A background subtraction algorithm for indoor monitoring surveillance systems

The use of the gray level intensity is a common practice for most of background subtraction algorithms due to speed matters in real time applications, and performance related considerations, yet using the RGB color representation could increase the efficiency of object detection thus the accuracy of the algorithm increases. In this paper, a non-parametric background subtraction algorithm based on samples modeling, adaptive threshold, and color layers combination is presented. The proposed framework showed an increase in performances regarding the accuracy and the robustness of the detection in indoor situations. The presented performance analysis supports the robustness of the algorithm to gradual illumination changes and ghost artifact.