基于MSR的雾天图像清晰化算法研究

研究雾天图像清晰化的问题,需提高图像增强的均匀性。针对雾天情况下,由于雾气的遮挡使得拍摄图像对比度降低,图像局部细节处不清晰,传统的直方图均衡化的雾天图像清晰化方法虽然能够增强图像对比度,但是图像局部细节增强不足,造成图像增强均匀性不高的问题。提出一种MSR的雾天图像清晰化算法,通过Sigmoid函数对图像作映射,拉伸图像的对比度,然后利用MSR算法,将图像小波分解为高频分量和低频分量,对高频分量取绝对值最大运算,低频分量加权平均,并避免了对图像进行全局直方图均衡化造成的图像增强不均匀,局部细节增强不足的问题。实验证明,提出的算法能够将雾天图像均匀增强,得到高清晰的图像,取得了满意的效果。

Research Fog- Degraded Images Clearness Algorithm Based on the MSR

Research the fog-degraded images clearness problem to improve the uniformity of the image enhancement.In foggy weather conditions,fog makes images contrast low and image local details not clear.The traditional fog-degraded images clearness algorithm based on histogram equalization can enhance image contrast,but the enhancement of image local details is insufficient,causing image enhancement uniformity is not high.This paper proposed a fog-degraded image clearness algorithm based on MSR.Through the Sigmoid function,the image was mapped and the image contrast was stretched.Then,the MSR algorithm was used to get the image wavelet decomposition for high frequency components and low frequency components.The high frequency components were calculate for the largest absolute value and the low frequency components were calculate for the weighted average.Experiments show that the algorithm can enhance the image contrast and get high-definition images.