计算机应用研究2017，Vol.34Issue(1)：283-286,301,5.DOI:10.3969/j.issn.1001-3695.2017.01.064
基于自适应分数阶微分的引导滤波及其应用
Guided image filter and application based on adaptive fractional-order differential
摘要
Abstract
Guided image filter has the function of edge preserving and smoothing,but the traditional guided image filter can lead to details losing and blurring.In order to keep high-frequency information and compute low-frequency structurally,this pa-per proposed guided image filter based on adaptive fractional-order differentiation.Firstly,it defined the fractional-order dif-ferential mask based on the fractional-order differential theory.Secondly,it designed fractional-order differential order function according to the image gradient,local information entropy and local variance,which could detecte the changes of texture and gradient.Then,it transferred local conditions to the guided image that ensured not only preserving egdes and denosing,but al-so keeping texture.The experiment results show,the proposed algorithm is skilled at preserving egdes and texture,and it can improve the accuracy of face recognition based on PCA and SVM in image preprocessing.关键词
保边平滑/分数阶微分/掩模/局部方差/人脸识别Key words
edge-preserving and smoothing/fractional-order differential/mask/local variance/face recognition分类
信息技术与安全科学引用本文复制引用
谢伟,游敏,周玉钦..基于自适应分数阶微分的引导滤波及其应用[J].计算机应用研究,2017,34(1):283-286,301,5.基金项目
国家自然科学基金资助项目（61501198，41101425，41201413）；湖北省自然科学基金面上资助项目（2014CFB461）；武汉市青年科技晨光计划资助项目（2014072704011248）；华中师范大学中央高校基本科研业务费资助项目 （）
