改进的Sobel算法和形态学小波变换相融合的图像边缘检测方法  

Image Edge Detection Method Based on Fusion of Improved Sobel Algorithm and Morphological Wavelet Transform

作　　者：殷婷婷 杨忠 孙瑞胜[2] 张棋 YIN Ting-ting;YANG Zhong;SUN Rui-sheng;ZHANG Qi(Jinling Institute of Technology,Nanjing 211169,China;Nanjing University of Science and Technology,Nanjing 210094,China;8511 Research Institute of China Aerospace Science and Industry Corporation,Nanjing 210007,China)

机构地区：[1]金陵科技学院智能科学与控制工程学院,江苏南京211169 [2]南京理工大学能源与动力工程学院,江苏南京210094 [3]中国航天科工集团8511研究所,江苏南京210007

出　　处：《金陵科技学院学报》2022年第2期15-21,共7页Journal of Jinling Institute of Technology

基　　金：金陵科技学院高层次人才科研启动基金项目(jit-b-202112)。

摘　　要：图像的边缘检测是图像处理领域的基本问题。针对小波变换模极大值法检出边缘不连续的问题,引入形态学的处理方式,增强边缘连续性。针对传统Sobel算法检测精度低且采用固定阈值的问题,提出一种结合自适应阈值和增加检测梯度的改进方法。结合形态学小波变换和改进的Sobel算法的优点,通过算法融合的方式,提出了一种将两者融合的图像边缘检测方法。仿真结果表明,该方法提高了边缘检测的精度和连续性,更好地满足了图像边缘定位稳定可靠的要求。Image edge detection is a fundamental problem in the field of image processing.Aiming at the problem of edge discontinuity detected by wavelet transform modulus maximum method,morphological processing method is introduced to enhance the edge continuity.For the problem of low detection accuracy and the use of fixed threshold of traditional Sobel algorithm,an improved method combining adaptive threshold and increasing detection gradient is proposed.Combining the advantages of morphological wavelet transform and improved Sobel algorithm,an image edge detection algorithm that combines the two is proposed by means of algorithm fusion.Simulation results show that the proposed method improves the accuracy and continuity of edge detection,better meeting the needs of stable and reliable image edge positioning.

关 键 词：图像边缘检测 模极大值 小波变换 SOBEL算法 形态学 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]