A People-Counting System Based on BP Neural Network

A people-counting system based on a back propagation (BP) neural network is proposed in this paper. The proposed system uses cheap photoelectric sensor to collect data and introduces BP neural network for counting and recognition, and it is effective and flexible for the purpose of performing people counting. In this paper, new methods for segmentation and feature extraction are developed to enhance the classification performance. Promising results were obtained and the analysis indicates that the proposed system based on BP neural network provides good results with low false rate and it is effective for people-counting.