Research on People Counting Method Based on the Density Map

With the rapid development of science and technology and the rise of deep learning, in recent years, image based people counting has become a hot research topic in the field of computer vision. The method based on the density map retains the spatial information of the crowd in the picture, and on this basis, the network model of deep learning training is also used to improve the accuracy of the people counting. This paper mainly analyzes the traditional counting methods, introduces the common datasets used in people counting. We mainly conducted research and analysis on recent years people counting methods of generating high quality population density map through CNN method, the end to end convolutional neural network model surprised us with the accuracy of the people counting. Finally, this paper prospects the future research trend of people counting. Qian Li | Peng Wei | Minjuan Shang "Research on People Counting Method Based on the Density Map" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-4 | Issue-6 , October 2020, URL: https://www.ijtsrd.com/papers/ijtsrd33599.pdf Paper Url: https://www.ijtsrd.com/engineering/computer-engineering/33599/research-on-people-counting-method-based-on-the-density-map/qian-li