DENSITY CLASSIFICATION AT LARGE SCALE CROWD USING SPATIO-TEMPORAL LOCAL BINARY PATTERN
Author(s):
Mangalagouri V.B, Prof. B. Naveen Kumar
Keywords:
Abstract
Expanding overall populace is prompting thick swarm (crowd) gathering at open spots. Because of mass social occasion at enormous scale, swarm related calamity has been often happened. Density is the status of group which is fundamental to arrange in visual observation framework fundamentally for security perspectives. The greater part of the current methods deals with recognition and following of people. Due to less pixels per focus on, discovery and following of people is a complex task in thick crowd situations. Therefore we propose system which introduces a novel procedure for huge scale swarm thickness grouping controlled by dynamic surface investigation. This methodology comprises of an intrigue focuses location pursued by spatio-temporial element extraction. A RIST-LBP design is proposed to remove dynamic surface of the moving swarm. Further, a multi-class bolster vector relapse is received for swarm thickness characterization. Our methodology has the benefit of low computational multifaceted nature with high effectiveness in genuine world uses of video observation.
Article Details
Unique Paper ID: 148441
Publication Volume & Issue: Volume 6, Issue 2
Page(s): 42 - 45
Article Preview & Download
Share This Article
Join our RMS
Conference Alert
NCSEM 2024
National Conference on Sustainable Engineering and Management - 2024