Open AccessMixed method of printed circuit board photoelectric image edge information extraction
Author(s) -
Naosheng Qiao,
Ping Sun
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 - 10.1088/1742-6596/1775/1/012004
Subject(s) - artificial intelligence , computer vision , noise (video) , edge detection , enhanced data rates for gsm evolution , canny edge detector , computer science , image (mathematics) , filter (signal processing) , image gradient , image processing , pattern recognition (psychology)
Due to the noise and fuzziness of printed circuit board (PCB) photoelectric image, a mixed method of image edge information extraction is proposed, which is based on median filter, improved maximum distance between categories (IMDBC) and improved Canny edge detection operator (ICEDO). Firstly, the basic principle and advantages of the maximum distance between categories (MDBC) for image segmentation are analyzed, then the improved MDBC is proposed. Secondly, the ICEDO is discussed, and the basic principle is analyzed. Then, the implementation steps of the image edge information extraction based on mixed method are discussed. The experiments using such noisy and fuzzy images acquired by charge coupled device (CCD) imaging system and microscope are actualized. The experiment results by using the mixed method have lesser noise points, and the image excellent quality coefficients are the highest in the experiment. The excellent quality coefficients are 0.8346 and 0.7981 respectively in two image edge information extraction experiments. Both of the subjective and objective experiment contrasts results verify that the edge information of image can be extracted better by using this method.