Performance Analysis of Spatial Color Information for Object Detection Using Background Subtraction

Object detection, which is a primary yet important step in applications including video surveillance, gets difficult when the video of outdoor environment is influenced by illumination and weather changes. Background Subtraction is the method frequently used for in such cases. In this paper, enhanced BS method namely Extended Frame Differencing Method (EFDM) and Extended Histogram Differencing Method (EHDM) using spatial color information are proposed to improve the accuracy and computational efficiency of object detection in real time and outdoor environment. The effects of different spatial color information for the proposed methods are compared and analyzed. The spatial information used in the analysis includes RGB, HSV, CIELab, CIELuv, YCrCb color models. The results show that EFDM and EHDM present better results with HSV spatial color information.