Open AccessAutomatic end-to-end veneer grading system based on machine vision
Author(s) -
Fengying Ma,
Jiyin Zhang,
Peng Ji
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 - 10.1088/1742-6596/1961/1/012029
Subject(s) - veneer , grading (engineering) , computer science , artificial intelligence , computer vision , image processing , engineering drawing , engineering , materials science , image (mathematics) , composite material , civil engineering
In order to realize automatic grading of veneers of different quality during the processing of veneer veneers, an end-to-end veneer automatic grading system based on YOLOV3 was designed. Using YOLOV3 network as the defect detection model, self-made veneer defect data set for model training. Firstly, the system obtains the veneer surface image; then the trained model detects the obtained veneer surface image, identifies scars and cavities in the surface defects, and marks the scars and cavities with rectangular boxes respectively; finally, the quality level of the veneer is determined according to the characteristic information of the detected scars and holes, and the level signal is transmitted to the PLC system to drive the grading actuator for grading. Experiments show that the end-to-end veneer automatic grading system based on machine vision can effectively carry out the automatic veneer grading, which is helpful to realize the intelligent and automated production of veneer processing and improve work efficiency.