基于小波变换的边缘检测技术  

Edge Detection Technology Based on Wavelet Transform

作　　者：刘光宇 刘彪 曾志勇 赵恩铭 邢传玺[3] LIU Guangyu;LIU Biao;ZENG Zhiyong;ZHAO Enming;XING Chuanxi(School of Engineering,Dali University,Dali Yunnan 671003,China;School of Physics and Photoelectric Engineering,Harbin Engineering University,Harbin 150001,China;School of Electrical Information Engineering,Yunnan University for Nationalities,Kunming 650031,China)

机构地区：[1]大理大学工程学院,云南大理671003 [2]哈尔滨工程大学物理与光电工程学院,哈尔滨150001 [3]云南民族大学电气信息工程学院,昆明650031

出　　处：《重庆科技学院学报（自然科学版）》2022年第3期74-78,共5页Journal of Chongqing University of Science and Technology：Natural Sciences Edition

基　　金：国家自然科学基金项目“基于无线传感器网络的云南高原湖泊声层析监测关键技术研究”(61761048);云南省地方本科高校基础研究联合专项资金项目“基于谱聚类的水下目标探测关键技术研究”(2019FH001-066);黑龙江省自然科学基金项目“基于自旋交换泵浦技术的高透过率、超窄带原子滤光器的研究”(LC2018026)。

摘　　要：传统的Roberts算子不适合用来提取存在陡峭边缘的图像,提出在小波变换的基础上,对图像的边缘信息进行提取。首先,对图像进行小波分解,采用Roberts算子,求得沿着0°、45°、90°、135°等方向的模值,并将其转换为幅值;然后,根据像素点所在的方向,寻找最大值点,将此点设置为边缘点;最后,将各方向的边缘点连接,形成边缘图像。通过仿真实验可知,采用该方法提取的边缘细节更加丰富。The traditional Roberts operator is not suitable for extracting images with steep edges.Therefore,it is proposed to extract the image edge information based on wavelet transform.Firstly,the image is decomposed by wavelet,and the modulus along the gradient direction of 0°,45°,90°and 135°is calculated by Roberts operator and converted into amplitude.Then find the maximum value point according to the direction of the quadrant where the pixel is located,and set this point as the edge point.Finally,the edge points in each direction are connected to form an edge image.The simulation results show that the edge details extracted by this method are richer.

关 键 词：小波变换 边缘检测 ROBERTS算子 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]