Counting Pedestrians Passing through a Line in Video Sequences based on Optical Flow Extraction

摘要

In recent researches on video processing-based pedestrian counting, a pedestrian density estimation method based on the number of optical flows has developed and attracted attentions. In this paper, we propose a simple pedestrian counting method by applying the number of optical flows for estimating the number of pedestrians passing through a specific line, called counting line, in crowded situations. Through experimental evaluations using actual video sequences, we show that the correlation coefficient between the number of optical flows and the number of pedestrians passing through the counting line is high. In addition, we show that the mean relative error of our proposed method is less than 0.068.