Video Anomaly Identification [ a Statistical Approach ]

T his article describes a family of unsupervised approaches to video anomaly detection based on statistical activity analysis. Approaches based on activity analysis provide intriguing possibilities for region-of-interest (ROI) processing since relevant activities and their locations are detected prior to higher-level processing such as object tracking, tagging , and classification. This strategy is essential for scalability of video analysis to cluttered environments with a multitude of objects and activities. Activity analysis approaches typically do not involve object tracking, and yet they inherently account for spatiotem-poral dependencies. They are robust to clutter arising from multiple activities and contamination arising from poor background subtraction or occlusions. They can sometimes also be employed for fusing activities from multiple cameras. We illustrate successful application of activity analysis to anomaly detection in various scenarios, including the detection of abandoned objects, crowds of people, and illegal U-turns. INTRODUCTION Video camera networks are ubiquitous. Over 30 million cameras produce close to 4 billion hours of video footage per week in the United States alone [1]. This proliferation is taking place since, unlike other sensing modalities, visible-light cameras provide excellent space-time resolution, long capture range, wide field of view, and low latency. Consequently, large-scale camera networks can provide pervasive, wide-area monitoring to protect national infrastructures. Such a protection necessitates careful video analysis in suitable data space (such as pho-tometric, behavior, and attitude) and with appropriate objectives, from counting objects and object localization to identifying suspicious movement of people or assets. Our focus in this article is on identifying anomalous activity in space and time. Currently, this type of video analysis