改进直方图均衡和Retinex算法在灰度图像增强中的应用

Application of improved histogram equalization and Retinex algorithm in gray image enhancement

为了提高光照条件变化下的图像增强效果,提出一种改进直方图均衡和Retinex算法的图像增强方法。对于待增强灰度图像,通过理想低通滤波获得图像低频分量,采用改进直方图均衡进行动态范围优化,利用引导图像滤波代替Retinex算法的高斯滤波对图像的高频分量进行估计,并对估计的结果进行线性放大处理。实验结果表明,相对单尺度和多尺度Retinex算法以及改进的直方图均衡化算法,提出的方法从主观和客观评价方面都获得了更好的图像增强效果,有效提高了图像的视觉效果和可懂度,并具有较强的鲁棒性。

In order to improve the performance of image enhancement under different illumination condi- tions, an effective image enhancement method was proposed, which was based on the improved histogram equalization and Retinex algorithm. First, the low frequency components of the image were obtained us- ing the ideal low-pass filter and its dynamic range was optimized by the improved histogram equalization. Second, the high frequency components of the image were estimated by the guided image filtering, which replaces the Gaussian filter in Retinex algorithm. Finally, the estimated results were amplified by the linear processing method. Experimental results show that the performance of the proposed method was better than the traditional Retinex algorithm and improved histogram equalization algorithm by the subjective and objective evaluation. Furthermore, the proposed method effectively improves the image＇s visual effects and intelligibility with strong robustness.