Study Of Visual Health Based On Optical Properties And Image Attributes Of Display Systems

Display devices such as TVs,smartphones,and tablet computers have become an indispensable part of modern people’s work and life.The demands on their performance are increasing.While novel displays such as high dynamic range and wide color gamut have improved display quality and viewing experience,the visual health hazards might be induced have also received increasing attention.Based on the perceptual aspects of human visual system,the study explored the comprehensive influence of optical properties and image attributes of display systems on visual discomfort and visual fatigue,based on factors including white field luminance,dark field luminance,color saturation,sharpness,chromaticity and spectral components.Mathematical models were established for evaluating visual discomfort and visual fatigue levels,which provide a reference for optimising the design and manufacture of display systems and improving the visual experience.The main contents of this thesis are as follows:The influence of white field luminance and color saturation of display system and the spatial structure of images on visual discomfort was studied.For the high dynamic range and wide color gamut displays,the external symptoms and internal mechanisms of visual discomfort were analysed,through subjective evaluations and changes in cortical hemodynamic responses.The results showed that moderate white field luminance(400 cd/m~2)and color saturation(73%NTSC color gamut)relieved visual discomfort and reduced the cognitive load of human visual system.Visual discomfort caused by the images with unnatural spatial distributions was more serious.The results also confirmed that the cortical hemodynamic responses were highly correlated to visual discomfort,and less correlated to subjective perceptions such as display quality or preference.The influence of chromaticities of ambient illumination and display system on visual discomfort was studied.Based on the spectral sensitivities of retinal photoreceptors,ambient environments with varied chromaticities and spectra were designed and optimised,to investigate the effects of S-cones and intrinsically photosensitive retinal ganglion cells activation on visual discomfort.A combination of eye-tracking and subjective questionnaires was used to investigate the changes in visual discomfort and visual cognitive properties caused by different chromatic combinations of ambient illumination and display system.The results showed that the ambient-display combinations with inconsistent chromaticities resulted in worse subjective preference,more unnatural color perception,and more obvious symptoms of visual discomfort.These evaluations were accompanied by more frequent blinks and shorter inter-blink intervals.Furthermore,it was found that retinal illuminance differences in ambient illumination had limited contribution to visual discomfort.The effects of white field luminance,dark field luminance,color saturation,and sharpness of display system on visual discomfort were studied,and a visual discomfort model based on multi-dimensional information fusion was proposed.By controlling these properties and recording the viewer’s subjective ratings,eye movements,and electrocardiogram,a multi-dimensional information fusion model based on Dempster-Shafer evidence theory was established.The model was optimized by reducing dimensions,assigning weight coefficients,and setting risk coefficients,to improve the performance effectively.The model demonstrated high sensitivity,specificity,precision,and accuracy,illustrating the feasibility of using multi-dimensional information to assess visual discomfort.The influence of smartphone displays’spectra on visual fatigue was studied,and a dynamic visual fatigue model was proposed.Based on the photobiological parameters of percieved illumination by retinal photoreceptors,the influence of spectral components was quantified.Subjective questionnaires,optical indicators,electrocardiogram,and electroencephalogram were combined to measure visual fatigue states comprehensively.The results showed that the shift of blue light peaks to longer wavelength reduced the stimulus on S-cones and intrinsically photosensitive retinal ganglion cells,thus relieved visual fatigue.Excessive light intensity in the middle and long wavelength stimulated M-and L-cones more,and affected the tear film stability.Based on the experimental results,a dynamic multi-dimensional information fusion model was established by characterising the accumulation of visual fatigue with time through Markov chains.The prediction performance of the model was great,and the sensitivity,specificity,precision and accuracy were high.