The Improved Codebook Algorithm Combined Updating Background And Changing Brightness Range

The establishment of a Codebook background model in RGB color space results in high computational complexity, low segmentation quality and poor immunity power. We proposed the improved brightness range Codebook algorithm under YUV space to overcome the problem of irrational brightness range definition of the algorithm, as well as the background updating method which combined a two-layer Codebook model with a short-sliding window. Firstly, the Codebook background model was built under YUV color space based on the characteristics of luminance and chrominance separation. Secondly, the brightness range was redefined using the average of the Y component of the codeword to tackle the problem of the irrational brightness range definition of the background model. Lastly, a two-layer Codebook model and a short-sliding window were combined to update the background in the process of moving object detection in order to improve the anti-jamming capability. The experiments demonstrated that the improved brightness range can be updated gradually with background changes. Comparing with the original algorithm, this method can update the background effectively, and improve the detection accuracy.