<![CDATA[The potential and usefulness of wood to meet the needs of human life are not in doubt. Demands us to continue to maintain the quality. Wood quality is closely related to wood defects. Manual defect checks in the wood industry are unreliable because they are prone to human error, For example, due to acute symptoms of headaches and tired eyes, technology in the form of image processing can help identify wood defects Swietenia Mahagoni. In this case, the method used is Euclidean distance with a ratio of k-means segmentation and thresholding on 42 images of wood defects consisting of 3 types of defects, namely growing skin defects, rotting knots, and healthy knots, every 14 images with data sharing. training for 30 images and testing for 12 images. The results of the k-means segmentation are then extracted on 6 features including metric, eccentricity, contrast, correlation, energy, and homogeneity using the Gray Level Co-occurrence Matrix (GLCM) extractor and classified by calculating the closest distance using the euclidean distance between the results of data feature extraction. testing of the value of feature extraction in the training data which is used as a previous database. It is the smallest value that indicates the type of defect. The success calculation is presented in the confusion matrix calculation and gets a success or accuracy value of 91.67%.]]>
