Detection and Segmentation of Occluded Vehicles Based on Skeleton Features

Abstract

Vehicle occlusion is an important problem for vehicle detection even in the high-definition camera. The difficulty is to obtain individual vehicle from one blob when some vehicles merged together in the blob in the foreground mask. We present a method that can detect the number of occluded vehicles in one blob and the segmentation points based on the effective skeleton nodes, and then segment occluded vehicles in one blob associated with concave spots. Experiments are performed on traffic video sequences captured at two different locations, and the results appear to have good accuracy.