Optical health analysis of visual comfort for bright screen display based on back propagation neural network. (November 2020)

• Record Type: Journal Article
• Authors: Wang, Kun; Ho, Chun-Heng; Tian, Chunpeng; Zong, Yan
• Is Part Of: Computer methods and programs in biomedicine. Volume 196(2020)
• Publication Date: 2020-11
• Subjects: Visual comfort -- Back propagation neural network -- Display ergonomics -- Optical health -- Human factors; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2020.105600
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: We applied BP neural network for analyzing LCD visual comfort. LCD brightness can be influenced by physiological and environmental factors. Our model can successfully predict the optimal visual level of LCDs. We optimally adjust LCD brightness and color temperature according to users. Improving visual comfort of bright LCDs enhances optical health. Abstract: Background: The visual comfort of liquid crystal display (LCD) is the subjective evaluation of the user. It is a multi-dimensional and multi-factor problem, which is affected by the luminous characteristics of the LCD screen, the physiological factors of the user, and some other environmental factors. Methods: Based on the theory of visual comfort under the guidance of ergonomics, this paper adopts a combination of objective measurement and subjective evaluation to obtain objective data such as blink frequency and pupil size changes, and subjective evaluation data on screen parameters. Correlation analysis was used to screen subjective and objective data, and an LCD visual comfort evaluation using the back propagation (BP) neural network was constructed with the aim of a concise evaluation of the LCD's own light-emitting characteristics, user's physiological factors, and environmental factors. Results: After testing, the model can successfully predict the optimal visual level of the screen. After training, the relative error between the predicted value of visual comfort and the actual evaluation value is mostly within 10%. Based on this model, the display brightness and color temperature control system combined with the ambient light sensor can automatically adjust the brightness of the screen and the temperature of color parameters in correlation to user's gender, age, and ambient light changes to achieve the effect of improving visual comfort. Setting and user parameter adjustment provide a new method. The maximum adjustment error of the system after testing is 5.378%. Conclusion: Our proposed technique can serve as a useful analysis platform for understanding and evaluating the visual comfort of the bright LCD screen at home or in the workplace, and enhancing optical health of humans.