雾天环境下低维图像模糊区域快速增强仿真

摘要

The quality of image collected under severe weather such as fog and haze is badly damaged due to the atmospheric scattering.It needs to make rapid enhancement to the hazy region of low dimensional image under the foggy weather environment.However,the current hazy region enhancement method cannot reflect the situation of gray space variation of pixel point neighborhood,it leads to bad effect of image enhancement.This paper proposes a hazy region rapid enhancement method of low dimensional image under foggy weather environment based on the fuzzy logic.Firstly,the integral low dimensional image from the gray plane is mapped to the fuzzy feature plane and the fuzzy entropy is used to work out the gray space change rate of image pixel.Then,the fuzzy entropy of all pixel points in the low dimensional image is obtained and the equivalence transformation model of image hazy region integrated with the region feature mapping theory is built to extract the image feature of hazy region.Moreover,the horizontal projection membership mean value of image hazy region feature is defined as the fuzzy threshold value corresponding to the transition point.Finally,the self-adaption fuzzy enhancement in the horizontal direction is achieved.The simulation results show that the method has better enhancement effect than the existing method.It improves the image degradation phenomena during the foggy weather apparently.%在雾、霾等恶劣天气下采集的图像会因为大气散射的作用而导致图像质量严重受损,为了解决上述问题,需要进行雾天环境下低维图像模糊区域快速增强.但是采用当前方法进行图像模糊区域增强时,无法反映图像模糊区域像素点邻域灰度值空间变化的情况,存在图像增强效果差的问题.为此提出一种基于模糊逻辑的雾天环境下低维图像模糊区域快速增强方法,先将整体的低维图像从灰度平面映射到模糊特征平面上,采用模糊熵计算出图像像素的灰度空间变化率,得到低维图像中的全部像素点的模糊熵值,并融合于区域特征映射原理建立图像模糊区域等价变换模型,提取模糊区域图像特征,将图像模糊区域特征的水平投影隶属度均值定义为渡越点对应的模糊阈值,对模糊区域进行水平方向自适应模糊增强.实验结果表明,所提方法增强效果好,极大的改善了雾天时图像的退化现象.