| 基于最佳阈值分割的舰船目标检测方法 |
| |
| 引用本文： | 王保云,杨昆.基于最佳阈值分割的舰船目标检测方法[J].云南师范大学学报(自然科学版),2014(4):55-60. |
| |
| 作者姓名： | 王保云 杨昆 |
| |
| 作者单位： | [1]云南师范大学信息学院,云南昆明650500 [2]西部资源环境地理信息技术教育部工程研究中心,云南昆明650500 |
| |
| 基金项目： | 国家高技术研究发展计划（863计划）资助项目（2012AA121403）. |
| |
| 摘 要： | 提出了一种基于最佳阈值分割的舰船目标检测方法,该方法通过构造分割阈值集合,并搜索特征约束条件下的最佳分割阈值进行目标分割,从而实现目标与背景区域的成功分离.使用该方法对不同卫星来源、不同分辨率的24 523幅光学遥感图像和516幅SAR图像进行了实验,对于1 155个目标图像的检测率高于95.0%,同时虚警率保持在较低水平,表明该方法对于遥感图像中的海上舰船目标检测效果较好、适应性强.
|
| 关 键 词： | 最佳阈值 分割 舰船检测 遥感图像 |
A New Ship Targets Detection Based on Optimal Method for Remote Sensing Images Threshold Segmentation |
| |
| WANG Bao-yun,YANG Kun.A New Ship Targets Detection Based on Optimal Method for Remote Sensing Images Threshold Segmentation[J].Journal of Yunnan Normal University (Natural Sciences Edition),2014(4):55-60. |
| |
| Authors: | WANG Bao-yun YANG Kun |
| |
| Institution: | 1. School of Information,Yunnan Normal University,Kunming 650500,China;2. The Engineering Research (;enter of GIS Technology in Western China,Ministry of Education,Kunming 650500,China) |
| |
| Abstract: | A new optimal threshold segmentation method is presented for ship targets detection. This method is designed to detect ship targets in remote sensing images, especially for complex sea surface. An experiment over 24 523 optical samples and 516 SAR samples, those from Spot, Quickbird, Ikonos, Landsat,and Radarsat 1,shows that the targets detection probability can be as high as 97.0% and the false alarm rate is low. Experiments testify the new method＇s robust ness with different aircraft product and different content images. |
| |
| Keywords: | Optimal threshold Segmentation Ship detection Remote sensing image |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
