Vision-Based Motion Detection Using C-NLPCA Approach

In this paper we propose an original method for visionbased motion detection. It allows an autonomous robot to explore an environment, acquiring information, about its own motion aiming to solve future positioning (navigation) problems. This system uses only visual information and no explicit heuristic information (landmarks) for the localization/ navigation process. It is based on the method of Cascaded Nonlinear Components Analysis (C-NLPCA), in order to get a dimensional reduction of the data (image series) which may consider its nonlinearity. For the verification of possible improvements brought by the use of C-NLPCA, a set of analyses were done and the results will be demonstrated in this paper. The C-NLPCA revealed to be efficient, propelling a new field of research.