Research on Key Technology of Online Detection for Particleboard

Abstract

In this paper, the Support Vector Machine (SVM) combined with an Adaboost online detection algorithm is proposed for the problem of particleboard defect detection. The algorithm has improved the LTP feature value algorithm, and the Adaboost is improved to the accuracy of the particleboard defect. At the same time, this paper proposes a feature extraction method based on improved SURF algorithm and Tamura texture characteristics, and verifies the effectiveness and rapidity of the method in feature extraction. Effectively reduce the redundancy of sample training information, improve the efficiency and accuracy of the defect type identification.