Pedestrian Recognition Based on Saliency Detection and Kalman Filter Algorithm in Aerial Video
Xingbao Wang, Liu Chunping, Gong Liu, Long Liu, Shengrong Gong · 2011
For the problem of low resolution, camera movement and target's detail fuzzy in aerial video when detecting and recognizing pedestrian, this paper proposes weighted region matching algorithm based on saliency detection and Kalman filter(KS-WRM). In the preprocessing stage, the KS-WRM algorithm uses saliency detection algorithm, which adds human subjective consciousness to segment pedestrians. The result is perfect and improves recognition accuracy. In the matching stage, the KS-WRM algorithm first uses Kalman filter algorithm to label candidate's region, and then selects candidates using weighted region matching algorithm in labeled region, which can avoid the problem of selecting candidates under supervision. As a result, it not only cuts down calculated amount, but also improves adaptive and real-time ability, then applies successfully in the video field. With a large number of experiments in aerial video of complex environment, it is demonstrated that the proposed method outperforms recent state-of-the-art methods.