Wang, Jing and Xu, Zhijie (2016) Spatio-temporal Texture Modelling for Real-time Crowd Anomaly Detection. Computer Vision and Image Understanding, 144. pp. 177-187. ISSN 1077-3142
Abstract

With  the  rapidly  increasing  demands  from surveillance  and  security  industries,  crowd  behaviour analysis has become one of the hotly pursued video event detection frontiers within the computer
vision  arena  in  recent  years.  This research has  investigated  innovative  crowd  behaviour detection 
approaches based on statistical crowd features extracted from video footages. In this paper, a new crowd  video  anomaly  detection  algorithm  has  been developed  based  on  analysing the extracted spatio‐temporal  textures. The algorithm has been designed  for  real‐time applications by deploying low-level statistical features and avoiding complicated machine learning and recognition processes. In the experiments, the system has been proved as a valid solution for detecting anomaly behaviours
without  strong  assumptions  on  the  nature  of  crowds,  for  example,  subjects  and  density.  The 
developed prototype shows improved adaptability and efficiency against chosen benchmark systems.

Information
Library
Documents
[img]
CVIU-14-714R1.pdf
Restricted to Repository staff only

Download (1MB)
Statistics
Add to AnyAdd to TwitterAdd to FacebookAdd to LinkedinAdd to PinterestAdd to Email