This book introduces several appearance-based place recognition pipelines based on different mapping techniques for addressing loop-closure detection in mobile platforms with limited computational resources. The motivation behind this book has been the prospect that in many contemporary applications efficient methods are needed that can provide high performance under run-time and memory constraints. Thus, three different mapping techniques for addressing the task of place recognition for simultaneous localization and mapping (SLAM) are presented. The book at hand follows a tutorial-based structure describing each of the main parts needed for a loop-closure detection pipeline to facilitate the newcomers. It mainly goes through a historical review of the problem, focusing on how it was addressed during the years reaching the current age. This way, the reader is initially familiarized with each part while the place recognition paradigms follow.

This thesis deals with the task of appearance-based mapping and place recognition for mobile robots. More specifically, this work aims to identify how location models can be improved by exploring several existing and novel location representations in order to better exploit the available visual information. Appearance-based mapping and place recognition presents a number of challenges, including making reliable data-association decisions given repetitive and self-similar scenes (perceptual aliasing), variations in view-point and trajectory, appearance changes due to dynamic elements, lighting changes, and noisy measurements. As a result, choices about how to model and compare observations of locations is crucial to achieving practical results. This includes choices about the types of features extracted from imagery, how to define the extent of a location, and how to compare locations. Along with investigating existing location models, several novel methods are developed in this work. These are developed by incorporating information about the underlying structure of the scene through the use of covisibility graphs which capture approximate geometric relationships between local landmarks in the scene by noting which ones are observed together. Previously, the range of a location generally varied between either using discrete poses or loosely defined sequences of poses, facing problems related to perceptual aliasing and trajectory invariance respectively. Whereas by working with covisibility graphs, scenes are dynamically retrieved as clusters from the graph in a way which adapts to the environmental structure and given query. The probability of a query observation coming from a previously seen location is then obtained by applying a generative model such that the uniqueness of an observation is accounted for. Behaviour with respect to observation errors, mapping errors, perceptual aliasing, and parameter sensitivity are examined, motivating the use of a novel normalization scheme and observation likelihoods representations. The normalization method presented in this work is robust to redundant locations in the map (from missed loop-closures, for example), and results in place recognition which now has sub-linear complexity in the number of locations in the map. Beginning with bag-of-words representations of locations, location models are extended in order to include more discriminative structural information from the covisibility map. This results in various representations ranging between unstructured sets of features and full graphs of features, providing a tradeoff between complexity and recognition performance.

Mapping and localization are two essential tasks in autonomous mobile robotics. Due to the unavoidable noise that sensors present, mapping algorithms usually rely on loop closure detection techniques, which entail the correct identification of previously seen places to reduce the uncertainty of the resulting maps. This book deals with the problem of generating topological maps of the environment using efficient appearance-based loop closure detection techniques. Since the quality of a visual loop closure detection algorithm is related to the image description method and its ability to index previously seen images, several methods for loop closure detection adopting different approaches are developed and assessed. Then, these methods are used in three novel topological mapping algorithms. The results obtained indicate that the solutions proposed attain a better performance than several state-of-the-art approaches. To conclude, given that loop closure detection is also a key component in other research areas, a multi-threaded image mosaicing algorithm is proposed. This approach makes use of one of the loop closure detection techniques previously introduced in order to find overlapping pairs between images and finally obtain seamless mosaics of different environments in a reasonable amount of time.

Simultaneous Localization and Mapping (SLAM) has been a long-standing research problem in robotics. It describes the problem of a robot mapping an unknown environment, while simultaneously localizing in it with the help of the incomplete map. This book describes a technique called Switchable Constraints.Switchable Constraints help to increase the robustness of SLAM against data association errors and in particular against false positive loop closure detections. Such false positive loop closure detections can occur when the robot erroneously assumes it re-observed a landmark it has already mapped or when the appearance of the observed surroundings is very similar to the appearance of other places in the map. Ambiguous observations and appearances are very common in human-made environments such as office floors or suburban streets, making robustness against spurious observations a key challenge in SLAM. The book summarizes the foundations of factor graph-based SLAM techniques. It explains the problem of data association errors before introducing the novel idea of Switchable Constraints. We present a mathematical derivation and probabilistic interpretation of Switchable Constraints along with evaluations on different datasets. The book shows that Switchable Constraints are applicable beyond SLAM problems and demonstrates the efficacy of this technique to improve the quality of satellite-based localization in urban environments, where multipath and non-line-of-sight situations are common error sources.

This book constitutes the refereed proceedings of the 6th Iberian Conference on Pattern Recognition and Image Analysis, IbPRIA 2013, held in Funchal, Madeira, Portugal, in June 2013. The 105 papers (37 oral and 68 poster ones) presented were carefully reviewed and selected from 181 submissions. The papers are organized in topical sections on computer vision, pattern recognition, image and signal, applications.

The two volume set LNAI 9413 + 9414 constitutes the proceedings of the 14th Mexican International Conference on Artificial Intelligence, MICAI 2015, held in Cuernavaca,. Morelos, Mexico, in October 2015. The total of 98 papers presented in these proceedings was carefully reviewed and selected from 297 submissions. They were organized in topical sections named: natural language processing; logic and multi-agent systems; bioinspired algorithms; neural networks; evolutionary algorithms; fuzzy logic; machine learning and data mining; natural language processing applications; educational applications; biomedical applications; image processing and computer vision; search and optimization; forecasting; and intelligent applications.