Single Image Subpixel Rendering Using a Residual Network for Micro-LED Display

The increase in pixel density exacerbates the process difficulty and manufacturing cost of flat panel displays, especially for Micro-LED displays. To alleviate this problem, this paper proposes a novel single-image subpixel rendering method using a residual network for RGBG square-arranged flat panel subpixel displays. The proposed method uses a residual network to achieve nonlinear mapping from the original image to the rendered image and designs a perceptual error loss function based on the contrast sensitivity function (CSF) of the human eye to find an optimal balance between luminance resolution and chromatic aliasing. The experimental results show that the proposed method has significant advantages in rendering subpixel images, effectively controlling color errors while maintaining image clarity. Compared with existing subpixel rendering algorithms, the proposed method outperforms several evaluation metrics, including the zoomed peak-signal-to-noise-ratio (PSNRz) and perceptual error measuring (PEM), demonstrating its significant potential to improve visual quality.