Probabilistic BPRRC: Robust Change Detection against Illumination Changes and Background Movements

Abstract

This paper presents PrBPRRC (Probabilistic Bipolar Radial Reach Correlation), a change detection method that is robust against illumination changes and background movements. Most of the traditional change detection methods are robust against either illumination changes or background movements; BPRRC is one of the illumination-robust change detection methods. We introduce a probabilistic background texture model into BPRRC and add the robustness against background movements and foreground invasions such as moving cars, walking pedestrians, swaying trees , and falling snow. We show the superiority of our PrBPRRC under the environment with illumination changes and background movements by using public datasets: ATON Highway data, Karlsruhe traffic sequence data, and PETS 2007 data.