A comparative study of transfer learning-based methods for inspection of mobile camera modules
- 주제(키워드) Camera module , Defect inspection , Machine vision , Transfer learning
- 등재 SCOPUS, KCI등재
- 발행기관 Institute of Electronics and Information Engineers
- 발행년도 2018
- URI http://www.dcollection.net/handler/ewha/000000156077
- 본문언어 영어
- Published As http://dx.doi.org/10.5573/IEIESPC.2018.7.1.070
- 저작권 이화여자대학교 논문은 저작권에 의해 보호받습니다.
초록/요약
We apply three transfer learning methods using the pretrained AlexNet convolutional neural network (CNN) model to detect defects in camera modules. In experiments, the performance of fine-tuning methods using random initial parameters in less than the two last fully connected layers while using predetermined weights as initial parameters for the remaining layers, showed better performance than other methods. We expect that the transfer learning-based CNN can be effectively applied to camera module inspection systems. © 2018 The Institute of Electronics and Information Engineers.
more