The Community for Technology Leaders
Green Image
Issue No. 02 - February (2012 vol. 34)
ISSN: 0162-8828
pp: 302-314
Shoushun Chen , Sch. of Electr. & Electron. Eng. (EEE), Nanyang Technol. Univ., Singapore, Singapore
P. Akselrod , Electr. Eng. Dept., Yale Univ., New Haven, CT, USA
Bo Zhao , Sch. of Electr. & Electron. Eng. (EEE), Nanyang Technol. Univ., Singapore, Singapore
J. A. Perez-Carrasco , Inst. Microelectron. Sevilla (IMSE), Sevilla, Spain
B. Linares-Barranco , Inst. Microelectron. Sevilla (IMSE), Sevilla, Spain
E. Culurciello , Electr. Eng. Dept., Yale Univ., New Haven, CT, USA
This paper proposes an algorithm for feedforward categorization of objects and, in particular, human postures in real-time video sequences from address-event temporal-difference image sensors. The system employs an innovative combination of event-based hardware and bio-inspired software architecture. An event-based temporal difference image sensor is used to provide input video sequences, while a software module extracts size and position invariant line features inspired by models of the primate visual cortex. The detected line features are organized into vectorial segments. After feature extraction, a modified line segment Hausdorff-distance classifier combined with on-the-fly cluster-based size and position invariant categorization. The system can achieve about 90 percent average success rate in the categorization of human postures, while using only a small number of training samples. Compared to state-of-the-art bio-inspired categorization methods, the proposed algorithm requires less hardware resource, reduces the computation complexity by at least five times, and is an ideal candidate for hardware implementation with event-based circuits.
visual perception, feature extraction, image sensors, learning (artificial intelligence), pattern classification, pattern clustering, software architecture, video signal processing, human postures, object feedforward categorization, real-time video sequence, address-event temporal-difference image sensors, event- based hardware, bio-inspired software architecture, primate visual cortex, line features, vectorial segments, feature extraction, line segment Hausdorff- distance classifier, on the fly cluster based size invariant categorization, on the fly cluster based position invariant categorization, training samples, state-of-the-art bio-inspired categorization methods, computation complexity, hardware implementation, event-based circuits, Feature extraction, Image segmentation, Image sensors, Libraries, Human factors, address-event image sensor., Human posture categorization, bio-inspired categorization, event-based circuits

J. A. Perez-Carrasco, Bo Zhao, P. Akselrod, Shoushun Chen, B. Linares-Barranco and E. Culurciello, "Efficient Feedforward Categorization of Objects and Human Postures with Address-Event Image Sensors," in IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 34, no. , pp. 302-314, 2012.
82 ms
(Ver 3.3 (11022016))