Collaborative target tracking in lopor with multi-camera

The moving target tracking results of multiple single cameras can rarely be efficiently collaborated by using the current multi-camera target tracking algorithms, in the layout of partly overlapping regions (LOPOR) with multi-camera. To solve this problem, a novel algorithm of moving target collaborative tracking with multi-camera, based on LOPOR with multi camera, is presented. The completely overlapping region in LOPOR with multi-camera is taken by the algorithm to calibrate multiple single cameras collectively. The results of single camera target tracking and tracking targets significance rank are collaborated. Finally, the results of single camera targets tracking are modified using the targets' overall information. Then, interaction and sharing of each single camera target tracking are realized. The experimental results show that the single camera target tracking can be collaborated efficiently using the algorithm in LOPOR with multi-camera, with high monitoring efficiency and accurate location. In this manner, moving targets were tracked sustainably over a long distance for a large scene. (C) 2016 Elsevier GmbH. All rights reserved.