融合小波变换和模糊集的东巴古籍图像增强算法  

Image enhancement algorithm based on wavelet transform and fuzzy set

作　　者：夏欣雨 朱春梅[1] 吴国新[1] XIA Xinyu;ZHU Chunmei;WU Guoxin(Key Laboratory of Modem Measurement ＆ Control Technology Ministry of Education, Beijing Information Science ＆ Technology University, Beijing 100192, China)

机构地区：[1]北京信息科技大学现代测控技术教育部重点实验室,北京100192

出　　处：《北京信息科技大学学报（自然科学版）》2018年第3期47-51,56,共6页Journal of Beijing Information Science and Technology University

基　　金：国家社科基金重大项目(12&ZD234)

摘　　要：图像增强是东巴古籍图像数字化的重要环节。针对东巴古籍图像对比度低、图像模糊等特点,提出一种融合改进的小波分析理论和模糊集理论的图像增强算法。该算法选用小波变换对东巴古籍图像进行二级小波分解,得到图像的低频系数和高频系数;对图像的高频系数采用改进的阈值滤波算法处理,对低频系数采用模糊集理论增强;通过小波逆变换重构图像达到对比度增强和背景抑制的目的。实验结果表明,该算法结合了模糊集增强和小波阈值滤波的优点,在增强图像细节的同时不仅抑制了噪声,且图像的对比度显著提高。Image enhancement is an important part of the image processing of Dongba manuscripts.For the low contrast and blurred image of Dongba manuscripts image,an image enhancement algorithm based on improved wavelet analysis theory and fuzzy set theory is proposed. Firstly, the wavelet transform is used to decompose the Dongba manuscripts image into three levels,and the corresponding low-frequency components and high-frequency ones are obtained. Then,the high frequency coefficients of images are processed by improved threshold filtering algorithm,and the low frequency coefficients are enhanced by fuzzy sets. Finally,wavelet transform is used to reconstruct the image so that the contrast enhancement and background suppression can be achieved. Experimental results show that the proposed algorithm combines the advantages of fuzzy set enhancement and wavelet threshold filtering,suppresses the noise while enhancing the details,and improves the image contrast.

关 键 词：小波变换 阈值滤波 模糊集 图像增强 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]