摘要
目前研究的遥感图像多方向舰船目标匹配检测方法对数据的匹配度较低,导致检测精度难以达到要求。为了解决上述问题,基于特征矢量研究一种检测遥感图像多方向舰船目标检测匹配方法。利用卷积神经网络对图像细节进行分析,根据分析结果得到底层特征,融合底层特征和高层特征,形成特征图,完成图像特征提取。分析图像像素灰度值,确定图像像素特征矢量,建立模糊子集,实现多方向舰船目标检测。实验结果表明,该方法能够很好地提高匹配度,确保检测精度达到用户要求。
The currently studied remote sensing image multi-directional ship target matching detection method has a low degree of matching to the data,which makes the detection accuracy difficult to meet the requirements.In order to solve the above problems,a multi-directional ship target detection and matching method based on feature vectors is studied.The convolutional neural network is used to analyze the details of the image,and the low-level features are obtained according to the analysis results,and the low-level features and the high-level features are merged to form a feature map to complete the image feature extraction.Analyze the gray value of image pixels,determine the feature vector of image pixels,establish fuzzy subsets,and achieve multi-directional ship target detection.Experimental results show that this method can improve the matching degree and ensure that the detection accuracy meets the user’s requirements.
作者
蒋玉婷
JIANG Yu-ting(Information Engineering Institute,Jiangsu Maritime Institute,Nanjing 211170,China)
出处
《舰船科学技术》
北大核心
2021年第22期166-168,共3页
Ship Science and Technology
基金
江苏省教育厅“青蓝工程”优秀教学团队-软件技术专业创新教学团队项目(2019【3】)
关键词
特征矢量
矢量匹配
匹配检测
遥感图像
多方向舰船
舰船目标
feature vector
vector matching
matching detection
remote sensing image
multi-directional ship
ship target
作者简介
蒋玉婷(1981-),女,工程硕士,副教授,研究方向为计算机应用、计算机软件,大数据、云计算等。
