@article{oai:kansai-u.repo.nii.ac.jp:00018923,
 author = {Lin, Pang-Chieh and Huang, Yi-Chen and Lin, Sheng-Jie and Kung, Huang-Kuang},
 journal = {Science and technology reports of Kansai University = 関西大学理工学研究報告},
 month = {Mar},
 note = {As a result of regulations and requests, most fasteners for the automotive industry require a 100% full quality control inspection. Conventional optical inspection machines are unable to efficiently provide 100% full quality control inspection as it is time consuming and difficult to easily detect defects and flaws. This paper focuses on the development and application of a convolution neural network (CNN) of an AI deep-learning technique for the internal thread measurement of misaligned fasteners. Integration of an optical hardware system and a software system platform is included. It is thus similar to upgrading the hardware and software system platform of conventional optical inspection machines. Utilizing the machine vision hardware, the system is capable of capturing an image of an internal thread of the fastener. In the software platform system, a CNN of deep learning is applied to detect and determine defects or flaws in the internal thread of the fastener.},
 pages = {35--52},
 title = {Application of AI deep-learning technique to the detection of internal misaligned and defective screw nuts},
 volume = {63},
 year = {2021}
}