Low light combining multiscale deep learning networks and image enhancement algorithm
Journal Title: Modern Innovations, Systems and Technologies - Year 2022, Vol 2, Issue 4
Abstract
Aiming at the lack of reference images for low-light enhancement tasks and the problems of color distortion, texture loss, blurred details, and difficulty in obtaining ground-truth images in existing algorithms, this paper proposes a multi-scale weighted feature low-light based on Retinex theory and attention mechanism. An image enhancement algorithm is proposed. The algorithm performs multi-scale feature extraction on low-light images through the feature extraction module based on the Unet architecture, generates a high-dimensional multi-scale feature map, and establishes an attention mechanism module to highlight the feature information of different scales that are beneficial to the enhanced image, and obtain a weighted image. High-dimensional feature map, the final reflection estimation module uses Retinex theory to build a network model, and generates the final enhanced image through the high-dimensional feature map. An end-to-end network architecture is designed and a set of self-regular loss functions are used to constrain the network model, which gets rid of the constraints of reference images and realizes unsupervised learning. The final experimental results show that the algorithm in this paper maintains high image details and textures while enhancing the contrast and clarity of the image, has good visual effects, can effectively enhance low-light images, and greatly improves the visual quality. Compared with other enhanced algorithms, the objective indicators PSNR and SSIM have been improved.
Authors and Affiliations
Xia Yu, Lin Bo, Chen Xin
Keywords
Related Articles
- EP ID EP711099
- DOI 10.47813/2782-2818-2022-2-4-0215-0232
- Views 107
- Downloads 0
Thermocatalytic gas sensor based on palladium nanoparticles synthesized by spark ablation
We have demonstrated the possibility of fabricating thermocatalytic gas sensors with a catalytic layer based on palladium nanoparticles synthesized by spark ablation using palladium ingots with a purity of 99.96 wt. % as...
Application of the modified ant colony algorithm for solving the problem of scheduling of distributed enterprises
The task of scheduling of distributed enterprises is to assign tasks to geographically remote enterprises and to determine a convenient work schedule for each of the enterprises. The goal of solving this problem is to mi...
Detailed development of the design of a double-acting double-sided moldboard body for rotary ploughs
The article is devoted to the problems of developing designs of double-sided mouldboard bodies for rotary ploughs, using the methods of geometrical modeling. A brief analysis of designs of double-sided mouldboard bodies...
Structure and properties of sintered aluminum-based alloys produced by extrusion
This article examines the structure and properties of extruded sintered aluminum-based alloys. Two methods for producing alloys have been studied: activated liquid-phase sintering and intensive hot processing. A comparis...
Performance analysis of a linter machine with a new inter-saw gasket
A new design of the inter-saw gasket is proposed in the article; it makes it possible to reduce the weight of the saw cylinder of a linter machine to 1.5 kg. The results of experimental studies (under production conditio...