Open AccessAutomatic segmentation of ceramic materials with relaxed possibilistic C-Means clustering for defect detection
Author(s) -
Kwang-Baek Kim,
Doo Heon Song,
Hyun Jun Park
Publication year - 2020
Publication title -
indonesian journal of electrical engineering and computer science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.241
H-Index - 17
eISSN - 2502-4760
pISSN - 2502-4752
DOI - 10.11591/ijeecs.v19.i3.pp1505-1511
Subject(s) - cluster analysis , visual inspection , segmentation , artificial intelligence , object (grammar) , pixel , computer science , automated x ray inspection , computer vision , process (computing) , span (engineering) , fuzzy clustering , image segmentation , fuzzy logic , segmentation based object categorization , pattern recognition (psychology) , scale space segmentation , image (mathematics) , image processing , engineering , structural engineering , operating system
Auromatic inspection system is necessary for reliable quality control if ceramic materials to avoid operator subjectivity and fatigue in visual inspection. Automatic segmentation from material’s image is then the most important process to develop such an inspection system. In this paper, we propose a Possibilistic C-Means pixel clustering algorithm with fuzzy stretching to form the defect object in segmentation. In experiment using 50 images containing a certain amount of defects, the proposed method was successful in 49 cases or 98% of opportunities. That performance is roughly twice better than that of standard K-means clustering in defect object formation Auromatic inspection system is necessary for reliable quality control if ceramic materials to avoid operator subjectivity and fatigue in visual inspection. Automatic segmentation from material’s image is then the most important process to develop such an inspection system. In this paper, we propose a Possibilistic C-Means pixel clustering algorithm with fuzzy stretching to form the defect object in segmentation. In experiment using 50 images containing a certain amount of defects, the proposed method was successful in 49 cases or 98% of opportunities. That performance is roughly twice better than that of standard K-means clustering in defect object formation.