Abstract
In recent years, the development of stereo displays progresses fast and related products are put out time to time to offer human a wholly new visual experience. Although stereo display has become a leader of markets, it still doesn't popularize because of the discomfort in viewing stereo displays. This subject has been discussed in many researches, but there are no integral standards proposed nowadays. Therefore, the purpose of this study was to establish a well standard on viewing stereoscopic images by ergonomics evaluations. It is expected to find out the best combination of independent variables so that the viewers may experience visual fatigue-free while watching stereo displays. Two stages of experiments were conducted to evaluate the discomfort from viewing static and dynamic stereoscopic images by both subjective and objective assessment methods in this study. The first stage focused on static stereoscopic images. The effects of five factors, display luminance, ambient illumination, average of disparity, brightness contrast ratio, and viewing time, on psychophysical visual fatigue, subjective comfort, and objective fatigue were investigated simultaneously by Taguchi Methods. Then, the second stage experiment with the same assessment for range of disparity, velocity, and brightness contrast ratio of dynamic stereoscopic images was conducted by mixture design method. The results showed that the interaction effect between ambient illumination and average of disparity was moderate on Critical Flicker Fusion (CFF) and on eye fatigue, between ambient illumination and viewing time was significant on eye fatigue and moderate on spectral power of high frequency component of accommodative microfluctuations (HFC) and nausea, and between average of disparity and viewing time was moderate on HFC and on eye fatigue. Moreover, viewing stereoscopic images under high ambient illumination was better in terms of objective fatigue, but under low ambient illumination was better in terms of subjective comfort. According to the results of static and dynamic experiments, images with low brightness contrast ratio were recommended. Furthermore, by Taguchi Methods, we got lots of information of variables, and the optimal combination for viewing stereo displays could be selected in the integrity based on different requirements.