计算机应用 ›› 2010, Vol. 30 ›› Issue (11): 2977-2979.

• 模式识别 • 上一篇    下一篇

SAR图像目标方位角联合估计方法研究

许心瑜1,刘拥军1,黎鑫1,高波2   

  1. 1. 中国电波传播研究所
    2. 总装工程兵科研一所
  • 收稿日期:2010-04-26 修回日期:2010-06-29 发布日期:2010-11-05 出版日期:2010-11-01
  • 通讯作者: 许心瑜

Research on combined method for SAR target azimuth estimation

  • Received:2010-04-26 Revised:2010-06-29 Online:2010-11-05 Published:2010-11-01

摘要: 为了精确估计任意角度的目标方位角,融合主导边界法与峰值拟合法,提出一种SAR目标方位角的联合估计方法。通过计算投影比以及目标区域同外接矩形面积之比,初步判断目标是否位于水平或垂直方位附近,自适应选择相应的估计算法进行处理。移动与静止目标获取与识别(MSTAR)公共数据库实测数据的仿真结果表明,该方法估计误差较小,可获得较高的估计准确度。方位角联合估计方法有利于增强目标方位角估计的适应性,提高目标分类和识别的效率。

关键词: 合成孔径雷达, 方位角估计, 主导边界法, 峰值拟合法, 联合估计方法

Abstract: In order to obtain accurate estimation for any azimuth, a combined estimation method for Synthetic Aperture Radar (SAR) targets was proposed, integrating leading contour fitting and peak value fitting. According to the projection ratio and the area proportion of object region to the envelop rectangle, different estimation algorithms were exploited based on the initial judgment of the aspect. The simulation results using Moving and Stationary Target Acquisition and Recognition (MSTAR) data indicate that the error of the proposed method is small, and thus it has high accuracy. The combined method can improve the adjustability of azimuth estimation, and then can facilitate target classification and recognition effectively.

Key words: synthetic aperture radar (SAR), azimuth estimation, leading contour fitting, peak value fitting, combined estimation method