摘要
针对单尺度Retinex算法会产生较严重纵向条纹噪声的问题,提出了一种混合图像增强算法。该算法利用模板去噪算法减少常规单尺度Retinex算法产生的纵向条纹噪声,并结合拉普拉斯算子增强图像细节,提高图像质量。采用实际综采工作面不同工况下的监控图像对该算法进行实验验证,结果表明,该算法具有图像清晰度高、对比度大与图像细节增强优的特点。
In view of problem of longitudinal stripe noise existed in single dimension Retinex algorithm,a mixed enhancement algorithm of image was proposed.The algorithm uses template denoising algorithm to reduce longitudinal stripe noise of conventional single scale Retinex algorithm,and adopts Laplace operator to enhance image details,so as to improve image quality.The method was verified by experiment using actual monitoring images under different working conditions of fully-mechanized coal mining face.The results show that the algorithm has characteristics of high image resolution,big contrast and optimal image detail enhancement.
出处
《工矿自动化》
北大核心
2016年第4期36-40,共5页
Journal Of Mine Automation
基金
中国博士后科学基金第58批面上资助项目(2015M581879)
关键词
综采工作面
图像增强
单尺度Retinex
纵向条纹噪声
拉普拉斯算子
fully-mechanized coal mining face
image enhancement
single dimension Retinex
longitudinal stripe noise
Laplacian operator
