ABNORMAL GROUP BEHAVIOR DETECTION FOR OUTDOOR ENVIRONMENT

Computer Science Enineering

The main objective of this paper is to detect an abnormal group event that occurs in the given input video streams. This is the most popular and one of the challenging tasks in Computer vision domain. The proposed method is based on image descriptor and the classification approach. The optical flow computation is the first step in proposed method and this step can be followed by HOFO descriptor generation. Publically available dataset is used to evaluate the proposed method. To detect the abnormal moving queues we are using features named as HOFO descriptor. The classification approach involves SVM (Support vector machine) method. We apply this method on various Benchmark Dataset to detect abnormal group event. The proposed method gives the effective and accurate results.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.