Development of a color machine vision method for wood surface inspection

The purpose of this thesi s is to present a case study of the development , implementation and performance analysi s of a color-based visual surface inspection method for wood properties. The main contribution of the study is to answer the need of design strategies, performance characterizatio n methods and case studies in the field of automated visual inspection, and especially wood surface inspection. In real time color-based inspection, the complexity of the methods is important . In this study, defect detection and recognition methodsbased on color histogram percentil e featuresareproposed. The color histogram percentil e features were noticed to be able to recognize well wood surface defects with relat ively low complexity. A common problem in visual inspection applications is the collection and labelling of training material since human made labellings can be errorneous . Further, the classifier s are relatively static when once trained, thus offering only littl e possibilities for adjusting classification. In the study, a self-organizing map (SOM) -based approach for classifier user interface in visual surface inspection problems is introduced. The approach relieves the labelling of training material , simplifies retraining, provides an illustrative an intuitive user interface and offers a convenient way of controlling classification. Thestudy is illustrated with four experiment s related to themethod development and analysis. In the first experiment , a simulator environment is used for determining the relationshi p of the defect detection and recognition and grading accuracy. The second experiment consider s the suitability of different color spaces for wood defect recognition under changing illumination. RGB color space gives the best results compared to grey-level and other color spaces . The third experiment presents the experimental wood surface inspection setup implementing the method developed in this study. Comparat ive performance analysi s results are presente d and the difficulties, mainly caused by segmentation of thedefects, arediscussed . The fourth experiment demonstrate s thesuitability of the method for parquet sorting and sh ows the potential of the non-s egmenting approach.

[1]  Matti Pietikäinen,et al.  Online color camera calibration , 1994, Proceedings of 12th International Conference on Pattern Recognition.

[2]  Quang-Tuan Luong,et al.  Color in Computer Vision , 1993, Handbook of Pattern Recognition and Computer Vision.

[3]  J. Cohen.  Dependency of the spectral reflectance curves of the Munsell color chips , 1964 .

[4]  Jorma Laaksonen,et al.  SOM_PAK: The Self-Organizing Map Program Package , 1996 .

[5]  Richard W. Conners,et al.  Identifying and Locating Surface Defects in Wood: Part of an Automated Lumber Processing System , 1983, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Anil K. Jain,et al.  A Survey of Automated Visual Inspection , 1995, Comput. Vis. Image Underst..

[7]  Dominik Paul,et al.  VISTA: Visual Interpretation System for Technical Applications-Architecture and Use , 1988, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  L. Maloney,et al.  Color constancy: a method for recovering surface spectral reflectance. , 1986, Journal of the Optical Society of America. A, Optics and image science.

[9]  E. C. Semple,et al.  Detection and sizing visual features in wood using tonal measures and a classification algorithm , 1989, Pattern Recognit..

[10]  Matti Pietikäinen,et al.  Accurate color discrimination with classification based on feature distributions , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[11]  L. Kronsjö.  Algorithms: Their Complexity and Efficiency , 1979 .

[12]  Hannu Kauppinen,et al.  The effect of illumination variations on color-based wood defect classification , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[13]  P. Alapuranen,et al.  Automatic visual inspection of wood surfaces , 1992, [1992] Proceedings. 11th IAPR International Conference on Pattern Recognition.

[14]  Antti J. Koivo,et al.  Robust image modeling for classification of surface defects on wood boards , 1989, IEEE Trans. Syst. Man Cybern..

[15]  Jouko Lampinen,et al.  Wood defect recognition with self-organizing feature selection , 1994, Other Conferences.

[16]  Sar Sardy,et al.  Experimental medical and industrial applications of neural networks to image inspection using an inexpensive personal computer , 1996 .

[17]  W. N. Sproson,et al.  Colour Science in Television and Display Systems , 1983 .

[18]  Tai-Hoon Cho,et al.  A computer vision system for analyzing images of rough hardwood lumber , 1990, [1990] Proceedings. 10th International Conference on Pattern Recognition.

[19]  Esko Herrala,et al.  Imaging spectrometer for process industry applications , 1994, Other Conferences.

[20]  Markku Pietikäinen.  Detection of knots in logs using x-ray imaging , 1996 .

[21]  Antti J. Koivo,et al.  Hierarchical classification of surface defects on dusty wood boards , 1994, Pattern Recognit. Lett..

[22]  Hannu Kauppinen,et al.  Color vision based methodology for grading lumber , 1994, Proceedings of 12th International Conference on Pattern Recognition.

[23]  John Rasure,et al.  The Khoros Application Development Environment , 1994 .

[24]  John W. Sammon,et al.  A Nonlinear Mapping for Data Structure Analysis , 1969, IEEE Transactions on Computers.

[25]  Pasi Koikkalainen,et al.  Progress with the Tree-Structured Self-Organizing Map , 1994, ECAI.

[26]  G. Wyszecki,et al.  Color Science Concepts and Methods , 1982 .

[27]  Morton Nadler,et al.  Pattern recognition engineering , 1993 .

[28]  Ari Visa,et al.  Unsupervised segmentation of surface defects , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[29]  David A. Butler,et al.  An evaluation of color spaces fordetecting defects in Douglas-fir veneer , 1992 .

[30]  W. Pölzleitner,et al.  Real-time surface grading of profiled wooden boards , 1992 .

[31]  Jouko Lampinen,et al.  Wood Defect Recognition: A Comparative Study , 1994 .

[32]  Hannu Kauppinen,et al.  Self-Organizing Map Based User Interface for Visual Surface Inspection , 1999 .

[33]  Jouko Lampinen,et al.  Self-Organizing Feature Extraction in Recognition of Wood Surface Defects and Color Images , 1996, Int. J. Pattern Recognit. Artif. Intell..

[34]  David T. Berry.  Colour recognition using spectral signatures , 1987, Pattern Recognit. Lett..

[35]  Charles W. McMillin,et al.  Lumber defect detection abilities of furniture rough mill employees , 1985 .

[36]  G. Healey,et al.  Global color constancy: recognition of objects by use of illumination-invariant properties of color distributions , 1994 .

[37]  Olle Hagman,et al.  Near-sensor real time spectral classification for industrial applications , 1996 .

[38]  Kevin W. Bowyer,et al.  Empirical evaluation techniques in computer vision , 1998 .

[39]  T. Kanade,et al.  Color information for region segmentation , 1980 .

[40]  Sehran Tatari,et al.  Automatic Recognition Of Defects In Wood , 1987, Other Conferences.

[41]  Glenn Healey,et al.  Markov Random Field Models for Unsupervised Segmentation of Textured Color Images , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[42]  Hannu Kauppinen,et al.  Recent Development in Wood Inspection , 1996, Int. J. Pattern Recognit. Artif. Intell..

[43]  Matti Pietikäinen,et al.  Edge-based texture measures for surface inspection , 1992, Proceedings., 11th IAPR International Conference on Pattern Recognition. Vol.II. Conference B: Pattern Recognition Methodology and Systems.

[44]  Matti Pietikaeinen,et al.  Color analysis of defects for automated visual inspection of pine wood , 1993, Electronic Imaging.

[45]  Jiri Matas,et al.  Colour-based object recognition under spectrally non-uniform illumination , 1995, Image Vis. Comput..