This paper analyzes the difference between the imaging mechanism of the infrared images and that of the visible light images, and find that it is important to extract the stable and reliable common feature for object recognition. Then we propose a target recognition algorithm based on histograms of oriented gradients (HOG) which evaluates normalized local histograms of image gradient orientations in a dense grid. Last we adopt linear SVM trained for a binary object/non-object classifier and detect the object in the forward-looking infrared (FLIR) images. The experiment results suggest that the proposed approach has high rates of detection. Furthermore, we study how to select positive and negative samples for a better performance.

Date Published: 8 December 2011
PDF: 5 pages
Proc. SPIE 8003, MIPPR 2011: Automatic Target Recognition and Image Analysis, 80030S (8 December 2011); doi: 10.1117/12.901811

Published in SPIE Proceedings Vol. 8003:
MIPPR 2011: Automatic Target Recognition and Image Analysis
Tianxu Zhang; Nong Sang, Editor(s)