Open AccessResearch on image segmentation and edge detection technology based on computer vision
Author(s) -
Jing Qi,
Haitao Yang
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 - 10.1088/1742-6596/1994/1/012035
Subject(s) - sobel operator , canny edge detector , deriche edge detector , computer vision , image gradient , artificial intelligence , edge detection , computer science , image segmentation , enhanced data rates for gsm evolution , segmentation , image (mathematics) , pattern recognition (psychology) , image processing
With the continuous development of computer vision technology, image segmentation technology and edge detection technology are becoming more and more mature. Firstly, this paper simulates the fixed threshold method and adaptive threshold method in image segmentation, and concludes that the fixed threshold method is applicable to images with obvious targets and backgrounds, and the adaptive threshold method is applicable to images with uneven distribution of light and dark, and secondly, the comparative experiment and result evaluation of edge detection of images by Sobel operator and Canny operator are used to conclude that the Sobel algorithm is more accurate in positioning edges, while Canny algorithm image edge information is more continuous and single edge response is better. Its edge detection is also better.