Automated Labeling of Log Features in CT Imagery of Multiple Hardwood Species

Daniel L. Schmoldt, Jing He, A. Lynn Abbott


Before noninvasive scanning, e.g., computed tomography (CT), becomes feasible in industrial sawmill operations, we need a procedure that can automatically interpret scan information in order to provide the saw operator with information necessary to make proper sawing decisions. To this end, we have worked to develop an approach for automatic analysis of CT images of hardwood logs. Our current approach classifies each pixel individually using a feed-forward artifical neural network (ANN) and feature vectors that include a small, local neighborhood of pixels and the distance of the target pixel to the center of the log. Initially, this ANN was able to classify clear wood, bark, decay, knots, and voids in CT images of two species of oak with 95% pixel-wise accuracy. Recently we have investigated other ANN classifiers, comparing 2-D versus 3-D neighborhoods and species-dependent (single species) versus species-independent (multiple species) classifiers using oak (Quercus rubra L. and Q. nigra L.), yellow-poplar (Liriodendron tulipifera L.), and black cherry (Prumus serotina Ehrh.) CT images. When considered individually, the resulting species-dependent classifiers yield similar levels of accuracy (96-98%). 3-D neighborhoods work better for multiple-species classifiers, and 2-D is better for the single-species case. Classifiers combining yellow-poplar and cherry data misclassify many pixels belonging to splits as clear wood, resulting in lower classification rates. If yellow-poplar was not paired with cherry, however, we found no statistical difference in accuracy between the single-and multiple-species classifiers.


Industrial inspection;segmentation;computed tomography;image analysis;log processing

Full Text:



  • There are currently no refbacks.