Open AccessDesign of defect detection system for semiconductor plastic packaging based on machine vision
Author(s) -
Zan-Yao Niu
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2006/1/012008
Subject(s) - machine vision , computer vision , artificial intelligence , feature (linguistics) , computer science , semiconductor , lens (geology) , engineering , materials science , optoelectronics , philosophy , linguistics , petroleum engineering
At present, the ability of defect detection system to detect the position of semiconductor plastic packaging defects is poor, so a defect detection system of semiconductor plastic packaging based on machine vision is designed. In terms of hardware, based on machine vision technology, the system framework is determined, the camera, lens and light source models are selected, and the detection target image acquisition parameters are determined. In terms of software, according to the target image obtained by machine vision technology, the image is preprocessed, segmented and feature extracted to identify and detect the target image and detect the semiconductor plastic packaging defects. Experimental results: compared with the system selected in this experiment, the design system can accurately detect the defects in the semiconductor plastic packaging and the location of the defects.