A Fast Passenger Detection Algorithm Based on Head Area Characteristics

For the passengers monitoring requirements of rail subway and public places, this paper presents a method which can detect passengers based on regional characteristics of human head. The method uses adaptive background updating algorithm to establish a background model. Then the target area can be extracted and marked by background difference method. According to the matching degree and refusing degree of the target area, human head can be identified. This method has excellent performance of real-time, and requires little computational cost. At the end, simulation results show correctness and effectiveness of the method proposed in this paper.