测试技术学报2018，Vol.32Issue(4)：312-317,6.DOI:10.3969/j.issn.1671-7449.2018.04.006
基于Gabor小波变换的图像轮廓角点检测
Image Contour Corner Detection Based on Gabor Wavelet Transform
摘要
Abstract
The traditional methods of corner detection of image contour mostly adopt Gaussian smoot-hing to the image contour.It makes the corner point positioning accuracy lower and sensitive to local subtle changes and noise.In order to solve this problem,a corner detection of image contour based on Gabor filter is proposed,which uses the good characteristics of Gabor filter in representation of gradient amplitude change information at edge and corner points.And the virtual part of Gabor filter was used to smooth the image contour.Different from the traditional algorithm that takes the geometrical character-istics of the contour as the measure of the corner point,the angular difference between the main direc-tion and the adjacent pixel direction of the contour pixel after Gabor filter smoothing is used as the cor-ner point measure,which effectively improves the robustness to noise and the accuracy of corner positio-ning.Experimental results show that the new algorithm has better localization and noise robustness.关键词
轮廓提取/角点检测/Gabor小波变换/滤波器虚部平滑/角点检测Key words
contour extraction/corner detection/Gabor wavelet transform/filter imaginary smoothing/corner measurement分类
信息技术与安全科学引用本文复制引用
马迎春,刘慧锋,张航..基于Gabor小波变换的图像轮廓角点检测[J].测试技术学报,2018,32(4):312-317,6.基金项目
山西省重点研发计划项目(201603D121005) （201603D121005）
