In the field of magnetic tile surface detection, artificial detection efficiency is low, and the traditional
image segmentation algorithm cannot show good performance when the gray scale of the magnetic tile itself is
small, or the image is affected by uneven illumination. In view of these questions, this paper puts forward a new
clustering segmentation algorithm based on texture feature. This algorithm uses Gabor function spectra to represent
magnetic tile surface texture and then uses a user-defined local product coefficient to modify Gabor energy
spectra to get the center number of fuzzy C-means (FCM) clustering. Moreover, the user-defined Gabor energy
spectra image is segmented by clustering algorithm. Finally, it extracts the magnetic tile surface defects according
to the changes of regional gray characteristics. Experiments show that the algorithm effectively overcomes the
noise interference and makes a good performance on accuracy and robustness, which can effectively detect crack,
damage, pit and other defects on the magnetic tile surface.
LI Dan (李丹), NIU Zhongbin (牛中彬), PENG Dongxu (彭冬旭)
. Magnetic Tile Surface Defect Detection Based on Texture Feature Clustering[J]. Journal of Shanghai Jiaotong University(Science), 2019
, 24(5)
: 663
-670
.
DOI: 10.1007/s12204-019-2101-7
[1] WONG W K, YUEN C W M, FAN D D, et al. Stitchingdefect defection and classification using wavelettransform and BP neural network [J]. Expert Systemswith Applications, 2009, 36(2): 3845-3856.
[2] HALFAWY M R, HENGMEECHAI J. Automated defectdetection in sewer closed circuit television imagesusing histograms of oriented gradients and supportvector machine [J]. Automation in Construction, 2014,38: 1-13.
[3] YU Y W, YIN G F, JIANG H H, et al. Defect extractionmethod of arc magnet surface images basedon adaptive morphological filtering [J]. Journal ofComputer-Aided Design & Computer Graphics, 2012,24(3): 351-356 (in Chinese).
[4] JIANG H H, LI X Q, YANG C L, et al. Defect detectionon magnetic tile surfaces based on improvedadaptive median filtering [J]. Modular Machine Toll &Automatic Manufacturing Technique, 2014 (2): 79-82(in Chinese).
[5] LI X Q, JIANG H H, YIN G F. Detection of surfacecrack defects on ferrite magnetic tile [J]. NDT & EInternational, 2014, 62: 6-13.
[6] LIN L J, HE M G, YIN Y Y, et al. Edge detection algorithmof magnetic tile crack based on wavelet modulusmaxima [J]. Journal of University of Electronic Scienceand Technology of China, 2015, 44 (2): 283-288(in Chinese).
[7] YANG C L, LIU P Y, YIN G F et al. Defect detectionin magnetic tile images based on stationary wavelettransform [J]. Ndt & E International, 2016, 83: 78-87.
[8] RAFAEL C G, RICHARD EW. Digital image processing[M]. RUAN Q, RUAN Y (trans). Beijing, China:Publish House of Electronics Industry, 2012: 534-553(in Chinese).
[9] LIU L, KUANG G Y. Overview of image texture featureextraction methods [J]. Journal of Image andGraphics, 2009, 14(4): 622-635 (in Chinese).
[10] HU M, ZHU H, WANG X H, et al. Expression recognitionmethod based on gradient Gabor histogram features[J]. Journal of Computer-Aided Design & ComputerGraphics, 2013, 25(12): 1856-1861 (in Chinese).
[11] RAJA K B, MADHESWARAN M, THYAGARAJAH K. Texture pattern analysis of kidney tissues for disorderidentification and classification using dominantGabor wavelet [J]. Machine Vision and Applications,2010, 21(3): 287-300.
[12] KUMAR S, RAY S K, TEWARI P. A hybrid approachfor image segmentation using fuzzy clustering and levelset method [J]. International Journal of Image, Graphicsand Signal Processing, 2012, 4(6): 1-7.
[13] ZHANG X F, ZHANG C M, TANGWJ, et al. Medicalimage segmentation using improved FCM [J]. ScienceChina: Information Sciences, 2012, 55(5): 1052-1061.
[14] ZHANG M, YU J. Fuzzy partitional clustering algorithms[J]. Journal of Software, 2004, 15(6): 858-868(in Chinese).
