Crack Severity Classification from Timber Cross-Sectional Images Using Convolutional Neural Network

Cedar and cypress used for wooden construction have high moisture content after harvesting. To be as building materials, they must undergo high-temperature drying. However, this process causes internal cracks that are invisible on the outer surface. These defects serious because reduce strength of timber, i.e., buckling joint durability. Therefore, severity should evaluated. A square timber was cut at an arbitrary position assessed based length, thickness, shape in cross-section; however, is time-consuming labor-intensive. we a convolutional neural network (CNN) to automatically evaluate from cross-sectional images. Previously, silver-painted images cross-sections so easier observe; task burdensome. Hence, study, attempted classify crack using ResNet (Residual Neural Network) unpainted First, ResNet50 employed trained with supervised data level. The classification accuracy then evaluated test (not training) reached 86.67%. In conclusion, confirmed proposed CNN could behalf humans.