Abstract: In this paper a fast Retinex-based color image enhancement method is proposed.Global and local adaptation of the human visual system are simulated.A global dynamic range adjustment is performed first.Then Mean Shift filtering is performed on the original image in the illuminant estimation step，and the spatial-varying illuminant is eliminated base on the center/surround hypothesis.Finally steps of color restoration and post-processing are carried out.Experimental results demonstrate that this method can efficiently reduce halo artifacts and color of the original image is better preserved.An accelerated implementation of Mean Shift is employed.Thus this method is much faster than previous methods.

Key words: image enhancement, Retinex, Mean Shift, halo artifact, color preservation

摘要： 提出了一种快速Retinex彩色图像增强方法，模拟了人类视觉系统全局和局部自适应性。首先对图像的亮度范围做全局调整；然后在光照估计步骤中对亮度进行Mean Shift滤波，用于其后光照影响的消除；最后对增强结果进行颜色恢复并进行了后处理。实验结果表明，方法能够有效压缩图像的动态范围，克服光照不均的影响，有效去除了光晕现象，且方法的颜色保持性较同类方法更好。使用了Mean Shift加速算法，方法运行速度快于此前提出的方法。

关键词: 图像增强, Retinex, Mean Shift, 光晕现象, 颜色保持