Crowd Abnormality Detection Using Optical Flow and GLCM-Based Texture Features

Detection of abnormal crowd behavior is one the important tasks in real-time video surveillance systems for public safety places such as subway, shopping malls, sport complexes and various other gatherings. Due to high density crowded scenes, detection becomes a tedious task. Hence, analysis hot topic research requires an approach with higher rate detection. In this work, focus on management present end-to-end model analysis. A feature extraction-based using contrast, entropy, homogeneity, uniformity features determine threshold normal activity has been proposed paper. The measured terms receiver operating characteristic curve (ROC) & area under (AUC) UMN dataset compared methods literature prove its worthiness. YouTube sequences also used anomaly