基于全局双约束的矿井尘雾图像增强方法
Mine dust image enhancement method based on global double constraints
冀常鹏 1贺丽娜 1代巍1
作者信息
- 1. 辽宁工程技术大学 电子与信息工程学院,辽宁 葫芦岛 125105
- 折叠
摘要
为提高煤矿尘雾图像的可观测性,提出一种基于全局双约束的 Retinex 算法的尘雾图像增强算法(GCFCDL-Retinex).首先,将输入图像进行内外循环,训练聚类和稀疏双重约束下的过完备字典,对图像中的噪声分量进行抑制;然后,通过Retinex算法对照度分量和反射分量进行估计及提取,并对提取的照度分量进行自适应Gamma校正;最后输出增强后的图像.研究结果表明:在煤矿井下的复杂环境中,所提出的图像增强算法能够有效提高矿井下尘雾图像的对比度和清晰度,去除真实粉尘,同时抑制图像光晕、边缘模糊的现象,增强后的图像色彩自然,视觉效果明显提升.研究结论为矿井下视频监控清晰化的工程应用提供理论依据.
Abstract
In order to improve the observability of dust haze images in coal mines,a dust haze image enhancement algorithm based on global double constraint Retinex algorithm was proposed.Firstly,the input image is cycled inside and outside,and the overcomplete dictionary under the double constraints of clustering and sparse is trained to suppress the noise component in the image.Then,the Retinex algorithm was used to estimate and extract illuminance and reflection components,and adaptive Gamma correction was performed for the extracted illuminance components.The final output of the enhanced image.The research results show that the proposed image enhancement algorithm can effectively improve the contrast and clarity of dust fog image in the complex underground environment of coal mine,remove the real dust,and suppress the phenomenon of image halo and edge blur.The enhanced image color is natural,and the visual effect is significantly improved.The research conclusion provides a theoretical basis for the engineering application of mine video monitoring sharpening.
关键词
图像增强/图像去噪/稀疏约束/聚类约束/Retinex算法Key words
image enhancement/image denoising/sparse constraint/clustering constraint/Retinex algorithm引用本文复制引用
基金项目
辽宁省教育厅基本科研项目(LJKMZ20220677)
出版年
2024
国家科技期刊平台
NSTL
万方数据