrinsèquement délicat dans le contexte de l’imagerie urbaine. Bien que ces deux axes aient été traités de manière extensive dans la littérature, les méthodes de reconstruction 3D structurée souffrent d’une complexité en temps de calculs restreignant significativement leur intérêt. D’autre part, les approches de reconstruction 3D large échelle produisent généralement une géométrie simplifiée, perdant ainsi des éléments de structures qui sont importants dans le contexte urbain. L’objectif de cette thèse est de concilier les avantages des approches de reconstruction 3D structurée à celles des méthodes rapides produisant une géométrie simplifiée. Pour ce faire, nous présentons “Patchwork Stereo”, un framework qui combine stéréoscopie photométrique utilisant une poignée d’images issues de points de vue éloignés, et un nuage de point épars. Notre méthode intègre une analyse simultanée 2D-3D réalisant une extraction robuste de plans 3D ainsi qu’une segmentation d’images top-down structurée et repose sur une optimisation par champs de Markov aléatoires. Les contributions présentées sont évaluées via des expériences quantitatives et qualitatives sur des données d’imagerie urbaine complexes illustrant des performances tant quant à la fidélité structurelle des reconstructions 3D que du passage à l’échelle / In this thesis, we address the problem of 3D reconstruction from a sequence of calibrated street-level photographs with a simultaneous focus on scalability and the use of structure priors in Multi-View Stereo (MVS).While both aspects have been studied broadly, existing scalable MVS approaches do not handle well the ubiquitous structural regularities, yet simple, of man-made environments. On the other hand, structure-aware 3D reconstruction methods are slow and scale poorly with the size of the input sequences and/or may even require additional restrictive information. The goal of this thesis is to reconcile scalability and structure awareness within common MVS grounds using soft, generic priors which encourage : (i) piecewise planarity, (ii) alignment of objects boundaries with image gradients and (iii) with vanishing directions (VDs), and (iv) objects co-planarity. To do so, we present the novel “Patchwork Stereo” framework which integrates photometric stereo from a handful of wide-baseline views and a sparse 3D point cloud combining robust 3D plane extraction and top-down image partitioning from a unified 2D-3D analysis in a principled Markov Random Field energy minimization. We evaluate our contributions quantitatively and qualitatively on challenging urban datasets and illustrate results which are at least on par with state-of-the-art methods in terms of geometric structure, but achieved in several orders of magnitude faster paving the way for photo-realistic city-scale modeling

Autonomous navigation in outdoor, unstructured environments is one of the major challenges presents in the robotics field. One of its applications, intelligent autonomous vehicles, has the potential to decrease the number of accidents on roads and highways, increase the efficiency of traffic on major cities and contribute to the mobility of the disabled and elderly. For a robot/vehicle to safely navigate, accurate detection of navigable areas is essential. In this work, we address the task of visual road detection where, given an image, the objective is to classify its pixels into road or non-road. Instead of trying to manually derive an analytical solution for the task, we have used machine learning (ML) to learn it from a set of manually created samples. We have applied both traditional (shallow) and deep ML models to the task. Our main contribution regarding traditional ML models is an efficient and versatile way to aggregate spatially distant features, effectively providing a spatial context to such models. As for deep learning models, we have proposed a new neural network architecture focused on processing time and a new neural network layer called the semi-global layer, which efficiently provides a global context for the model. All the proposed methodology has been evaluated in the Karlsruhe Institute of Technology (KIT) road detection benchmark, achieving, in all cases, competitive results. / A navegação autônoma em ambientes externos não estruturados é um dos maiores desafios no campo da robótica. Uma das suas aplicações, os veículos inteligentes autônomos, tem o potencial de diminuir o número de acidentes nas estradas e rodovias, aumentar a eficiência do tráfego nas grandes cidades e contribuir para melhoria da mobilidade de deficientes e idosos. Para que um robô/veículo navegue com segurança, uma detecção precisa de áreas navegáveis é essencial. Neste trabalho, abordamos a tarefa de detecção visual de ruas onde, dada uma imagem, o objetivo é classificar cada um de seus pixels em rua ou não-rua. Ao invés de tentar derivar manualmente uma solução analítica para a tarefa, usamos aprendizado de máquina (AM) para aprendê-la a partir de um conjunto de amostras criadas manualmente. Nós utilizamos tanto modelos tradicionais (superficiais) quanto modelos profundos para a tarefa. A nossa principal contribuição em relação aos modelos tradicionais é uma forma eficiente e versátil de agregar características espacialmente distantes, fornecendo efetivamente um contexto espacial para esses modelos. Quanto aos modelos de aprendizagem profunda, propusemos uma nova arquitetura de rede neural focada no tempo de processamento e uma nova camada de rede neural, chamada camada semi-global, que fornece eficientemente um contexto global ao modelo. Toda a metodologia proposta foi avaliada no benchmark de detecção de ruas do Instituto de Tecnologia de Karlsruhe, alcançando, em todos os casos, resultados competitivos.

A análise em tempo real de pares de imagens estereoscópicas para extração de características dimensionais da cena tem apresentado crescente interesse, possibilitando robusta navegação robótica e identificação de objetos em cenários dinâmicos. A presente dissertação propõe um método que emprega a análise pixel a pixel e observação de janelas, em pares de imagens estereoscópicas, para extração de denso mapa de disparidades. A arquitetura de processamento proposta é única em sua constituição, misturando elementos de processamento concorrente e seqüencial. O algoritmo estrutura-se em processamento pipeline, permitindo sua implementação em dispositivos de lógica programável e obtenção de resultados em tempo real. / Real-time analysis of stereo images for extraction of dimensional features has been focus of great interest, providing means for autonomous robot navigation and identification of objects in dynamic environments. This work describes a method based in pixel-to-pixel and windows based matching analysis, in stereo images, for constructing dense disparity maps. The proposed processing structure is unique, mixing concurrent and sequential elements. Pipelines structure is employed, targeting implementation in FPGA devices and enabling real-time results.

O objetivo deste trabalho é reconhecer gestos em tempo real apenas com o uso de câmeras, sem marcadores, roupas ou qualquer outro tipo de sensor. A montagem do ambiente de captura é simples, com apenas duas câmeras e um computador. O fundo deve ser estático, e contrastar com o usuário. A ausência de marcadores ou roupas especiais dificulta a tarefa de localizar os membros. A motivação desta tese é criar um ambiente de realidade virtual para treino de goleiros, que possibilite corrigir erros de movimentação, posicionamento e de escolha do método de defesa. A técnica desenvolvida pode ser aplicada para qualquer atividade que envolva gestos ou movimentos do corpo. O reconhecimento de gestos começa com a detecção da região da imagem onde se encontra o usuário. Nessa região, localizamos as regiões mais salientes como candidatas a extremidades do corpo, ou seja, mãos, pés e cabeça. As extremidades encontradas recebem um rótulo que indica a parte do corpo que deve representar. Um vetor com as coordenadas das extremidades é gerado. Para descobrir qual a pose do usuário, o vetor com as coordenadas das suas extremidades é classificado. O passo final é a classificação temporal, ou seja, o reconhecimento do gesto. A técnica desenvolvida é robusta, funcionando bem mesmo quando o sistema foi treinado com um usuário e aplicado a dados de outro. / Our aim in this work is to recognize gestures in real time with cameras, without markers or special clothes. The capture environment setup is simple, uses just two cameras and a computer. The background must be static, and its colors must be different the users. The absence of markers or special clothes difficults the location of the users limbs. The motivation of this thesis is to create a virtual reality environment for goalkeeper training, but the technique can be applied in any activity that involves gestures or body movements. The recognition of gestures starts with the background subtraction. From the foreground, we locate the more proeminent regions as candidates to body extremities, that is, hands, feet and head. The found extremities receive a label that indicates the body part it may represent. To classify the users pose, the vector with the coordinates of his extremities is compared to keyposes and the best match is selected. The final step is the temporal classification, that is, the gesture recognition. The developed technique is robust, working well even when the system was trained with an user and applied to another users data.

The purpose of this project is to implement, analyze and compare visual odometry approaches to help the localization task in autonomous vehicles. The stereo visual odometry algorithm Libviso2 is compared with a proposed omnidirectional multi-camera approach. The proposed method consists of performing monocular visual odometry on all cameras individually and selecting the best estimate through a voting scheme involving all cameras. The omnidirectionality of the vision system allows the part of the surroundings richest in features to be used in the relative pose estimation. Experiments are carried out using cameras Bumblebee XB3 and Ladybug 2, fixed on the roof of a vehicle. The voting process of the proposed omnidirectional multi-camera method leads to some improvements relatively to the individual monocular estimates. However, stereo visual odometry provides considerably more accurate results. / O objetivo deste mestrado é implementar, analisar e comparar abordagens de odometria visual, de forma a contribuir para a localização de um veículo autônomo. O algoritmo de odometria visual estéreo Libviso2 é comparado com um método proposto, que usa um sistema multi-câmera omnidirecional. De acordo com este método, odometria visual monocular é calculada para cada câmera individualmente e, seguidamente, a melhor estimativa é selecionada através de um processo de votação que involve todas as câmeras. O fato de o sistema de visão ser omnidirecional faz com que a parte dos arredores mais rica em características possa sempre ser usada para estimar a pose relativa do veículo. Nas experiências são utilizadas as câmeras Bumblebee XB3 e Ladybug 2, fixadas no teto de um veículo. O processo de votação do método multi-câmera omnidirecional proposto apresenta melhorias relativamente às estimativas monoculares individuais. No entanto, a odometria visual estéreo fornece resultados mais precisos.

"Patient motion, which causes artifacts in reconstructed images, can be a serious problem in Single Photon Emission Computed Tomography (SPECT) imaging. If patient motion can be detected and quantified, the reconstruction algorithm can compensate for the motion. A real-time multi-threaded Visual Tracking System (VTS) using optical cameras, which will be suitable for deployment in clinical trials, is under development. The VTS tracks patients using multiple video images and image processing techniques, calculating patient motion in three-dimensional space. This research aimed to develop and implement an algorithm for feature matching and stereo location computation using multiple cameras. Feature matching is done based on the epipolar geometry constraints for a pair of images and extended to the multiple view case with an iterative algorithm. Stereo locations of the matches are then computed using sum of squared distances from the projected 3D lines in SPECT coordinates as the error metric. This information from the VTS, when coupled with motion assessment from the emission data itself, can provide a robust compensation for patient motion as part of reconstruction."