基于同态滤波的平面视觉图像色彩增强算法

Color Enhancement Algorithm of Plane Vision Image Based on Homomorphic Filtering

针对因光照环境导致图像色块像素分布不均、失真以及视觉效果不佳等问题,在同态滤波的基础上提出平面视觉图像色彩增强算法。根据RGB色彩空间的分量变化缺陷,将需要处理的图像色彩空间转换为HSV形式,并分块处理。对所有图像在DCT变换的基础上进行同态滤波计算,并通过图像构成条件创建对应的同态滤波模型。采用压缩低频、提升高频的方法获取清晰图像,构造巴特沃思型的同态滤波器,缓解图像色彩不均匀和视觉效果失真的情况,最后运用基于HSV的可见度函数去除图像中存有的块效应。仿真结果表明,所提算法能够有效地消除图像的块效应,保护图像边缘信息,具有较高的适用性和有效性。

Due to uneven distribution and distortion caused by illumination environment,a color enhancement algorithm for planar vision image based on homomorphic filtering was put forward.According to the defect of the change of component in RGB color space,the image color space to be processed was transformed into HSV and processed in blocks.Based on DCT transform,the homomorphic filtering was performed on all images.According to image constitution conditions,the homomorphic filtering model was built.The method of compressing low frequency and increasing high frequency was adopted to obtain clear image.Moreover,the Butterworth homomorphic filter was constructed to alleviate the uneven color and visual distortion.Finally,the visibility function based on HSV was used to remove the block effect in image.Simulation results prove that the proposed algorithm can effectively eliminate the blocking effect and protect the edge information of image,so it has high applicability and effectiveness.