互联网 qkzz.net
全刊杂志网:首页 > 女性 > 文章正文
刊社推荐

基于聚类的SAR图像快速目标检测潘 卓 高 鑫 王岩飞 范俐捷


  摘 要:针对sar图像目标检测效率低、虚警概率高及sar图像的特点,改进了mean shift聚类算法,并与双参数cfar检测技术相结合,提出了一种能够快速而准确的sar图像目标检测算法。通过聚类预处理sar图像,降低了背景杂波对目标检测的影响及检测的虚警率,并且聚类后的sar图像具有一定的结构,将图像结构的概念引入到目标检测中,避免了对图像逐点检测,大大提高了检测速度。实验结果表明,该方法具有检测速度快、虚警概率低的特点。

  关键词:合成孔径雷达图像; 目标检测; 恒虚警率检测; mean shift聚类

  中图分类号:tp751 文献标志码:a 文章编号:1001-3695(2008)08-2416-04

  clustering-based target fast detection for sar imagery

  pan zhuo1,2, gao xin1, wang yan-fei1, fan li-jie1,2

   (1.institute of electronics, chinese academy of sciences, beijing 100080, china;2.graduate school, chinese academy of sciences, beijing 100039, china)

  abstract:to solve the inefficient and high false alarm probability problem of the target detection in sar images, this paper proposed a fast target-detection scheme for sar images, which combined improved mean shift clustering and two-parameter cfar detection technique. according to the sar image character, the mean shift clustering algorithm was improved. clustering sar images for preprocessing, which reduced the effect of clutter and eliminated many false target detections from background. furthermore, performed an initial clustering incorporates the concept of image structure into the target detection process, which the pixel-by-pixel detection was avoided. numerical experiments show that the novel method performs better accurately and faster speed.

......
很抱歉,暂无全文,若需要阅读全文或喜欢本刊物请联系《计算机应用研究》杂志社购买。
欢迎作者提供全文,请点击编辑
分享:
 

了解更多资讯,请关注“木兰百花园”
分享:
 
精彩图文


关键字
支持中国杂志产业发展,请购买、订阅纸质杂志,欢迎杂志社提供过刊、样刊及电子版。
关于我们 | 网站声明 | 刊社管理 | 网站地图 | 联系方式 | 中图分类法 | RSS 2.0订阅 | IP查询
全刊杂志赏析网 2017