@inproceedings{354ac18a67d7473b89058302cd42d737,
title = "Morphological independence for landmark detection in vision based SLAM",
abstract = "Morphologically independent vectors correspond to approximations to the vertices of the convex hull covering the data vectors in high dimensional space. We use Morphological Associative Memories (MAM) for the induction of sets of morphologically independent vectors from data. Simultaneous Localization and Mapping (SLAM) is the process of simultaneously building a map of the environment and localizing the mapping agent. In this paper we explore the realization of non-metric SLAM using a visual information based approach relying on morphologically independent images induced from a mobile robot camera image stream. The selected images are proposed as the landmarks for localization, building simultaneously a qualitative map of the environment. We report results of some experiments on data gathered from an indoor ambient.",
author = "Ivan Villaverde and Manuel Gra{\~n}a and Alicia D'Anjou",
year = "2007",
doi = "10.1007/978-3-540-73007-1\_102",
language = "English",
isbn = "9783540730064",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "847--854",
booktitle = "Computational and Ambient Intelligence - 9th International Work-Conference on Artificial Neural Networks, IWANN 2007, Proceedings",
address = "Germany",
note = "9th International Work-Conference on Artificial Neural Networks, IWANN 2007 ; Conference date: 20-06-2007 Through 22-06-2007",
}