Automatic visual inspection of wood surfaces

A prototype software system for visual inspection of wood defects has been developed. The system uses a hierarchical vector connected components (HVCC) segmentation which can be described as a multistage region-growing type of segmentation. The HVCC version used in experiments uses RGB color vector differences and Euclidean metrics. The HVCC segmentation seems to be very suitable for wood surface image segmentation. Geometrical, color and structural features are used in classification. Possible defects are classified using combined tree-kNN classifier and pure kNN-classifier. The system has been tested using plywood boards. Preliminary classification accuracy is 85-90% depending on the type of defect.<<ETX>>

[1]  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.

[2]  Matti Pietikäinen,et al.  A hybrid computer architecture for machine vision , 1990, [1990] Proceedings. 10th International Conference on Pattern Recognition.

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

[4]  Antti J. Koivo,et al.  Hierarchical classification of surface defects on dusty wood boards , 1990, [1990] Proceedings. 10th International Conference on Pattern Recognition.

[5]  Matti Pietikäinen,et al.  Color segmentation by hierarchical connected components analysis with image enhancement by symmetric neighborhood filters , 1990, [1990] Proceedings. 10th International Conference on Pattern Recognition.

[6]  Larry S. Davis,et al.  A new class of edge-preserving smoothing filters , 1987, Pattern Recognit. Lett..