Application of Combined Local Object Based Features and Cluster Fusion for the Behaviors Recognition and Detection of Abnormal Behaviors

In this paper, we propose a novel framework for behaviors recognition and detection of certain types of abnormal behaviors, capable of achieving high detection rates on a variety of real-life scenes. The new proposed approach here is a combination of the location based methods and the object based ones. First, a novel approach is formulated to use optical flow and binary motion video as the location based feature. Next, the spectral clustering is implicated to categorize the similar behavioral features, and a new cluster fusion method which combines the obtained results of the clustering with the two lateral features is also proposed here. Then, in each cluster, the velocity and the trajectory are used as the object based features. In addition, the hidden Markov model is used as the behavior model. The most important outcome of this paper is that with the help of the mentioned object based features, we can detect the abnormal behaviors which cannot be identified using the previously reported location based features. Finally, a framework that performs abnormal behavior detection via statistical methods is presented.