[1]孙利杰,赵景秀,郑美珠,等.一种新的基于小波包与人类视觉系统的边缘检测算法[J].南京师大学报(自然科学版),2010,33(04):134-138.
　Sun Lijie,Zhao Jingxiu,Zheng Meizhu,et al.A New Edge Detection Algorithm Based on Wavelet Packet and Human Visual System[J].Journal of Nanjing Normal University(Natural Science Edition),2010,33(04):134-138.
点击复制

一种新的基于小波包与人类视觉系统的边缘检测算法()
分享到：

《南京师大学报（自然科学版）》[ISSN:1001-4616/CN:32-1239/N]

卷:
第33卷
期数:
2010年04期
页码:
134-138
栏目:
计算机科学
出版日期:
2010-12-20

文章信息/Info

Title:
A New Edge Detection Algorithm Based on Wavelet Packet and Human Visual System
作者:
孙利杰;赵景秀;郑美珠;高忠;
曲阜师范大学计算机科学学院, 山东日照276826
Author(s):
Sun Lijie; Zhao Jingxiu; Zheng Meizhu; Gao Zhong
School of Computer Science,Qufu Normal University,Rizhao 276826,China
关键词:
小波包;  人类视觉系统;  中值滤波;  小波包系数;  零交叉点
Keywords:
w ave let packet;  HVS;  them edian filter ing;  the wave le t packe t co efficients;  zero-crossing
分类号:
TP391.41
摘要:
针对小波变换的局限性和人类视觉系统的敏感性不同,提出了一种新的基于小波包与人类视觉系统的边缘检测算法.首先对加噪的载体图像应用中值滤波进行图像的平滑处理并用小波包对图像的高频和低频部分进行分解;然后根据人类视觉系统模型的特性进行小波包系数的选择,并利用Log算子进行边缘的零交叉点检测;最后通过二维小波包重构函数得到图像近似部分的边缘.理论数据与实验结果表明,应用文中提出的边缘检测算法,通过与经典的Roberts、Sobel、Prewitt和Log 4种边缘检测算子比较,不但更符合人类的认知,而且在含有丰富细节和微小变化的区域, 新方法更具有应用价值.
Abstract:
To recognize the lim it o fw avelet transform and the sensitive diversity of hum an v isual system, an approach us ing a new edge detection algor ithm based on w avelet packe t and hum an v isua l system is proposed. F irstly, them ed ian fi-l tering is easy to remove im age no ise, and the deno ising im age is transfo rm ed by w av elet packet; The edge detection reg ions a re by se lecting thew ave let packet coeffic ients based on the characteristic o f hum an v isua lmode,l and then useLog a lgo rithm to se lec t the zero-crossing; The detected edge image is obta ined by reverse w avelet packe t. The expe rim en tal resu lts show that the a lgo rithm compared w ith Roberts、Sobel、Prew itt and Laplacian of Gauss ian edge detec tion a lgor ithm s not only agrees mo re w ith hum an?? s recogn ition, but a lso can get an even be tter applica tion on some reg ions inc lud ing abundant loca l deta ils and som e tiny areas that change.

参考文献/References:

[ 1]Lee J, H ara lick R, Shap iro L. M orpho log ic edge detection[ J]. IEEE Trans on Robotics and Automa tion, 1987, 3( 2): 142- 156.
[ 2]刁麓弘, 于斌, 李华. 一种基于菲涅尔衍射的边缘检测算法[ J] . 计算机学报, 2006, 12( 29) : 2 138-2 142.
[ 3]H ou Z J, W e iG W. A new approach to edge detection[ J]. Pattern Recognition, 2002, 35( 7): 1 559-1 570.
[ 4] T remb la is B, Augereau B. A fast m ult-i sca le edg e detec tion a lgo rithm [ J]. Pattern Recogn ition Lette rs, 2004, 25( 26): 603-618.
[ 5]Zhang L, Bao P. Edge de tection by scalem ultiplication in wave let dom ain[ J] . Pa ttern Recognition Letters, 2002, 23( 14): 1 771-1 784.
[ 6]王竹青, 王珂, 袁国良, 等. 多层次自适应空间系数高斯小波图像边缘检测[ J] . 中国图像图形学报, 2009, 14 ( 7): 1 347-1 353.
[ 7] Rakesh R R, Chaudhur i P, M urthy C A. Thresholding in edge detection: A statistica l approach [ J]. IEEE T ransactions on Im age Process ing, 2004, 13( 7) : 927-936.
[ 8]Kon ish i S, Yu illeA L, Coughlan JM, et a.l Sta tistica l edg e detection: Lea rning and eva luating edge Cues[ J]. IEEE Transac tions on Pattern Ana lysis andM ach ine Inte llig ence, 2003, 25( 1) : 57-74.
[ 9]王磊, 杨付正, 常义林, 等. 基于空间复杂度掩盖的边缘检测[ J]. 中国图像图形学报, 2008, 1( 13): 100-103.
[ 10]张德丰. M a tlab小波分析[M ]. 北京: 机械工业出版社, 2009.
[ 11]董彬, 林小竹, 徐凤. 基于人类视觉系统的小波域数字水印算法[ J]. 计算机工程, 2006, 24( 32): 138-143.

备注/Memo

备注/Memo:
基金项目: 山东省自然科学基金( ZR2009GM009 )、山东省博士基金. 通讯联系人: 孙利杰, 硕士研究生, 研究方向: 图像处理与数字水印. E-mail:dpslj-good@ 126. com
更新日期/Last Update: 2013-04-08