Performance analysis of a backlight dimming method using weighted mean-square-error based on joint edge-saliency characteristics

摘要

In this paper, a new backlight dimming method using weighted mean-square-error (MSE) based on joint edge-saliency characteristics is proposed. In contrast to conventional backlight dimming methods that cannot accurately evaluate the saturation error and determine the optimal clipping point to achieve the best trade-off between image quality and power consumption, the proposed method, with consideration of human visual characteristics, analyzes the joint edge-saliency characteristics based on spatio-temporal saliency map and edge-strength map of the input images, and weights the MSE accordingly. It can therefore evaluate the allowable saturation error and determine the optimal clipping point. Simulation results show that the proposed method outperforms other benchmark methods in the case of overall consideration of image quality and power consumption. Additionally, the proposed method maximizes the power reduction rate while successfully preserving image details in the regions of interest and maintaining high perceived image quality. Our results also show that the average computation time of the proposed algorithm was reduced by up to 30.184%, 19.162% and 8.027%, respectively, compared with I~2GEC, SMVA~2 and SPBD algorithms.