Low-power Display Optimization Technology Based On Visual Perception Characteristics For Self-luminescence Display

Displays,as one essential kind of carrier for information transmission,play an important role in all areas of industry and daily life.According to the light emitting principle,the current display technologies can be divided into two types,i.e.the self-luminescence display technologies and the non-self-luminescence display technologies.LCD(Light Crystal Display)is a type of mainstream non-self-luminescence displays,but with the development of display technologies,self-luminescence displays,especially OLED(Organic Light Emitting Diode)displays,gradually occupy an important position in this field due to their various advantages,appearing a tendency to be the third generation of displays by replacing the current LCD.With the trend of much larger screen OLED displays,the energy consumption issue has been more and more prominent,which has also aroused extensive research interests of scholars at home and abroad.With regard to energy consumption of self-luminescence displays,deep analyses were focused on domestic and foreign research status.Then,a low-power display optimization algorithm for self-luminescence displays based on visual perception characteristics was proposed in this study,aiming at reducing the energy consumption through image analysis and processing without much quality loss of the original images.The experimental comparison results demonstrated the effectiveness of the proposed algorithm.The algorithm can be applied to self-luminescence displays,which can effectively reduce the energy consumption and achieve the goal of display effect optimization.The main contents of this thesis are as follows.(1)Based on the domestic and foreign status of energy-saving display algorithms of selfluminescence displays,the concept of the algorithm in this study was designed,and the characteristics of visual perception of human vision were analyzed,especially the characteristics of visual saliency,relative brightness perception and memory color,which served as the theoretical basis of the algorithm in this thesis.At the same time,the color model used in the algorithm,the indexes for evaluating the algorithm and another energysaving display algorithms were introduced.(2)To address the deficiencies of existing researches in low-power display and improving display quality,an low-power display algorithm based on visual perception characteristics was proposed,by which the human visual system's perception of relative brightness and visual saliency were fully considered.The brightness of each pixels was compressed by a non-linear function according to its visual significance results,i.e.the saliency,which can guarantee smooth transition of the image brightness.The relationship between energy-saving rate and energy-saving coefficient in the algorithm was also discussed,so that the proposed algorithm can meet different energy-saving requirements.The effect of the algorithm was analyzed through objective indexes and subjective evaluation.The results show that the algorithm of this thesis can maintain the brightness of the visual subject while saving energy,without leading to color distortion.The image display effect can be still guaranteed for the energy-saving rate of 20% to 30%.(3)In order to improve the display effect,a display optimization algorithm was formed by implementing two kinds of improvement in terms of skin color protection and blue sky color enhancement.Firstly,the existing detection algorithms of skin color area and blue sky area were studied and actualized.In view of their advantages and disadvantages,the algorithms in this study were designed and actualized respectively for detecting skin color area and blue sky area.The simulation results of 24 Kodak Standard Test Images show that the proposed detection algorithms have greatly improved the false detection rate and the missing detection rate.The missing detection rate of the blue sky area almost reaches at 0.After that,the protection method for the skin color region and the enhancement method for the blue sky region were proposed based on the memory color theory,as the optimization of the lowpower display algorithm,which could significantly improve the effect of the algorithm when processing typical image contents.(4)The implementation process of the low-power display optimization algorithm was summarized,and the comparison of this algorithm with the other three energy-saving algorithms of OLED displays were performed on the same energy-saving rate(i.e.20%).Subjective experiment was carried out and the calculation of z-score standard fraction was used to analyze experimental results.The results show that the proposed low-power display optimization algorithm can maintain the original image display effect and reduce the energy consumption at the same time.The mean PSNR value between original image and optimized version is 27.The standard fractions of subjective experiments on five standard test images are all positive.For some individual images,due to the use of background brightness compression and enhancement techniques based on memory color theory,the display effect of images processed by the proposed algorithm is even better than their corresponding original images,under the premise that energy saving is achieved.