Face acquiring optimization based on video sensor network

In video sensor network, acquiring the optimal facial images is of great significance for improving the efficiency of pedestrian target recognition and enhancing the performance of video sensor network. In this paper a new optimal facial images acquiring method based on collaborative information processing in video sensor network is proposed to improve facial images obtaining efficiency and reduce network energy consumption. The Gaussian background subtraction algorithm is employed to extract foreground target information against background. The target's position in the real scenario is determined using collaborative multi-angle positioning method. The ARMA model is utilized to filter the target trajectory. Then the face quality score of each node is calculated using the proposed face quality assessment function with the target's moving state information. The optimal face images are acquired according to the quality score. The experimental results show that the proposed method can achieve the target trajectory tracking accurately, obtain the optimal face images and reduce the energy consumption of video sensor network.