No description available.

Dans plusieurs tâches de la robotique, la vision est considérée comme l’élément essentiel avec lequel la perception de l’environnement ou l’interaction avec d’autres utilisateurs peut se réaliser. Néanmoins, les artefacts potentiellement présents dans les images capturées rendent la tâche de reconnaissance et d’interprétation de l’information visuelle extrêmement compliquée. Il est de ce fait, très important d’utiliser des primitives robustes, stables et ayant un taux de répétabilité élevé afin d’obtenir de bonnes performances. Cette thèse porte sur les problèmes de localisation et de détection de fermeture de boucle d’un robot mobile en utilisant la saillance visuelle. Les résultats en termes de précision et d’efficacité des applications de localisation et de détection de fermeture sont évalués et comparés aux résultats obtenus avec des approches de l’état de l’art sur différentes séquences d’images acquises en milieu extérieur. Le principal inconvénient avec les modèles proposés pour l’extraction de zones de saillance est leur complexité de calcul, ce qui conduit à des temps de traitement important. Afin d’obtenir un traitement en temps réel, nous présentons dans ce mémoire l’implémentation du détecteur de régions saillantes sur la plate forme reconfigurable DreamCam. / In several tasks of robotics, vision is considered to be the essential element by which the perception of the environment or the interaction with other users can be realized. However, the potential artifacts in the captured images make the task of recognition and interpretation of the visual information extremely complicated. It is therefore very important to use robust, stable and high repeatability rate primitives to achieve good performance. This thesis deals with the problems of localization and loop closure detection for a mobile robot using visual saliency. The results in terms of accuracy and efficiency of localization and closure detection applications are evaluated and compared to the results obtained with the approaches provided in literature, both applied on different sequences of images acquired in outdoor environnement. The main drawback with the models proposed for the extraction of salient regions is their computational complexity, which leads to significant processing time. To obtain a real-time processing, we present in this thesis also the implementation of the salient region detector on the reconfigurable platform DreamCam.

Visual place recognition is the task of finding same places in a set of database images for a given set of query images. This becomes particularly challenging for long-term applications when the environmental condition changes between or within the database and query set, e.g., from day to night. Visual place recognition in changing environments can be used if global position data like GPS is not available or very inaccurate, or for redundancy. It is required for tasks like loop closure detection in SLAM, candidate selection for global localization, or multi-robot/multi-session mapping and map merging. In contrast to pure image retrieval, visual place recognition can often build upon additional information and data for improvements in performance, runtime, or memory usage. This includes additional data-inherent knowledge about information that is contained in the image sets themselves because of the way they were recorded. Using data-inherent knowledge avoids the dependency on other sensors, which increases the generality of methods for an integration into many existing place recognition pipelines. This thesis focuses on the usage of additional data-inherent knowledge. After the discussion of basics about visual place recognition, the thesis gives a systematic overview of existing data-inherent knowledge and corresponding methods. Subsequently, the thesis concentrates on a deeper consideration and exploitation of four different types of additional data-inherent knowledge. This includes 1) sequences, i.e., the database and query set are recorded as spatio-temporal sequences so that consecutive images are also adjacent in the world, 2) knowledge of whether the environmental conditions within the database and query set are constant or continuously changing, 3) intra-database similarities between the database images, and 4) intra-query similarities between the query images. Except for sequences, all types have received only little attention in the literature so far. For the exploitation of knowledge about constant conditions within the database and query set (e.g., database: summer, query: winter), the thesis evaluates different descriptor standardization techniques. For the alternative scenario of continuous condition changes (e.g., database: sunny to rainy, query: sunny to cloudy), the thesis first investigates the qualitative and quantitative impact on the performance of image descriptors. It then proposes and evaluates four unsupervised learning methods, including our novel clustering-based descriptor standardization method K-STD and three PCA-based methods from the literature. To address the high computational effort of descriptor comparisons during place recognition, our novel method EPR for efficient place recognition is proposed. Given a query descriptor, EPR uses sequence information and intra-database similarities to identify nearly all matching descriptors in the database. For a structured combination of several sources of additional knowledge in a single graph, the thesis presents our novel graphical framework for place recognition. After the minimization of the graph's error with our proposed ICM-based optimization, the place recognition performance can be significantly improved. For an extensive experimental evaluation of all methods in this thesis and beyond, a benchmark for visual place recognition in changing environments is presented, which is composed of six datasets with thirty sequence combinations.