IDENTIFICATION OF KNOTTY CORE IN PINUS RADIATA LOGS FROM COMPUTED TOMOGRAPHY IMAGES USING ARTIFICIAL NEURAL NETWORK

The feasibility of identifying Knotty core in images of X-ray computed tomography (CT) of pruned radiata pine logs (Pinus radiata D. Don), was evaluated using a supervised classification method based on artificial neural networks (ANN). The classification process also considers the identification of the clear wood and knots. Thirty pruned radiata pine logs were scanned in a multi-slice scanner medical X-ray, where the resulting CT images were obtained every 5 mm. A total of 270 CT images were classified using the ANN, and the resulting thematic maps were filtered with a median filter of 7 x 7. The accuracy of the classification process of the CT images was obtained from a confusion matrix and Kappa statistics. The results indicated that the Knotty core can be identified and separated with an accuracy of 92.7%, while for the overall accuracy was obtained a value of 85.0%. After filtering thematic maps, the precision values increased to 96.3% and 92.3% for the defective core and overall accuracy, respectively. Kappa values were 0.607 and 0.764 for thematic maps and thematic maps filtered, respectively. These values indicate that there is a strong degree of agreement between reference data and classification process. The results suggest that it is feasible to apply artificial neural networks as classification procedure to identify the Knotty core in CT images of pruned radiata pine logs.