.594 北京科技大学学报 2006年第6期 Mlach Intell,1989,11(7).674 [2]Nowak R D.Baraniuk R G.Wavelet-based transformations for nonlinear signal processing.IEEE Trans Signal Process, 1999,47(7):1852 [3]Dougherty E R.An introduction to morphological image pro- cessing-Bellingham:SPIE Press,1992 [4]Goutsias J.Schonfeld D.Morphological representation of dis- 图8形态小波在线检测出的裂纹图像 crete and binary images.IEEE Trans Signal Process,1991. 39(6):1369 Fig-8 Online detected crack image by morphological wavelet [5]Chatzis V.Pitas I.A generalized fuzy mathematical morphol- ogy and its application in robust 2 D and 3 D object represen- 3 结语 tation.IEEE Trans Image Process.2000.9(10):1798 [6]Lazzaroni F.Leonardi R.Signoroni A.High performance em- 由于形态小波是在数学形态学以及小波理论 bedded morphological wavelet coding.IEEE Signal Process, 基础上发展起来的,它具有数学形态学的形态滤 Lct,2003,10(10):293 波特性以及小波的检测突变信号的能力,因此在 [7]Heijmans HJA M.Goutsias J.Nonlinear multiresolution sig 图像检测中,可以用于检测一些在灰度值上存在 nal decomposition schemes:Part II morphological wavelets 着突变,并且存在着明显形态特征的目标,如表面 IEEE Trans Image Process,2000.9(11):1897 裂纹等.实验证明了形态小波变换方法对于检测 [8]Cha H.Chaparro L F.A morphological wavelet transform// International Conference on Systems Engineering in the Service 中厚板表面裂纹缺陷取得了很好的效果,检测的 of Humans:Systems.Man and Cybernetics.South Korea. 时间短,一般在几十毫秒左右,同时去除噪声的能 1993,501 力强, [9]Lam W K.Ip H R.Li C K.Approaches to decompose flat structuring element for fast overlapping search morphological 参考文献 algorithm//Proceedings of the 14th International Conference [1]Mallat S C.A theory for multi-resolution signal decomposi- on Pattern Recognition.1998:1461 tion:the wavelet representation.IEEE Trans Pattern Anal Application of morphological wavelet in surface crack defect detection of medium plates WU Guifang,XU Ke),XU Jinwu,YANG Chaolin),SUN Hao2),WANG Chunmei2) 1)Mechanical Engineering School,University of Science and Technology Beijing.Beijing 100083.China 2)Technology Center of Jinan IronSteel Group Corporation.Jinan 250101 China ABSTRACI It is difficult to extract crack defects from a defect image by traditional detection techniques, for the image of surface crack defects on a medium plate is influenced greatly by background noise and oxi- dized scale.Morphological wavelet,which combines the advantages of wavelet decomposition with mathe- matical morphology,was introduced to decompose the defect image and detect catastrophe signals.Accord- ing to the similar linear character of crack defects in shape,the crack defects can be accurately extracted from the image.Experimental results show that the morphological wavelet transform presented here is more effective than other techniques for the detection of surface crack defects of a medium heavy plate. KEY WORDS medium plate;surface crack;defect detection;morphological wavelet图8 形态小波在线检测出的裂纹图像 Fig.8 Online detected crack image by morphological wavelet 3 结语 由于形态小波是在数学形态学以及小波理论 基础上发展起来的它具有数学形态学的形态滤 波特性以及小波的检测突变信号的能力因此在 图像检测中可以用于检测一些在灰度值上存在 着突变并且存在着明显形态特征的目标如表面 裂纹等.实验证明了形态小波变换方法对于检测 中厚板表面裂纹缺陷取得了很好的效果检测的 时间短一般在几十毫秒左右同时去除噪声的能 力强. 参 考 文 献 [1] Mallat S G.A theory for mult-i resolution signal decomposition:the wavelet representation.IEEE Trans Pattern Anal Mach Intell198911(7):674 [2] Nowak R DBaraniuk R G.Wavelet-based transformations for nonlinear signal processing.IEEE Trans Signal Process 199947(7):1852 [3] Dougherty E R.An introduction to morphological image processing.Bellingham:SPIE Press1992 [4] Goutsias JSchonfeld D.Morphological representation of discrete and binary images.IEEE Trans Signal Process1991 39(6):1369 [5] Chatzis VPitas I.A generalized fuzzy mathematical morphology and its application in robust2—D and3—D object representation.IEEE Trans Image Process20009(10):1798 [6] Lazzaroni FLeonardi RSignoroni A.High-performance embedded morphological wavelet coding.IEEE Signal Process Lett200310(10):293 [7] Heijmans H J A MGoutsias J.Nonlinear mult-i resolution signal decomposition schemes:Part Ⅱ morphological wavelets. IEEE Trans Image Process20009(11):1897 [8] Cha HChaparro L F.A morphological wavelet transform∥ International Conference on Systems Engineering in the Service of Humans:SystemsMan and Cybernetics.South Korea 1993:501 [9] Lam W KIp H RLi C K.Approaches to decompose flat structuring element for fast overlapping search morphological algorithm∥Proceedings of the 14th International Conference on Pattern Recognition.1998:1461 Application of morphological wavelet in surface crack defect detection of medium plates W U Guif ang 1)XU Ke 1)XU Jinw u 1)Y A NG Chaolin 1)SUN Hao 2)WA NG Chunmei 2) 1) Mechanical Engineering SchoolUniversity of Science and Technology BeijingBeijing100083China 2) Technology Center of Jinan Iron & Steel Group CorporationJinan250101China ABSTRACT It is difficult to extract crack defects from a defect image by traditional detection techniques for the image of surface crack defects on a medium plate is influenced greatly by background noise and oxidized scale.Morphological waveletwhich combines the advantages of wavelet decomposition with mathematical morphologywas introduced to decompose the defect image and detect catastrophe signals.According to the similar linear character of crack defects in shapethe crack defects can be accurately extracted from the image.Experimental results show that the morphological wavelet transform presented here is more effective than other techniques for the detection of surface crack defects of a medium heavy plate. KEY WORDS medium plate;surface crack;defect detection;morphological wavelet ·594· 北 京 科 技 大 学 学 报 2006年第6期