Title: Study on extraction method of image's line shape feature parameters based on Canny operator

Authors: Hai Wang

Addresses: School of Art, Hunan City University, Yi Yang 413000, Hunan, China

Abstract: There are some problems in the extraction of image line shape feature parameters, such as low extraction accuracy and serious noise interference. The image line morphological features are divided into step type, slope type, triangle roof type, double step type and double roof type, and the judgment formula is set according to different types of image lines. Canny operator is adopted to extract line shape feature parameters. The best grey value of image line is obtained through the transformation of maximum value, average grey value and average weight. The image is processed by wavelet transform and the Canny operator is used to enhance the line detail information. According to the signal-to-noise ratio, the best location, unilateral response, etc., the line morphological features of the image are extracted. The results show that the error of the new method in this paper is about 4%.

Keywords: canny operator; line shape parameter of image; component method; median filter; signal-to-noise-ratio.

DOI: 10.1504/IJICT.2022.126487

International Journal of Information and Communication Technology, 2022 Vol.21 No.4, pp.369 - 385

Received: 19 Oct 2020
Accepted: 13 Dec 2020
Published online: 26 Oct 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article