基于双边滤波的单尺度Retinex图像增强算法

Single-scale Retinex Image Enhancement Based on Bilateral Filtering

提出了一种基于双边滤波的单尺度Retinex图像增强方法,在有效增强图像的同时,很好地保持了图像边缘及高亮部分的细节.采用双边高斯滤波来估计图像亮度,同时对原始图像进行局部对比度增强,有效地消除了光晕现象并增强了局部对比度.通过Weber定律在增强过程中对图像亮度进行控制,防止增强后的图像过亮,保持图像高亮部分的细节.实验结果表明:该方法增强效果明显,适用于灰度图像和彩色图像,并能防止颜色失真,使灰度和彩色图像达到较好的视觉效果.

A novel method for image enhancement with single-scale Retinex based on bilateral filtering is proposed, which remains the details of image edge and high brightness part in the process of image enhancement. The bilateral Gauss filter is utilized to estimate the illumination image. Simultaneously the proposed algorithm strengthens the partial contrast gradient to the original image, removes the halo phenomenon and strengthens the partial contrast gradient effectively. On the other hand, the brightness of image is controlled through Weber＇ s law in strengthening process to prevent the image to be excessively bright and maintain the details of high brightness part. The experimental result indicates that the enhancement effect of the proposed algorithm is obvious. The method, which is applied to both the gray and color images, can prevent color distortion, and make the target images achieve better visual effect.