Low-light Image Enhancement Based on Convolutional Analysis Sparse Representation and Phase Congruency

doi:10.3969/j.issn.0372-2112.2020.01.022

您当前的位置：

首页 >

文章列表页 >

Low-light Image Enhancement Based on Convolutional Analysis Sparse Representation and Phase Congruency

更新时间：2025-07-08

- Low-light Image Enhancement Based on Convolutional Analysis Sparse Representation and Phase Congruency
- Acta Electronica Sinica Vol. 48, Issue 1, Pages: 180-188(2020)
- 作者机构：
  
  陆军炮兵防空兵学院信息工程系,安徽,合肥,230031
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China (No.61379105);Natural Science Foundation of Anhui Province (No.1908085MF208)
- DOI：10.3969/j.issn.0372-2112.2020.01.022
  CLC： TP391
- Published Online：25 January 2020，
  
  Published：2020
- 稿件说明：
移动端阅览
Low-light Image Enhancement Based on Convolutional Analysis Sparse Representation and Phase Congruency[J]. Acta Electronica Sinica, 2020, 48(1): 180-188.
DOI：

Low-light Image Enhancement Based on Convolutional Analysis Sparse Representation and Phase Congruency[J]. Acta Electronica Sinica, 2020, 48(1): 180-188. DOI： 10.3969/j.issn.0372-2112.2020.01.022.

摘要

针对低照度图像存在的对比度低、视觉效果差等问题，提出一种基于卷积分析稀疏表示和相位一致性的低照度图像增强方法.该方法基于Retinex模型，在估计照度图像时采用卷积分析稀疏表示进行约束，所用滤波器一部分人工设定，一部分由样本训练自动获得；在计算反射图像时利用单演相位一致性特征，施加相位一致性残余最小约束来恢复细节；通过联合约束并进行优化，得到的反射图像即为最终的增强结果.对大量低照度图像进行实验，并与当前先进方法相比，结果表明，本文方法不仅提高了图像的亮度与对比度，增强了细节，而且在多个客观评价指标上都优于其他方法.

Abstract

Low-light images suffer from low visibility and poor visual quality. To improve the quality of low-light images

a method based on convolution analysis sparse representation and phase congruency was proposed. This method is based on the Retinex model and improves the problem of insufficient constraints.More concretely

we used the convolutional analysis sparse representation whose filters were hand-crafted and learned from the input to estimate the illumination image. Then

by using the monogenic phase congruency

the reflection image was calculated via the phase congruency residual minimization to enhance weak details. Through joint constraints and optimization

the resulting reflection image served as the final enhancement result.Experiments on a number of challenging low-light images are presented to reveal the efficacy of our method and show its superiority over several state-of-the-arts on both subjective and objective assessments.

关键词

Keywords

references

Views

207

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A Robust Multi-Source Remote-Sensing Image Registration Method Based on Feature Matching

Medical Image Registration Based on Principal Phase Congruency

Low-Light Image Enhancement Method Based on U-Net Generative Adversarial Network

Related Author

LING Zhi-gang

LIANG Yan

CHENG Yong-mei

PAN Quan

SHEN He

LU Zhen-tai

FENG Yan-qiu

FENG Qian-jin

Related Institution

College of Electrical and Information Engineering, Hunan University

College of Automation, Northwestern Polytechnical University

College of Electrical and Information EngineeringHunan UniversityChangshaHunan 410082China

College of AutomationNorthwestern Polytechnical UniversityXi'anShaanxi 710072China

Key Lab for Medical Imaging of Southern Medical University,Guangzhou,Guangdong 5105151

⁰