A Detection Method for Wood Surface Defect Based on Feature Fusion

Abstract

The wood defect detection is an important link in the process of furniture manufacturing. In this paper, a detection method for wood surface defect based on feature fusion was proposed. Firstly, the histogram of oriented gradient (HOG) was extracted from wood surface defect images, and then kernel principal component analysis (KPCA) was used to reduce the dimension of the extracted HOG features. Finally, by fusing the dimension reduced HOG-KPCA and the gray-level co-occurrence matrix (GLCM) which extracted form wood surface defect images, HOG-KPCA-GLCM features were obtained. And the support vector machine (SVM) is used as classifier. The effects of HOG dimension and weight ratio on classication performance had been experimetally investigated. The results suggested that the detection accuracy had been improved by using the proposed method on the test set, recording an accuracy of 91.26%.