Research on central issues of crowd density estimation

Crowd density estimation is the fundamental content and central issue in most public video monitoring systems, and it is also a hot spot in computer vision area. Recent state-of-the-art method is based on image processing. Traditional methods can be divided into two main directions, one is based on pixel statistics, and the other is based on texture analysis. In our paper, we combine these two methods and cut the image pixels into small blocks, for low density blocks we adopt the rapid pixel statistics, and for high density blocks we use the combination of pixel statistics and texture analysis to get better estimation. Experiments show that we get more precise result.