Chinese Journal of Liquid Crystals and Displays, Volume. 37, Issue 11, 1467(2022)
Adaptive color optimization algorithm of image engine in display system
Fig. 1. Standard convolutional are replaced by depthwise convolution and pointwise convolution
Fig. 3. Schematic diagram of the principle of dilated convolution.（a）1-dilated convolution，each element has a receptive field of 3×3；（b）2-dilated convolution，each element has a receptive field of 5×5；（c）3-dilated convolution，each element has a receptive field of 7×7.
Fig. 8. Mean scores of original and optimized images for image quality
Table 1. System configuration
View tableView in ArticleTable 1. System configuration
系统环境 版本信息 操作系统 Ubuntu 16.04 LTS CPU Intel Core i7-8700k GPU NVIDIA GeForce GTX 1080Ti 内存 32 GB Python版本 3.5
Table 2. Comparison of objective evaluation indexes in test set between our method and others
View tableView in ArticleTable 2. Comparison of objective evaluation indexes in test set between our method and others
Index Traditional
method
DPED FPIE MIRNet Proposed PSNR 22.81 22.74 22.44 24.02 26.82 SSIM 0.78 0.76 0.79 0.83 0.82 NIMA Score 4.23 4.25 4.32 4.76 4.83
Table 3. Grading standard of visual perception experiment
View tableView in ArticleTable 3. Grading standard of visual perception experiment
Grade Image quality 5 Excellent 4 Good 3 Fair 2 Poor 1 Bad
Table 4. Results of ANOVA analysis for different factors on image quality score
View tableView in ArticleTable 4. Results of ANOVA analysis for different factors on image quality score
Factor Image quality score df F Sig. Process 1 809.280 <0.001 Image 59 7.509 <0.001 Image×Process 59 2.309 <0.001
Get Citation
Copy Citation Text
Bo-wen ZHANG, Zhen-ping XIA, Bo ZHOU, Yu SONG, Feng-yun MA, Yi BAI. Adaptive color optimization algorithm of image engine in display system[J]. Chinese Journal of Liquid Crystals and Displays, 2022, 37(11): 1467
Category: Research Articles
Received: Apr. 23, 2022
Accepted: --
Published Online: Nov. 3, 2022
The Author Email: Zhen-ping XIA (xzp@usts.edu.cn)
