TY - GEN
T1 - A Defect Detection Method for Optical Fiber Preform Based on Machine Vision
AU - Ren, Xinzhen
AU - Zhou, Wenju
AU - Gu, Xiaogang
AU - Liu, Qiang
PY - 2021/10/19
Y1 - 2021/10/19
N2 - In order to ensure reliable product quality, fiber optics need to be detected for defects during the manufacturing process. Most domestic manufacturers use manual visual inspection and comparison with tool cards for defect detection. However, manual inspection is slow and subjective, and problems such as missed inspections and false inspections are extremely easy to occur, which seriously affects the quality of products. This paper proposes a defect detection method for optical fiber preforms based on machine vision. First, rely on the optical fiber preform defect detection experimental platform to obtain the full-angle image of the optical fiber preform. Then, the boundary of the optical fiber preform in the full-angle image is determined by the proposed algorithm and the full-angle image is preprocessed. Further, the defects are tracked through the full-angle image and the tracking information is recorded. Finally, the location, size and category of the defect are calculated by tracking information. It is verified by experiments that this algorithm can realize the detection of defects in the optical fiber preform.
AB - In order to ensure reliable product quality, fiber optics need to be detected for defects during the manufacturing process. Most domestic manufacturers use manual visual inspection and comparison with tool cards for defect detection. However, manual inspection is slow and subjective, and problems such as missed inspections and false inspections are extremely easy to occur, which seriously affects the quality of products. This paper proposes a defect detection method for optical fiber preforms based on machine vision. First, rely on the optical fiber preform defect detection experimental platform to obtain the full-angle image of the optical fiber preform. Then, the boundary of the optical fiber preform in the full-angle image is determined by the proposed algorithm and the full-angle image is preprocessed. Further, the defects are tracked through the full-angle image and the tracking information is recorded. Finally, the location, size and category of the defect are calculated by tracking information. It is verified by experiments that this algorithm can realize the detection of defects in the optical fiber preform.
U2 - 10.1007/978-981-16-7210-1_32
DO - 10.1007/978-981-16-7210-1_32
M3 - Conference Contribution (Conference Proceeding)
SN - 9789811672095
T3 - Communications in computer and information science
SP - 334
EP - 343
BT - 7th International Conference on Life System Modeling and Simulation, LSMS 2021
ER -