METHOD AND APPARATUS FOR VIDEO SURVEILLANCE BASED ON DETECTING ABNORMAL BEHAVIOR USING EXTRACTION OF TRAJECTORIES FROM CROWD IN IMAGES

摘要

According to an embodiment of the present invention, a video surveillance apparatus comprises: a foreground extraction unit for extracting a foreground from video frames; a feature point trajectory extraction unit for extracting feature point trajectories represented by length information, coordinate information, and direction information based on feature points extracted from the foreground using the Kanade-Lucas-Tomasi algorithm; a trajectory selection unit for selecting feature point trajectories to determine crowd behavior with regard to each of the extracted feature point trajectories based on a length similarity, a coordinate similarity, and a direction similarity to other feature point trajectories; and a trajectory classification unit for classifying the selected feature point trajectories into a normal trajectory and an abnormal trajectory.