Abstract
Automatic surveillance of abnormal events is a major unsolved problem in city management. By successful implementation of automatic surveillance of abnormal events, a significant amount of human resources in video monitoring can be economized. One solution to this application is computer vision technology. This approach utilizes an image processing algorithm to extract specific features and then uses discriminator algorithms to give an alert. In this paper, we propose to apply a particle filter-based algorithm to feature series extracted from videos in order to give alerts when abnormal events occur. The whole process consists of feature series generation and particle filter tracking. To represent the features of a video, an L2-norm extractor is designed based on the optical flow. Then, the particle filter keeps track of these feature series. The occurrence of abnormal events will cause the shift of feature series and a large error in PF tracking. This, in turn, will allow computers to understand and define the occurrences of anomalies. Experiments on UMN dataset show that our algorithm reaches 90% accuracy in frame-level detection.
Highlights
Pedestrian gathering is a principal cause of many serious accidents such as crushing and trampling accidents
We use a particle filter to do prediction of L2-norm of image sequence, which assumes that normal event frame is easy to predict while anomaly is hard to
We summarize the contributions to our paper as follows: 1) Inspired by related works [14]–[16], we propose a particle filter-based prediction pipeline for anomaly detection, which tries to tackle the abnormal events by comparing their L2-norm with their expectation
Summary
Pedestrian gathering is a principal cause of many serious accidents such as crushing and trampling accidents. It was based on an assumption that normal event frame is easy to predict while anomaly is hard to Another classical approach for this problem is to use probability-based statistical models. Instead of hand-crafted features, deep learning methods often use convolutional neural networks to extract features that carry target information, such as motion and appearance These features will be sent to an auto-encoder and will be enforced to reconstruct the normal behavior with small error. We summarize the contributions to our paper as follows: 1) Inspired by related works [14]–[16], we propose a particle filter-based prediction pipeline for anomaly detection, which tries to tackle the abnormal events by comparing their L2-norm with their expectation. The results show that our method can distinguish some specific abnormal events from normal ones, as well as detect the start and the end of anomaly in videos
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
