基于光照分量提取的光照不均匀图像校正-《集美大学学报（自然科学版）》

文章信息/Info

Title:: Illumination Nonuniform Image Correction Based on Illumination Component Extraction

Author(s):: TAO Sheng1; HU Yue2; (1.School of Science，Jimei University,Xiamen 361021,China；2.Chengyi University College，Jimei University，Xiamen 361021,China)

Keywords:: illumination nonuniform image; illumination component; power transformation; correction; Lab space

摘要:: 针对光照不均匀的灰度图像，在频域采用多尺度高斯函数提取出光照分量，并对光照分量进行对数变换，再根据提取出的光照分量和变换后的光照分量对原始图像的灰度值进行幂次变换，从而实现光照不均匀灰度图像的校正。对于真彩色RGB图像，先将该图像从RGB空间转换到Lab空间，然后对L分量的灰度值进行幂次变换校正，最后将校正后的图像从Lab空间转换回RGB空间。Matlab仿真结果表明，频域提取光照分量比空域处理速度更快，幂次变换校正后的图像质量得到明显改善，与伽马函数校正相比，在客观评价标准和主观视觉效果两方面均能够取得更好的结果。

Abstract:: For the gray image with uneven illumination，the multi-scale Gaussian function is used to extract the illumination component in the frequency domain，and the logarithmic transformation is carried out on the illumination component.According to the extracted illumination component and the transformed illumination component，the power transformation is carried out on the gray value of the original image，so as to realize the correction of uneven illumination gray image.For a true color image，the image is converted from RGB space to Lab space first，then the gray value of the L component is corrected by power transformation，and finally the corrected image is converted from Lab space to RGB space.Matlab simulation results show that the processing speed of light component extracted in frequency domain is faster than that in spatial domain，and the image quality after power transform correction is improved significantly.Compared with gamma function correction，it can achieve better results in both objective evaluation standard and subjective visual effect.