Nowadays due to advancements in Computer Vision algorithms, there is an increasing demand for cost effective and reliable performance evaluation of its algorithms. For objective evaluation of camera object and lane detection algorithms, generation of ground truth is important. Ground truth refers to the set of measurements known to be accurate thereby acting as reference. In context of Computer Vision, it includes a set of labeled images. However, obtaining it by direct observation is not possible. Thus, its generation is an intriguing challenge and numerous efforts have been made in accordance to that. Most commonly, it is generated manually thereby consuming lot of time, effort and money. The existing or proposed semi-automatic approaches fail to address the issues of manual ground truth generation. Currently in the Bosch Validation toolchain, synchronized reference ground truth for Bosch Camera sensor is not present and the annotation is done manually. The term 'reference ground truth' in this thesis refers to the information provided by the reference sensors. Hence, this thesis focuses on the feasibility of Ibeo reference sensors and Ibeo tool chain to obtain synchronized reference GT for enabling automatic preannotation process. The report thus highlights the details related to research, formulated concepts, employed synchronization approach and the developed label toolbox for facilitating fullautomatic pre-annotation process with Reference Sensors System. Upon visualizing the results, the need for cross calibration between Bosch camera and reference sensors is specified. Moreover, with the help of results and findings, requirements are derived to improve the proposed annotation process and to obtain more accurate reference ground truth for evaluation of Front Camera algorithms.

L'extraction des paramètres de surface est une activité essentielle des Sciences de la vie et de la Terre. Ce mémoire propose une nouvelle méthodologie pour l'analyse des paramètres biophysiques et géophysiques, appelée Réalité Terrain Étendue, et qui mêle les avantages des relevés terrain et de la télédétection. Nous nous sommes en particulier attachés aux avantages de la télédétection basse altitude et d'un système micro-drone multi-caméras pour la cartographie de la dynamique de la végétation à l'échelle des individus. Cette problématique pose de nombreuses contraintes sur notre système car l'identification des arbrisseaux nécessite des capteurs innovants et une adaptation aux cycles phénologiques pour améliorer leur capacité de discrimination. La télédétection basse altitude semble être une solution intéressante en terme de résolution spatiale et de souplesse opérationnelle, et le développement des micro-drones civils permet des outils d'autant plus performants et fiables pour les missions terrain. Nous avons donc mis en place un système complet de drone avec une charge utile spécifique emportant simultanément trois appareils photographiques pour l'acquisition à la demande d'images obliques, stéréoscopiques ou multispectrales et permettant le développement de nouvelles méthodes d'identification de la végétation. Enfin, en participant à un relevé terrain du Muséum national d'Histoire naturelle, nous avons validé l'intérêt de notre système pour la cartographie de la dynamique de la végétation. Ce travail s'ouvre sur de nombreuses applications et perspectives de recherche, comme l'extraction de paramètres biophysiques par stéréo-restitution et l'agriculture de précision / Retrieval of land surface parameters is an essential part of life and Earth sciences activities, as they are the key to understand the complex phenomena that take place in the biosphere. This thesis proposes a new methodology for biophysical and geophysical surface para-meters analysis, that we call Extended Ground Truth, and that mix the advantages of the field ope-rations and those of remote sensing. We have especially focused our work on the benefits of low altitude remote sensing with a multi-cameras Unmanned Aerial System for vegetation dynamics mapping at subject scale. This problematic raises strong constraints on the system, as the identification of small bushes implies new type of sensors and the adaptation to the phenological cycles to improve the discrimination capacity of the sensors. Low altitude remote sensing seems a good solution in terms of resolution and operational flexibility and the technical revolution behind the spreading of civilian micro-UAV allows all the more capable and reliable tools for field operations. We therefore developed an Unmanned Aerial System with a specific payload that can lift up three digital cameras at once, allowing on demand oblique, stereoscopic or multispectral synchronous acquisitions and thus the development of new identification methods of the vegetation. Finally, we have participated in a field operation on a test site of the National Natural History Museum and confirmed the interest of our tool for vegetation dynamics mapping. This work leads to many other researches, per example in the field of biophysical parameters extraction from stereoscopic images, and other application fields like precision farming

<p>Mammographic screening is an effective way to detect breast cancer. Early detection of breast cancer depends to a high degree on the adequacy of the mammogram from which the diagnosis is made. Today, most of the analysis of the mammogram is performed by radiologists. Computer-aided diagnosis (CAD) systems have been proposed as an aid to increase the efficiency and effectiveness of the screening procedure by automatically indicating abnormalities in the mammograms. However, in order for a CAD system to be stable and efficient, the input images need to be adequate. Criteria for adequacy include: high resolution, low image noise and high image contrast. Additionally, the breast needs to be adequately positioned and compressed to properly visualise the entire breast and especially the glandular tissue.</p><p>This thesis addresses questions regarding the automatic determination of mammogram adequacy with the focus on breast positioning and segmentation evaluation. The goal and, thus, the major technical challenge is to develop algorithms that support fully automatic quality checks. The relevant quality criteria are discussed in Chapter 2. The aim of this discussion is to compile a comprehensive list of necessary criteria that a system for automatic assessment of mammographic adequacy must satisfy. Chapter 3 gives an overview of research performed in computer-aided analysis of mammograms. It also provides basic knowledge about image analysis involved in the research area of computerized mammography in general, and in the papers of this thesis, in particular. In contrast, Chapter 4 describes basic knowledge about segmentation evaluation, which is a highly important component in image analysis. Papers I–IV propose algorithms for measuring the quality of a mammogram according to certain criteria and addresses problems related to them. A method for automatic analysis of the shape of the pectoralis muscle is presented in Paper I. Paper II proposes a fully automatic method for extracting the breast border. A geometric assumption used by radiologists is that the nipple is located at the point on the breast border being furthest away from the pectoralis muscle. This assumption is investigated in Paper III, and a method for automatically restricting the search area is proposed. There has been an increasing need to develop an automated segmentation algorithm for extracting the glandular tissue, where the majority of breast cancer occur. In Paper IV, a novel approach for solving this problem in a robust and accurate way is proposed. Paper V discusses the challenges involved in evaluating the quality of segmentation algorithms based on ground truths provided by an expert panel. A method to relate ground truths provided by several experts to each other in order to establish levels of agreement is proposed. Furthermore, this work is used to develop an algorithm that combines an ensemble of markings into one surrogate ground truth.</p>

Mammographic screening is an effective way to detect breast cancer. Early detection of breast cancer depends to a high degree on the adequacy of the mammogram from which the diagnosis is made. Today, most of the analysis of the mammogram is performed by radiologists. Computer-aided diagnosis (CAD) systems have been proposed as an aid to increase the efficiency and effectiveness of the screening procedure by automatically indicating abnormalities in the mammograms. However, in order for a CAD system to be stable and efficient, the input images need to be adequate. Criteria for adequacy include: high resolution, low image noise and high image contrast. Additionally, the breast needs to be adequately positioned and compressed to properly visualise the entire breast and especially the glandular tissue. This thesis addresses questions regarding the automatic determination of mammogram adequacy with the focus on breast positioning and segmentation evaluation. The goal and, thus, the major technical challenge is to develop algorithms that support fully automatic quality checks. The relevant quality criteria are discussed in Chapter 2. The aim of this discussion is to compile a comprehensive list of necessary criteria that a system for automatic assessment of mammographic adequacy must satisfy. Chapter 3 gives an overview of research performed in computer-aided analysis of mammograms. It also provides basic knowledge about image analysis involved in the research area of computerized mammography in general, and in the papers of this thesis, in particular. In contrast, Chapter 4 describes basic knowledge about segmentation evaluation, which is a highly important component in image analysis. Papers I–IV propose algorithms for measuring the quality of a mammogram according to certain criteria and addresses problems related to them. A method for automatic analysis of the shape of the pectoralis muscle is presented in Paper I. Paper II proposes a fully automatic method for extracting the breast border. A geometric assumption used by radiologists is that the nipple is located at the point on the breast border being furthest away from the pectoralis muscle. This assumption is investigated in Paper III, and a method for automatically restricting the search area is proposed. There has been an increasing need to develop an automated segmentation algorithm for extracting the glandular tissue, where the majority of breast cancer occur. In Paper IV, a novel approach for solving this problem in a robust and accurate way is proposed. Paper V discusses the challenges involved in evaluating the quality of segmentation algorithms based on ground truths provided by an expert panel. A method to relate ground truths provided by several experts to each other in order to establish levels of agreement is proposed. Furthermore, this work is used to develop an algorithm that combines an ensemble of markings into one surrogate ground truth.

A COMPARATIVE EVALUATION OF FOREGROUND / BACKGROUND SEGMENTATION ALGORITHMS Pakyurek, Muhammet M.Sc., Department of Electrical and Electronics Engineering Supervisor: Prof. Dr. G&ouml / zde Bozdagi Akar September 2012, 77 pages Foreground Background segmentation is a process which separates the stationary objects from the moving objects on the scene. It plays significant role in computer vision applications. In this study, several background foreground segmentation algorithms are analyzed by changing their critical parameters individually to see the sensitivity of the algorithms to some difficulties in background segmentation applications. These difficulties are illumination level, view angles of camera, noise level, and range of the objects. This study is mainly comprised of two parts. In the first part, some well-known algorithms based on pixel difference, probability, and codebook are explained and implemented by providing implementation details. The second part includes the evaluation of the performances of the algorithms which is based on the comparison v between the foreground background regions indicated by the algorithms and ground truth. Therefore, some metrics including precision, recall and f-measures are defined at first. Then, the data set videos having different scenarios are run for each algorithm to compare the performances. Finally, the performances of each algorithm along with optimal values of their parameters are given based on f measure.

The objective of this dissertation is to develop a Multi-Camera Lane Departure Warning (MCLDW) system and a framework to evaluate it. A Lane Departure Warning (LDW) system is a safety feature that is included in a few luxury automobiles. Using a single camera, it performs the task of informing the driver if a lane change is imminent. The core component of an LDW system is a lane detector, whose objective is to find lane markers on the road. Therefore, we start this dissertation by explaining the requirements of an ideal lane detector, and then present several algorithmic implementations that meet these requirements. After selecting the best implementation, we present the MCLDW methodology. Using a multi-camera setup, MCLDW system combines the detected lane marker information from each camera's view to estimate the immediate distance between the vehicle and the lane marker, and signals a warning if this distance is under a certain threshold. Next, we introduce a procedure to create ground truth and a database of videos which serve as the framework for evaluation. Ground truth is created using an efficient procedure called Time-Slicing that allows the user to quickly annotate the true locations of the lane markers in each frame of the videos. Subsequently, we describe the details of a database of driving videos that has been put together to help establish a benchmark for evaluating existing lane detectors and LDW systems. Finally, we conclude the dissertation by citing the contributions of the research and discussing the avenues for future work.

La validation des algorithmes de tractographie et l’optimisation des paramètres choisis en référence à une vérité terrain, sont primordiales avant l’utilisation de ces méthodes en routine clinique. D’une part, nous présentons une méthode de comparaison quantitative de reconstructions issues de la tractographie à celles reconstruites depuis la dissection par un scanner laser. Cette comparaison permet d’évaluer les reconstructions de différents modèles et algorithmes de tractographie (déterministe, probabiliste) en les confrontant à une vérité terrain acquise par le scanner laser (surfaces triangulées). La transformation des données sous des formats communs était nécessaire avant leur comparaison quantitative. Deux méthodes de comparaison, surface-surface et volume-volume ont été proposées. D’autre part, nous présentons une méthode d’optimisation par algorithme génétique (AG), des paramètres de tractographie déterministe. L’AG est un algorithme itératif d’optimisation basé sur la sélection naturelle, il est capable d’optimiser des problèmes complexes ayant plusieurs paramètres. Etant donné la vérité terrain d’un faisceau, l’AG se propose de trouver le jeu de paramètres optimal donnant le meilleur résultat de tractographie. Les méthodes de comparaisons et d’optimisation ont été appliquées à un faisceau d’objet test puis à deux faisceaux disséqués à partir d’un cerveau humain post mortem après acquisitions IRM et scanner laser. / Tractography validation and optimization of tracking parameters against a ground truth are mandatory before a large clinical use. First, we present a method to quantitatively compare tractography reconstructions to a ground truth derived from laser scanner acquisitions of dissected specimens. This comparison allows evaluation of multiple models and tractography approaches (deterministic, probabilistic…). The ground truth used for this comparison was acquired from dissected specimens using a surface laser scanner, which produces triangulated surface meshes. Data transformation to a common format was necessary before quantitative comparisons. Two comparison methods were proposed, surface-to-surface and volume-to-volume. Second, we propose a method for automatic optimization of deterministic tractography parameters using a genetic algorithm (GA). The GA is an iterative optimization algorithm based on natural selection, which is able to optimize complex problems having several parameters. For a given ground truth fasciculus, the GA was expected to find the set of tractography parameters producing the best tractography result according to the ground truth. The comparison and optimization methods were applied to a synthetic bundle derived from a phantom and to two dissected white matter tracts of a human post mortem brain.

Hlavním přínosem této práce jsou dva nové přístupy pro zvýšení bezpečnosti biometrických systémů založených na rozpoznávání podle otisků prstů. První přístup je z oblasti testování živosti a znemožňuje použití různých typů falešných otisků prstů a jiných metod oklamání senzoru v průběhu procesu snímání otisků. Tento patentovaný přístup je založen na změně barvy a šířky papilárních linií vlivem přitlačení prstu na skleněný podklad. Výsledná jednotka pro testování živosti může být integrována do optických senzorů.  Druhý přístup je z oblasti standardizace a zvyšuje bezpečnost a interoperabilitu procesů extrakce markantů a porovnání. Pro tyto účely jsem vytvořila metodologii, která stanovuje míry sémantické shody pro extraktory markantů otisků prstů. Markanty nalezené testovanými extraktory jsou porovnávány oproti Ground-Truth markantům získaným pomocí shlukování dat poskytnutých daktyloskopickými experty. Tato navrhovaná metodologie je zahrnuta v navrhovaném dodatku k normě ISO/IEC 29109-2 (Amd. 2 WD4).

No description available.

Objectivity and neutrality of current makers of online maps is hampered by the fact that the most important of them who form the public opinion are largely commercial giants. Sensitivity of approaches is desirable especially in specific situations involving territorial disputes. The textual part of the thesis brings theoretical insights to the problematics of map makers policies and their preservation or violation. At the same time is describes critical cartography as a current opposition against governmental and cooperative mapping, of which counter-mapping processes are regularly used in art, too. Practical outcome attempts to reflect specific approaches of map makers depending on particular territorial disputes. It takes place mostly in online environment, using the functions of Google Street View or Google Earth and forms of counter-mapping on the OSM platform. The goal is to bring alternative realities complicating map makers policies and governing apparatus to map projects with the international coverage.