Real-Time Target Detection Algorithm Based on Background Modeling

Identifying moving objects from sequence video is a fundamental and critical task in intelligent surveillance applications. In this paper, we design an intelligent video surveillance terminal based on DSP+FPGA structure. Considering the hardware resources, the real-time target detection algorithm is the most important factor. In fact, the common approach which identified moving objects from the portion of a video frame for static camera is background modeling algorithm. However, this paper introduces and simulates five background modeling algorithms, analyzes the simulation results, tests the real-time nature of each algorithm, and selects the best algorithm transplant to hardware platform to achieve a real-time intelligent video surveillance.