基于分块DCT同态滤波的彩色图像增强算法  

Block-DCT based homomorphic filtering algorithm for color image enhancement

作　　者：张亚飞[1] 谢明鸿[2] 

机构地区：[1]南京工程学院通信工程学院,江苏南京211167 [2]南京电子技术研究所,江苏南京210039

出　　处：《计算机工程与设计》2013年第5期1752-1756,共5页Computer Engineering and Design

基　　金：南京工程学院科研基金项目(YKJ200907)

摘　　要：对于光照不均的彩色图像,由于传统的同态滤波器没有充分考虑图像的空间局部特性,同态滤波后的图像在局部对比度增强效果上,不能令人满意。为此,结合DCT变换的优点,提出了基于分块DCT变换的同态滤波算法,可以获得很好的局部对比度增强效果。为了消除分块滤波引起的块效应,提出了基于重叠块的块效应消除方法。同时,为了保留原始图像的色彩信息,提出将图像从RGB颜色空间转换到HSV颜色空间,仅对亮度分量进行基于DCT变换的同态滤波及块效应的去除,色调和饱和度分量保持不变。实验结果表明,该方法不但有效改善了光照不均对图像的影响,而且使图像色彩保持良好,图像中的细节更为清晰,性能上优于传统的同态滤波。For the color image of non-uniform illumination, because the traditional homomorphic filtering doesn＇t fully take into account the local characteristics of the image, the local contrast enhancement effect of the image after homomorphic filtering is not satisfactory. To this end, combined with the advantages of the DCT, the homomorphic filtering algorithm based on block discrete cosine transformation （DCT） is proposed, which can provide a better enhancement effect on the local contrast. In order to eliminate the block effect caused by block filtering, the method of block effect elimination based on overlapping blocks is presented. Meanwhile, in order to retain the color information of the original image, the image is converted from the RGB color space to HSV color space. The only luminance component is enhanced by the homomorphic filtering based on block-DCT and block effect removal. The hue and saturation components are invariant. The experimental results show that this method is not only effective in improving the impact of the non-uniform illumination of the image, and to maintain a good image color and clearer image details. The performance of homomorphic filtering based on block-DCT is superior to traditional one.

关 键 词：彩色图像增强 光照不均 离散余弦变换 同态滤波 块效应 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]