Analyzing Internal Hardwood Log Defect Prediction Equation Accuracy
The type, size, and location of internal defects dictate the grade and value of lumber sawn from hardwood logs. Significant correlations have been documented among external log defect indicators and internal defect features. A series of prediction models for four hardwood species have been developed based on these correlations. A recently developed high-resolution laser log scanner that locates external log defects uses these models to predict internal defect locations and sizes based on detected external defect data. The prediction models are specific for species and defect type and allow users to estimate internal defect size, shape, depth, and position. Forest Service scientists found that the models achieved approximately 70 percent accuracy when predicting the occurrence of a knot defect. In addition, the models accurately predicted the sawing of 95 of 105 boards with no knot defects, an accuracy rate of 90.5 percent. These methods have the potential to allow full optimization of the sawing of a log based on internal defect data using laser scanning technology, thus reducing waste and improving the value of lumber sawn.