模糊集与非线性增益相结合的自适应图像增强算法  

Image enhancement algorithm based on fuzzy sets and nonlinear gain

作　　者：全永奇 李太君[1] 邓家先[1] 谢凯明[1] 

机构地区：[1]海南大学信息科学技术学院,海口570228

出　　处：《计算机应用研究》2016年第1期311-315,共5页Application Research of Computers

基　　金：海南省自然科学基金资助项目(613155);海南省地方标准实施与应用的关键技术研究项目(SF201455);海南省科技兴海专项资金资助项目(XH201311)

摘　　要：针对部分图像对比度低、细节模糊不清等问题,提出了一种模糊集增强与非线性增益相结合的自适应图像增强算法,使用双正交小波变换对原始图像进行分解,低频子带系数采用改进的模糊集增强算法,以提升图像的整体对比度;对高频子带,先采用贝叶斯萎缩法估计噪声与信号的阈值,再使用一种非线性增益函数增强图像细节并抑制噪声。对算法中影响增强效果的关键参数进行了研究,并提出了一种模糊集增强算子的阈值选取算法,能够实现不同图像自适应参数选择;将信息熵作为非线性增益函数的参数选取准则,并针对算法中排序算法运算量过多导致算法时间过长的情况,提出了一种替代求解方法,极大地提高了算法效率。对算法进行仿真,结果表明:算法能够有效提升对比度、增强图像细节并抑制噪声,可以明显改善图像的视觉效果,具有参数自适应、算法效率高等优点。Aiming at the low contrast and low detailed nature of partial image, this paper proposed an adaptive image enhance- ment algorithm based on fuzzy sets and nonlinear gain function ,which decomposed the input image into low-frequency sub-band and high-frequency sub-bands through DWT（ discrete wavelet transform）. Low-frequency sub-bands could improve the contrast degree through fuzzy sets enhancement method while the high-frequency sub-bands coefficients enhanced by using Bayesian shrinkage threshold and nonlinear gain function, which could suppress the image noise and effectively highlight the image de- tail. Through seriously researches on the parameters which heavily influence the efficiency of image enhancement, this paper proposed a kind of threshold selection for fuzzy enhancement operator, which could achieve parameter selection adaptively for different images. This algorithm took information entropy as criterion for parameters selection. It proposed an approximate solu- tion to improve the efficiency aiming at the sorting algorithm which was so ruath-intensive that time-consuming. Simulation re- suits indicate that the proposed algorithm can enhance the image contrast degree, suppress the noise, sharpen structural and de- tails, leading to noticeable improvement in visual quality, it also has advantages of parameters selection adaptively and higher efficiency.

关 键 词：图像增强 模糊集增强 非线性增益 自适应增强 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TP301.6[自动化与计算机技术—计算机科学与技术]