Fuzzy logic-based segmentation of manufacturing defects on reflective surfaces
Neural Computing and Applications, cilt.29, sa.8, ss.107-116, 2018 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 29 Sayı: 8
- Basım Tarihi: 2018
- Doi Numarası: 10.1007/s00521-017-2862-6
- Dergi Adı: Neural Computing and Applications
- Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
- Sayfa Sayıları: ss.107-116
- Anahtar Kelimeler: Defect detection, Fuzzy logic, Glass defect inspection, Image segmentation, Reflective surface analysis, Texture analysis
- Ankara Hacı Bayram Veli Üniversitesi Adresli: Evet
Özet
Automatic defect detection on reflective surfaces is a compelling process. In particular, detection of tiny defects is almost impossible for human eye or simple machine vision methods. Therefore, the need for fast and sensitive machine vision methods has gained importance. In this study, an effective defect detection method is presented for reflective surfaces such as glass, tile, and steel. Defects on the surface of the product are determined automatically without the need for human intervention. The proposed system involves illumination unit, digital camera, and defect detection algorithm. Firstly, color image is taken by digital camera. Then, properties of taken image are selected. At this stage, ambient condition of lighting devices is very important. Reflections are minimized thanks to the true lighting. Selected properties are: red, green, and blue values, and luminance value. These properties are applied to fuzzy inputs. Information from the inputs is evaluated according to determined rules. Finally, each pixel is classified as black or white. Thirty-two glass pieces are tested using the proposed system. The proposed method was compared with commonly used methods. The success rate of the proposed algorithm is 83.5% and is higher than that of other algorithms.