Detecting moving objects, ghosts, and shadows in video streams

TLDR

Background subtraction methods are widely used for moving object detection in video applications, but accurately modeling and updating the background while handling shadows remains a major challenge. This work proposes a general-purpose method that integrates statistical assumptions with object-level knowledge of moving objects, ghosts, and shadows. The method processes pixels belonging to moving objects, ghosts, and shadows differently, employing color information for background subtraction and shadow detection to enhance segmentation and selective background updates. Experiments show the approach is fast, flexible, and precise, achieving high pixel accuracy and rapid adaptation to background changes.

Abstract

Background subtraction methods are widely exploited for moving object detection in videos in many applications, such as traffic monitoring, human motion capture, and video surveillance. How to correctly and efficiently model and update the background model and how to deal with shadows are two of the most distinguishing and challenging aspects of such approaches. The article proposes a general-purpose method that combines statistical assumptions with the object-level knowledge of moving objects, apparent objects (ghosts), and shadows acquired in the processing of the previous frames. Pixels belonging to moving objects, ghosts, and shadows are processed differently in order to supply an object-based selective update. The proposed approach exploits color information for both background subtraction and shadow detection to improve object segmentation and background update. The approach proves fast, flexible, and precise in terms of both pixel accuracy and reactivity to background changes.