Defect Detection on Texture using Statistical Approach


  • Siana Halim Industrial Engineering Petra Christian University Surabaya - Indonesia



Hill estimator, kernel density estimate, image histogram, wavelet, texture, defect.


In this paper we present several techniques for detecting simple defect on the texture. The simple defect means, that the defect can be detected via image histogram or via wavelet of the image histogram. Hill estimator is one of the techniques that we suggest to use to solve this problem, since it does not need estimate parameters for estimating the image density


.Haindl, M., Texture Synthesis, CWI Quaterly,1991,pp. 305-331.

.Cavalin, P., Oliveira, L. S., Koerich, A. L., and Britto Jr, A.S., Wood Defect Detection using Grayscale Images and an Optimized Feature Set. IEEE Industrial Electronics, IECON 2006 -32nd Annual conferenceProceeding, Paris 2006.

.Wand, M.P and Jones, M.C; Kernel Smoothing, Chapman & Hall, 1995.New Jersey: Pearson Prentice Hall. 2005

.Moahseri, B. B. M, Azadinia, S. and Mehbodniya, A New Voting Approach to Texture Defect Detection Based on Multiresolutional Decomposition, World Academy of Science, Engineering and Technology, 65, 2010, pp. 887-891.

Timm, F. and Barth, E., Non-parametric Texture Defect Detection using Weibull Features. Image Processing: Machine Vision Applications IV, 7877, Proceeding of SPIE. SPIE-IS& T. San Francisco, USA 2011.

Franke, J., and Halim, S., Wild Bootstrap Tests: Regression Models for Comparing Signal and Images, IEEE Signal Processing Magazine, July 2007, pp. 31-37

Franke, J., and Halim, S., A Bootstrap Test for Comparing Images in Surface Inspection, presented at DFG-Priority Program 1114 Technical Report 150.

Nason, G.P., Wavelet Methods in Statistics, Springer, New York, 2008

Wand, M.P and Jones, M.C; Kernel Smoothing, Chapman & Hall, 1995

Otsu, N., A Threshold Selection Method from Gray-Level Histogram. IEEE Transaction on System Man Cybernet, SMC-9(1), 1979,pp.62-66

Hill, B. M., A Simple General Approach to Inference about the Tail of Distribution. Annals of Statistics, 3, 1975, pp. 1063-1174.

Ng Hui Fang, Automatic Thresholding for defect detection, Pattern Recog¬nition Letters, 27(14), 2006,pp.1644-1649.