Work

Research of Low-illumination Image Enhancement Algorithm Based on Retinex Theory

Year: 2023
Type: conference-paper
Abstract: The image enhancement based on Retinex theory can simultaneously improve brightness, details and color retaining, which has a comprehensive improvement in image quality. It is the focus of research on low-illumination image enhancement. The scale parameters of conventional Retinex theory-based methods are too single, making it unable to enhance various types of pictures adaptively. To solve the problem above, an algorithm based on multi-scale Retinex with color restore, which has adaptive weights is proposed. The algorithm calculates the optimal scale parameters and adaptive weights using the detail information of the sub blocks. And it defines the normalized brightness of each pixel in the image, avoiding the problem of excessive enhancement in the normal brightness area. The effectiveness is proved in comparison with some methods proposed.
Author: Shiqing Chen
Language: English
Open Access status: bronze
OpenAlex ID: W4399253901
FWCI:
Cites: 8
Cited by:
Related to: 10