Backlight Unit Excessive Dimming Based on Perceptual Image Analysis
- Authors
- Yang, Seungwoo; Shin, Hyungsik; You, JaeHee
- Issue Date
- Feb-2021
- Publisher
- MDPI
- Keywords
- Backlight dimming; Bright sensitivity; Contrast response function; Human visual system; LCD; Perceptual image quality
- Citation
- APPLIED SCIENCES-BASEL, v.11, no.3, pp.1 - 17
- Journal Title
- APPLIED SCIENCES-BASEL
- Volume
- 11
- Number
- 3
- Start Page
- 1
- End Page
- 17
- URI
- https://scholarworks.bwise.kr/hongik/handle/2020.sw.hongik/15628
- DOI
- 10.3390/app11031039
- ISSN
- 2076-3417
- Abstract
- A method to maximize backlight dimming of liquid crystal displays (LCDs) based on human visual system are proposed to minimize power consumption of display panels. Based on images, the proposed method optimizes global, local, and red, green, blue (RGB) backlight dimming by enhancing dimming about 12% on average for various images while maintaining tolerable degradation of perceived image qualities in pixel saturation areas. The method considers and utilizes the brightness sensitivity and contrast response functions of human visual system using the mean luminance and the contrast of an image, which are mathematically modelled to allow optimization for display panels and application areas. A simulator that can calculate various dimming cases with the evaluations of numerical and perceptual image qualities as well as power consumption amount is introduced. With pattern and real photo images, the degree of power savings and the preservation of image quality of the proposed method are verified to outperform conventional approaches with high scores of the mean opinion score (MOS) and the structural similarity index measure (SSIM) over 0.97 while saving more than 10% of power dissipation.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - College of Engineering > School of Electronic & Electrical Engineering > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/hongik/handle/2020.sw.hongik/15628)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.