基于纹理复杂度的自适应分数阶微分算法  

Adaptive Fractional Differential Algorithm Based on Texture Complexity

作　　者：汪成亮[1] 乔鹤松[1] 陈娟娟[2] 

机构地区：[1]重庆大学计算机学院,重庆400030 [2]重庆师范大学计算机与信息科学学院,重庆400047

出　　处：《计算机工程》2012年第7期177-178,181,共3页Computer Engineering

基　　金：国家自然科学基金资助项目(61004112);中国博士后科学基金资助项目(20080430750)

摘　　要：图像分数阶微分算子具有较强的纹理细节信息增强能力,但最佳分数阶微分的阶数需要人为指定。为此,分析传统的分数盒维计算方法并对其进行改进,提出一种基于纹理复杂度的自适应分数阶微分算法。选择可以表示纹理细节复杂程度的分数维作为参数,自适应确定微分的阶数。实验结果表明,改进算法提取图像边缘的效果较好。Fractional differential operator can enhance the high frequency signals of the image,while retaining the low frequency profile information,so it has a strong detail enhancement capability.But the best order of fractional differential needs to be specified by the researchers,affecting the actual application of the fractional differential.This paper analyzes the disadvantage of the algorithm and proposes the improved method.It chooses the fractal dimension,which can express the complexity of the detail of the texture,as a parameter adaptive to determine the order of differential.Experimental results show that the improved algorithm has better result in the edge extraction of the image.

关 键 词：分数阶微分 分数维 纹理复杂度 纹理增强 边缘提取 

分 类 号：TN911.73[电子电信—通信与信息系统]