Moving object detection with background subtraction and shadow removal

This paper presents a moving object detection method with background subtraction and shadow removal in the RGB color space. For background subtraction, the metrically trimmed mean is employed as a robust estimate of the background model and the mean absolute deviation (MAD) is adopted as a scale estimate. A cascading chromaticity difference estimator, brightness difference estimator, and spatial analysis are explored to discriminate the shadow and the moving object. Our experimental results validate the good performance of the proposed method.