关键词: 关键词: 低光照图像增强, 拉普拉斯金字塔, 图像处理, 卷积神经网络, Transformer

Abstract: Abstract: The quality of low-light images is poor and low-light image enhancement (LLIE) aims to improve the visual quality. Most of LLIE algorithms focus on enhancing luminance and contrast, while neglecting details. To solve this issue, a progressive enhancement algorithm for low-light images based on layer guidance was proposed, which enhanced images to a suitable illumination level and reconstructed clear details. First, to reduce the complexity and improve the efficiency, the low-light image was decomposed into several frequency components by Laplace pyramid (LP) decomposition. Secondly, since different frequency components exhibit pixel-level correlation, a Transformer-based fusion model and a lightweight fusion model were respectively proposed for layer guidance. The Transformer-based model was applied between the low-frequency and the lowest high-frequency components. The lightweight fusion model was applied between two neighbouring high-frequency components. By doing so, the layer guidance features were continuously fed to a higher layer and components were enhanced in a coarse-to-fine manner. Finally, the LP was used to reconstruct the enhanced image with normal illumination and clear details. The experimental results show, the proposed algorithm is 2.3 dB higher than DSLR(Deep Stacked Laplacian Restorer) on LOL(LOw-Light)-v1 dataset and 0.55 dB higher than UNIE(Unsupervised Night Image Enhancement) on LOL-v2 dataset. Compared with other state-of-the-art LLIE algorithms, this algorithm has shorter runtime and achieves significant improvement in objective and subjective quality, which is more suitable for real-world.

Key words: Keywords: low-light image enhancement, Laplace pyramid, image processing, Convolutional Neural Network, Transformer