Novel Multi-Pedestrian Detection and Tracking Framework Combining Machine and Deep Learning Schemes to Anisotropic Diffusion-Based Models

An automatic multiple pedestrian detection and counting framework is introduced in this research paper. The proposed system combines successfully several computer vision and nonlinear anisotropic diffusion-based models. Its detection component performs a boosted cascade classifier-based walking person localization process that is followed and improved by a segmentation of the detected sub-images, which is performed applying a novel partial differential equation (PDE) - based geodesic active contour (GAC) scheme. The obtained pedestrian detections are then counted successfully in the analyzed traffic video by applying a tracking-by-detection scheme using a deep learning-based feature extraction to them. The people detection and tracking simulation results that are finally discussed here demonstrate the effectiveness of the described framework.