摘要
船舶视觉系统是船舶航行的重要辅助部分,视觉系统的工作效果很大程度上取决于边缘检测方法的准确性。常见的图像边缘检测方法通过像素梯度变化检测目标边缘,在对复杂图像处理时容易受到噪声干扰,导致检测精度和效率较低。为解决上述问题,提出船舶视觉系统目标图像边缘检测方法。对图像进行背景区域与目标区域的划分,提取目标所在区域。选择图像边缘的Harr和HOG特征后,利用蚁群算法检测图像边缘。对比仿真实验结果表明,研究的边缘检测方法检测准确率均高于85%,并且检测效率均高于对比方法,具有明显的应用优势。
Ship vision system is an important auxiliary part of ship navigation, and the working effect of vision system depends largely on the accuracy of edge detection method. Common image edge detection methods detect target edges by pixel gradient changes, which are susceptible to noise interference when processing complex images, resulting in low detection accuracy and efficiency. To solve the above problems, the target image edge detection method for ship vision system is studied. The image is divided into background region and target region, and the region where the target is located is extracted. After selecting the Harr and HOG features of image edges, the image edges are detected using ant colony algorithm. The results of the comparison simulation experiments show that the detection accuracy of the studied edge detection methods are both higher than 85%, and the detection efficiency are both higher than the comparison methods, which have obvious application advantages.
作者
顾建华
严国军
贲能军
邱婷婷
GU Jian-hua;YAN Guo-jun;BEN Neng-jun;QIU Ting-ting(Yancheng Industry Vocational and Technical College,Yancheng 224000,China;Yancheng Biological Engineering Higher Vocational Technology School,Yancheng 224000,China)
出处
《舰船科学技术》
北大核心
2021年第10期64-66,共3页
Ship Science and Technology
基金
苏北科技专项(SZ-YC2018067)
江苏省教育科学“十三五”规划职教立项课题(D/2018/03/04)
江苏省现代教育技术研究课题(2018-R-62252)
关键词
船舶视觉系统
目标图像
图像边缘
边缘检测
蚁群算法
marine vision systems
target images
image edges
edge detection
ant colony algorithm
作者简介
顾建华(1972-),女,工程硕士,副教授,主要研究方向为计算机网络应用及计算机应用系统。
