Standing tree health assessment using contact–ultrasonic testing and machine learning

• DOI: 10.1016/j.compag.2023.107816
• Publication Date: 2023-04-03T15:51:25Z
• Authors: Mohsen Mousavi; Mohammad Sadegh T...; Amir H. Gandomi
• Keywords: Tree (set theory)

The problem of hole-defect detection in standing trees is solved. An ultrasonic device (Pundit PL-200) was employed to collect signals from various wood specimens both the lab and field. collected were then processed through Variational Mode Decomposition algorithm derive effective features. In order solve classification more efficiently, obtained characteristics analyzed PCA determine most useful Several machine learning algorithms a one-dimensional convolutional neural network (1D-CNN) set problems based on data (1) with artificial defects (2) billets natural selected harvested at sites two states WA NSW, Australia. results demonstrate effectiveness proposed method for classifying materials their health state, where testing accuracy 100% least 92.2% fields achieved. Fine Gaussian SVM found perform effectively derived fields. It also shown that 1D-CNN reliable generalizing solution