Anomalies detection from video surveillance using support vector trained deep neural network classifier
by S. Giriprasad; S. Mohan; S. Gokul
International Journal of Heavy Vehicle Systems (IJHVS), Vol. 25, No. 3/4, 2018

Abstract: Intelligent video surveillance plays a crucial role in various applications for detecting the abnormal activities. The surveillance system uses many significant technologies for detecting the anomalies in different applications but it fails to manage the accuracy while detecting the anomalies from huge crowd. This paper introduces an effective image processing technology-based classifier for recognising and detecting the abnormality from the crowd effectively. Initially, the videos are captured using the surveillance camera, and the background has been subtracted by the robust background principal analysis method. After extracting the background from the image, the different principal bow sift descriptors are extracted. Subsequently the similar descriptors are grouped using the bee-based collaborative filtering approach. Finally, the anomaly classification is done by support vector machine training-based deep neural networks. Then the excellence of the system is evaluated by using the implementation results and the obtained results are compared with the traditional classifiers.

Online publication date: Mon, 24-Sep-2018

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Heavy Vehicle Systems (IJHVS):
Login with your Inderscience username and password:

    Username:        Password:         

Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email