Towards Intelligent Crowd Behavior Understanding through the STFD Descriptor Exploration

Realizing the automated and online detection of crowd anomalies from surveillance CCTVs is a research-intensive and application-demanding task. This research proposes a novel technique for detecting crowd abnormalities through analyzing the spatial and temporal features of input video signals. This integrated solution defines an image descriptor (named spatio-temporal feature descriptor - STFD) that reflects the global motion information of crowds over time. A CNN has then been adopted to classify dominant or large-scale crowd abnormal behaviors. The work reported has focused on: 1) detecting moving objects in online (or near real-time) manner through spatio-temporal segmentations of crowds that is defined by the similarity of group trajectory structures in temporal space and the foreground blocks based on Gaussian Mixture Model (GMM) in spatial space; 2) dividing multiple clustered groups based on the spectral clustering method by considering image pixels from spatio-temporal segmentation regions as dynamic particles; 3) generating the STFD descriptor instances by calculating the attributes (i.e., collectiveness, stability, conflict and crowd density) of particles in the corresponding groups; 4) inputting generated STFD descriptor instances into the devised convolutional neural network (CNN) to detect suspicious crowd behaviors. The test and evaluation of the devised models and techniques have selected the PETS database as the primary experimental data sets. Results against benchmarking models and systems have shown promising advancements of this novel approach in terms of accuracy and efficiency for detecting crowd anomalies.