Pedestrian Detection based on Improved Shape Context in Infrared images

摘要

this paper presents an approach toward pedestrian detection applied to infrared images using improved Shape Context feature (SC). To facilitate the task of Shape Context feature extracting, we apply Bilateral Filtering for infrared images to preserve the information of large sharp edges. Then, we study an image descriptor based on an improved Shape Context feature (SC), which is more robust to object deformation. For a test image, improved SC features are extracted and matched to the codebook. A voting scheme then obtains object locations from the matching results. Experimental results with OSU Infrared Image Database demonstrate the accuracy and robustness of our algorithm, and this method is promising.