This paper presents a FPGA-based design that aims at apply in real-time vision processes, specially Optical flow estimation processes. The objective of this work is to provide critical information for autonomous flights in a micro air robot. As motion field is one of the crucial information for the safety of robot. Based on these motion information, obstacles avoidance, for example, could be added to increase the autonomous degree of the robot. The structure of this paper is as follows. In section 2, the Lucas and Kanade optical flow algorithm is presented and few considerations on its implementation are proposed. The next sections (3 and 4) introduce the data flow design and proposes the hardware implementation of the process. Finally, experimental results on realistic image sequence obtained by an implementation on the authors' smart camera (SeeMos) is given in section 5.
Hardware implementation of a real time Lucas and Kanade optical flow
Hardware-Implementierung für den Lucas- und Kanade-Algorithmus zur Echtzeitverarbeitung der optischen Bildfolge
2009
5 Seiten, 4 Bilder, 3 Tabellen, 15 Quellen
(nicht paginiert)
Conference paper
Storage medium
English
Evaluation of Lucas-Kanade based optical flow algorithm
IEEE | 2018
|Lucas-Kanade 20 Years On A Unifying Framework
British Library Online Contents | 2004
|Lucas-Kanade based entropy congealing for joint face alignment
British Library Online Contents | 2012
|Lucas/Kanade Meets Horn/Schunck Combining Local and Global Optic Flow Methods
British Library Online Contents | 2005
|