Defect detection on wooden surfaces using color-based texture analysis and genetic feature selection

We describe a scanning system developed for the classification of hardwood surface texture. The system uses color CCD imaging sensors to analyse the surface of parquet material in terms of the texture formed by grain lines (orientation, spatial frequency, and color), various types of colorization, and other defects like knots, heart wood, cracks, holes, etc. The analysis requires two major tracks: the assignment of a tile to its texture class (like A, 13, C, 1, 2, 3, Waste), and the detection of defects that decrease the commercial value of the tile (heart wood, knots, etc.). In the industrial scenario we describe, many of the features defining a class cannot be described mathematically. Consequently a focus was the design of a learning architecture, where prototype texture samples are presented to the system, which then automatically finds the internal representation necessary for classification.