基于改进Canny和灰度矩的水下图像边缘检测方法研究  

Research on underwater image edge detection method based on improved Canny and gray scale moment

作　　者：陈振娅 刘增力[1] CHEN Zhenya;LIU Zengli(Faculty of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650500,Yunnan,China)

机构地区：[1]昆明理工大学信息工程与自动化学院,云南昆明650500

出　　处：《农业装备与车辆工程》2024年第3期107-110,120,共5页Agricultural Equipment & Vehicle Engineering

摘　　要：针对水下图像对比度差、模糊,致使检测出的边缘存在不连续和伪边缘的问题,提出一种基于改进的Canny和亚像素的水下图像边缘检测方法。采用改进的Canny算法检测水下图像的像素级边缘,在此基础上采用灰度矩提取图像的亚像素边缘特征,提高边缘的定位精度和检测率。实验结果表明,提出的算法相较于传统的像素级边缘检测算法在边缘轮廓的提取、减少伪边缘、提升边缘精度方面有较大优势,尤其是对深海环境中光源单一以及图像对比度差的物体边缘的检测具有更显著的效果和定位精度。Aiming at the problem of poor contrast and blurring of underwater images,resulting in discontinuities and pseudo-edges in the detected edges,an improved underwater image edge detection method based on Canny and subpixel was proposed.The improved Canny algorithm was used to detect the pixel-level edges of underwater images,and the sub-pixel edge features of the images were extracted by using gray-scale moments on this basis to improve the localization accuracy and detection rate of edges.The experimental results showed that the proposed algorithm had greater advantages over the traditional pixel-level edge detection algorithm in the extraction of edge contours,reduction of pseudo-edges,and improvement of edge accuracy,especially for the detection of edges of objects with single light source and poor image contrast in deep sea environment with more significant effect and localization accuracy.

关 键 词：CANNY算子 亚像素 水下图像 边缘检测 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]