Target detection and tracking based on embedded multicore DSP system

摘要

In this paper, a multicore DSP parallel implementation strategy for infrared target detection and tracking is proposed. Tracking-Learning-Detection is considered a highly efficient algorithm for tracking a single target. Although this algorithm can re-track a target when the target is occluded by other targets, many shortcomings still exist. This paper deals with the issue of target tracking by fusing local spectrum suppression based on human visual attention mechanism with tracking-learning-detection algorithm. Specifically, the area of the target is estimated to reduce the detection region and to increase the processing speed. Experimental results conducted on VIVID benchmark video library demonstrate that the proposed method can detect properly and track accurately infrared target in complex scenes.