基于分数阶微分的边缘检测

被引:49
作者
杨柱中 [1 ]
周激流 [2 ]
黄梅 [1 ]
晏祥玉 [1 ]
机构
[1] 四川大学电子信息学院
[2] 四川大学计算机学院
关键词
分数阶微分; 边缘检测; 微分阶数; 掩模模板; 峰值信噪比(PSNR);
D O I
10.15961/j.jsuese.2008.01.029
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
为了提高图像边缘提取的信噪比,更有效和准确检测图像边缘,由信号的微分特性得出分数阶微分算子较传统1阶和2阶微分算子具有更高的信噪比,然后根据经典的G-L分数阶微分定义推导出的分数阶差分方程,构建了近似的分数阶Tiansi模板。实验证明,基于分数阶微分的边缘提取算子,可以有效提取边缘和有比传统的算子更高的信噪比。
引用
收藏
页码:152 / 157
页数:6
相关论文
共 18 条
[11]  
Fractional differentiation for edge detection. Mathieu,B.,Melchior,P.,Oustaloup,A.,Ceyral,Ch. Signal Processing . 2003
[12]  
Statisticaledge detection:learning and evaluating edge cues. Konishi S,Yuille A L,Coughlan J M,et al. IEEE Transactions on Pattern Analysis and Machine Intelligence . 2003
[13]  
分数阶微积分在现代信号分析与处理中应用的研究[D]. 蒲亦非.四川大学 2006
[14]  
Fractional differential equations. Podlubny I. Mathe-matics in Science and Engineering . 1999
[15]  
Multiscale Edge Detection on Range Images via Normal Changes. Chwen Jye Sze,,Hong Yuan,Mark Liao,H,ai Lung Hung,,Kuo Chin Fan,,Jun Wei Hsieh. IEEE Transactions On Circuits And Systems—Ii: Analog And Digital Signal Processing . 1998
[16]  
Directional filtering in edge detection. Andrew P P. IEEE Transactions on Image Processing . 1998
[17]  
A computational approach to edge detection. Canny J. IEEE Trans-actions on Pattern Analysis and Machine Intelli-gence . 1986
[18]  
Adaptive determination of filter scales foredge detection. Jeong H,Kim C. IEEE Trans on Pattern Analysis andMachine Intelligence . 1992