Back to Search Start Over

A New Edge Detection Algorithm Based on Canny Idea.

Authors :
Yingke Feng
Jinmin Zhang
Siming Wang
Source :
AIP Conference Proceedings; 2017, Vol. 1890 Issue 1, p1-7, 7p, 2 Black and White Photographs, 1 Diagram, 2 Charts
Publication Year :
2017

Abstract

The traditional Canny algorithm has poor self-adaptability threshold, and it is more sensitive to noise. In order to overcome these drawbacks, this paper proposed a new edge detection method based on Canny algorithm. Firstly, the media filtering and filtering based on the method of Euclidean distance are adopted to process it; secondly using the Freichen algorithm to calculate gradient amplitude; finally, using the Otsu algorithm to calculate partial gradient amplitude operation to get images of thresholds value, then find the average of all thresholds that had been calculated, half of the average is high threshold value, and the half of the high threshold value is low threshold value. Experiment results show that this new method can effectively suppress noise disturbance, keep the edge information, and also improve the edge detection accuracy. [ABSTRACT FROM AUTHOR]

Details

Language :
English
ISSN :
0094243X
Volume :
1890
Issue :
1
Database :
Complementary Index
Journal :
AIP Conference Proceedings
Publication Type :
Conference
Accession number :
125543210
Full Text :
https://doi.org/10.1063/1.5005213