AUTOMATIC SURFACE INSPECTION FOR DIRECTIONAL TEXTURES USING PRINCIPAL COMPONENT ANALYSIS
A global image restoration scheme using principal component analysis (PCA) is proposed in this paper. This PCA-based image restoration scheme can be used for inspecting the defects in directionally textured surfaces automatically. Decomposing the gray level of image pixels into an ensemble of row vectors, we first transform the original input space into principal component space. The repetitive and periodical primitives are well reconstructed by first k major components and their corresponding weight vectors, named truncated component solution (TCS). Then the local defects will be revealed by applying image subtraction between the original image and the TCS. As a consequence, the directional textures are eliminated and only local defects are preserved if they initially are embedded in the surface. The maximal change of curvature of the scree plot (MCCSP) criterion is developed which enables users extract a proper value of k automatically. Experiments on a variety of products with directional texture surface demonstrate the effectiveness and robustness of the proposed method.
Directional tezture Principal component analysis Defect inspection Machine vision
Der-Baau Perng Ssu-Han Chen
Department of Industrial Engineering and Management, National Chiao-Tung University, 1001, Universit Department of Industrial Engineering and Management, National Chiao-Tung University,1001, University
国际会议
上海
英文
1-5
2009-08-02(万方平台首次上网日期,不代表论文的发表时间)