Hybrid tracking model and GSLM based neural network for crowd behavior recognition
Hybrid tracking model and GSLM based neural network for crowd behavior recognition作者机构:Department of Computer Engineering and ApplicationsGLA UniversityMathuraIndia
出 版 物:《Journal of Central South University》 (中南大学学报(英文版))
年 卷 期:2017年第24卷第9期
页 面:2071-2081页
核心收录:
学科分类:12[管理学] 080802[工学-电力系统及其自动化] 0808[工学-电气工程] 08[工学] 0810[工学-信息与通信工程] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 0806[工学-冶金工程] 081104[工学-模式识别与智能系统] 0805[工学-材料科学与工程(可授工学、理学学位)] 0703[理学-化学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术(可授工学、理学学位)]
主 题:crowd video crowd bohavior tracking recognition neural network gravitational search algorithm
摘 要:Crowd behaviors analysis is the‘state of art’research topic in the field of computer vision which provides applications in video surveillance to crowd safety,event detection,security,*** presents some of the works related to crowd behavior detection and *** crowd behavior detection,varying density of crowds and motion patterns appears to be complex occlusions for the *** work presents a novel crowd behavior detection system to improve these *** proposed crowd behavior detection system is developed using hybrid tracking model and integrated features enabled neural *** object movement and activity in the proposed crowded behavior detection system is assessed using proposed GSLM-based neural *** based neural network is developed by integrating the gravitational search algorithm with LM algorithm of the neural network to increase the learning process of the *** performance of the proposed crowd behavior detection system is validated over five different videos and analyzed using *** experimentation results in the crowd behavior detection with a maximum accuracy of 93%which proves the efficacy of the proposed system in video surveillance with security concerns.