LCD pixel defect detection using shallow CNN based approach

ÇELİK A., KÜÇÜKMANİSA A., Sumer A., TAŞYAPI ÇELEBİ A., URHAN O.

2021 International Conference on INnovations in Intelligent SysTems and Applications, INISTA 2021, Kocaeli, Türkiye, 25 - 27 Ağustos 2021 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/inista52262.2021.9548540
  • Basıldığı Şehir: Kocaeli
  • Basıldığı Ülke: Türkiye
  • Anahtar Kelimeler: Defect detection, LCD, Pixel defect, Shallow CNN
  • Kocaeli Üniversitesi Adresli: Evet

Özet

© 2021 IEEE.Pixel level defects in LCDs are inevitable due to the manufacturing process of these panels. Thus, it is required to detect this kind of defects during the manufacturing process of LCDs or final products i.e. TV, smart phone or tablets. Conventionally this kind of quality control operations are performed by human operators which makes the process error prone. In this work, we propose a computer vision based automatic inspection system to detect pixel level defects. The proposed approach benefits from convolutional neural networks to detect these kinds of defects. Our experiments reveal that the proposed approach is able to provide superior detection performance.