摘要
光照不均匀、光学显微镜的放大作用、光学衍射等使微视觉图像存在动态范围大、整体像素灰度值偏低、微细结构与背景对比度低、信噪比低等不足.为增强图像质量,文中利用对比度设计高斯平滑卷积核的尺度优化准则以合理估计光照分量,提出了尺度优化的Retinex增强算法,并用该算法对模拟图像和实拍的同轴光光强变化的微视觉图像进行增强.结果表明,该增强算法在压缩动态范围内既能提高对比度,又能较好地保留图像细节信息,增强效果优于同态滤波算法.
Due to the existence of non-unifiorm illumination, magnification effect of optical microscope and optical diffraction, etc. , micro-vision images are of large dynamic range, low overall pixel gray value, low contrast of fine structure to background and low signal-to-noise ratio. In order to improve the image quality, a Retinex enhaneemenf algorithm with scale optimization is proposed. This algorithm uses the contrast to design an optimization criterion of Gauss smoother convolution kernel scale for the reasonable estimation of illumination component. The algorithm is then applied to the enhancement of simulated images and micro-vision images generated under various coaxial light intensities. The results show that the proposed algorithm is superior to the homomorphic fihering one because it improves the contrast and reserves the image details within the dynamic compressing range.
出处
《华南理工大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2013年第2期7-11,共5页
Journal of South China University of Technology(Natural Science Edition)
基金
国家自然科学基金重大研究计划项目(91223201)
NSFC-广东省自然科学联合基金资助项目(U0934004)
华南理工大学中央高校基本科研业务费专项资金资助项目(2012ZP0004)
关键词
显微图像
图像增强
RETINEX算法
光照不均匀
尺度优化
microscope image
image enhaneement
Retinex algorithm
non-uniform illumination
scale optimization
