基于Laplacian塔式分解的自适应图像模糊增强

Adaptive Image Fuzzy Enhancement Based on Laplacian Pyramid Decomposition

为解决部分数字图像对比度偏低、细节模糊等问题,提出了一种非线性拉伸与模糊增强相结合的自适应图像增强算法。使用Laplacian塔式分解对图像进行分解,高频子带系数采用贝叶斯萎缩阈值和非线性增益函数进行处理,以增强后高频子带系数信息熵为目标,自适应选取控制增益曲线形状参数和控制增益强度参数;低频子带系数采用模糊增强算法进行处理,对隶属度函数和模糊增强算子进行改进,并提出了一种模糊增强算子中阈值参数的自适应选取算法。实验结果表明,该算法能有效提高图像对比度和清晰度,突显图像细节信息,且实现了增强参数的自适应选择,更有利于图像的检测与识别。

In order to solve the problems of low contrast and definition in some digital images, an adaptive image enhancement algorithm combined nonlinear extension and fuzzy enhancement was proposed. An input image was de-composed into low-frequency sub-band and high-frequency sub-bands through Laplacian pyramid decomposition. Bayesian shrinkage threshold and nonlinear gain function were used for high-frequency sub-bands. The information entropy of coefficients of enhanced high-frequency sub-bands was regarded as a target, which aim to achieve the con-trol gain curve shape parameter and control gain intensity parameter adaptive selection. Coefficients of low-frequency sub-band were enhanced according to fuzzy enhancement method. The membership functions and fuzzy enhancement operator were improved and propose an adaptive algorithm which can achieve fuzzy enhancement operator threshold pa-rameter adaptive selection. Experimental results show that, the proposed algorithm can improve the contrast and defini-tion of images effectively, highlight the details of images and realize the adaptive selection of enhancement parameters. It is better for the detection and recognition of images.