Anomaly prediction over human crowded scenes via associate-based data mining and k-ary tree hashing

Abstract

Anomaly detection and behavioral recognition are key research areas widely used to improve human safety. However, in recent times, with the extensive use of surveillance systems and the substantial increase in the volume of recorded scenes, the con ventional analysis of categorizing anomalous events has proven to be a difcult task. As a result, machine learning researchers require a smart surveillance system to detect anomalies. Tis research introduces a robust system for predicting pedestrian anomalies. First, we acquired the crowd data as input from two benchmark datasets (including Avenue and ADOC). Ten, diferent denoising techniques (such as frame conversion, background subtraction, and RGB-to-binary image conversion) for unfltered data are carried out. Second, texton segmentation is performed to identify human subjects from acquired denoised data. Tird, we used Gaussian smoothing and crowd clustering to analyze the multiple subjects from the acquired data for further estimations. Te next step is to perform feature extraction to multiple abstract cues from the data. Tese bag of features include periodic motion, shape autocorrelation, and motion direction fow. Ten, the abstracted features are mapped into a single vector in order to apply data optimization and mining techniques. Next, we apply the associate-based mining approach for optimized feature selection. Finally, the resultant vector is served to the k-ary tree hashing classifer to track normal and abnormal activities in pedestrian crowded scenes.