Crowd Analysis System for Images of CCTV Camera
Vishakha L. Bansod1, Asha Ambhaikar2
1Vishakha L. Bansod, PhD Scholar, Kalinga University, Naya Raipur; India.
2Dr. Asha Ambhaikar, Professor and Dean Student welfare, Kalinga University, Naya Raipur; India
Manuscript received on January 02, 2020. | Revised Manuscript received on January 15, 2020. | Manuscript published on January 30, 2020. | PP: 1113-1118 | Volume-8 Issue-5, January 2020. | Retrieval Number: E6283018520/2020©BEIESP | DOI: 10.35940/ijrte.E6283.018520
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: It is important to monitor and analyze the crowd for peaceful event organization and minimum number of persons killed or injured in a stampede, war or accident in public or religious places. Ultimately, crowd management strategies should be adopted for crowd safety. Detecting the abnormal crowd behaviour is the interesting research area. Traditionally to monitor the crowd closed circuit television is used. Video analyses are used in many application areas. This work is a paper of traditional approach and is a study of different attributes of crowd which include crowd counting and estimation of density, detection of crowd motion. It also presents the study about behaviour understanding of crowd and crowd tracking. We have also presents crowd analysis using different deep learning methods. This paper showed results of crowd analysis using convolutional neural network and different recurrent neural network training model.
Keywords: Abnormal Behaviour, Convolution Neural Network (CNN), Crowd Analysis, Deep Learning, Recurrent Neural network (RNN),
Scope of the Article: Deep Learning.