Anomaly Detection With Particle Filtering for Online Video Surveillance

With growing security threats, many online and offine frameworks have been proposed for anomaly detection in video sequences. However, existing techniques are either computationally very expensive or lack desirable accuracy. This research work proposes a novel particle filtering based framework which detects frames with anomalous activities upon the posterior probability of sequence. The method also regions frames. We propose prediction measurement models to accurately detect Novel model likelihood assigning weights these particles proposed. These efficiently utilise variable sized cell structure creates sub-regions scenes Furthermore, they extract information from frame form size, motion location features. is tested on UCSD LIVE datasets compared state-of-the-art algorithms literature. algorithm outperforms state-of-the art terms reduced Equal Error Rate (EER) comparatively lesser processing time.