In recent years, the invention of new display technologies not only has enriched the way and content of visual perception, but also has promoted the development of information society. However, with the widely use of various displays and the excessive reliance on information equipments, visual health has become a more and more important issue. Visual fatigue is widely regarded as one of the key factors influencing on visual health. Besides, with the rapid development of LED light sources, the requirement for the corresponding ambient illumination has become more and more critical. But for indoor lighting, LEDs have some problems which may affect people’s health, like glare and higher proportion of blue light. So the safe and comfortable issue is once again in the spotlight.
In this dissertation, the research for the visual fatigue caused by High-Dynamic Range (HDR) display and white LED light source has been carried out though the combination of subjective questionnaire, opthalmological parameters and physiological signals which are confirmed effectively.
The visual fatigue caused by long-term (8h) visual display terminal (VDT) work under fluorescent and LED luminaire has been investigated. The results show that visual fatigue becomes more serious with increasing time spent at VDT work under the two light sources. Compared to fluorescent luminaire, the LED luminaire produces significantly greater near point accommodation, blink amplitude and oxygen saturation but less “dry eyes” symptom, best corrected distance visual acuity and high frequency electrocardiogram power as time passed, but all are of small effect size. Although the participants carry out eight hours of VDT work under the two light sources, at the end the symptom “tired eyes” is rated as moderate while all the other reported symptoms are rated as slight. Thus, LED light source could be used in office lighting. On this basis, a model for evaluating or predicting visual fatigue under LED light sources has been developed. And a simplified model which also shows high accuracy is proposed as well. Using these models visual fatigue could be well evaluated, and long-term measurement can be avoided and instead, predicted by short-term evaluation results.
The influence of long-term (8h) light effect on the visual fatigue for VDT and paper reading tasks under different illuminance (300lx, 600lx, 1200lx) of LED light source has been explored. The results show that the visual fatigue increases with time. In comparison with paper reading tasks, VDT reading task could induce more severe visual fatigue. Within the range of 300 - 1200lx, the impact of illuminance on visual fatigue is small.
The effect of HDR and Standard-Dynamic Range (SDR) display on display quality and visual fatigue has been investigated. The results show that the image quality, image contrast, sharpness, immersion, preference, naturalness and comfort of HDR image are significant higher than SDR image’s. And the results of color saturation show the opposite. HDR display causes more “tired eyes” symptom (around the critical threshold of 0.05) and low frequency electrocardiogram power, compared with SDR display. That is, the impact of display format on visual fatigue is slight. The visual fatigue occurring after one hour of HDR display and SDR display is rather small. On that basis, the ambient illumination for watching HDR display is tried to be optimized. The results show that the optimized ambient illumination could alleviate the impact of the display on visual system effectively, but not for the physiology. By the above research, a visual fatigue recognition model based on Bayesian Network is proposed, which takes display format and ambient illumination as the variables of the input layer of the model; regards best corrected distance visual acuity, eye blink and low frequency electrocardiogram power as the variables of the output layer; and uses soberness and fatigue as the variables of the hidden layer. It could increase the accuracy of visual fatigue recognition.