| 基于广义最小最大凹惩罚项的ISAR稀疏成像方法 |
| |
| 作者姓名： | 杨力 魏中浩 张冰尘 卢晓军 |
| |
| 作者单位： | 1. 中国科学院电子学研究所 中国科学院空间信息处理与应用系统技术重点实验室, 北京 100190;
2. 中国科学院大学, 北京 100049;
3. 中国国际工程咨询公司, 北京 100048 |
| |
| 基金项目： | 国家自然科学基金（61331017，61571419）资助 |
| |
| 摘 要： | 介绍一种基于广义最小最大凹（generalized minimax concave，GMC）惩罚项的ISAR稀疏成像方法。该方法的惩罚项形式与L1范数最小化方法不同，不仅使最小二乘损失函数凸性最小，而且避免了L1范数最小化方法系统性幅值低估问题。通过仿真实验说明GMC算法在ISAR成像中的幅度保持特性。利用Yak-42飞机的实际数据进行ISAR成像，结果表明GMC算法在成像精度方面优势明显，具有更好的成像效果。
|
| 关 键 词： | ISAR 广义最小最大凹惩罚项 L1范数最小化 |
| 收稿时间： | 2018-01-08 |
| 修稿时间： | 2018-03-20 |
ISAR sparse imaging algorithm based on generalized minimax concave penalty |
| |
| Authors: | YANG Li WEI Zhonghao ZHANG Bingchen LU Xiaojun |
| |
| Institution: | 1. Key Laboratory of Spatial Information Processing and Application System Technology of CAS, Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China;
2. University of Chinese Academy of Sciences, Beijing 100049, China;
3. China International Engineering Consulting Corporation, Beijing 100048, China |
| |
| Abstract: | A sparse imaging algorithm of ISAR based on generalized minimax concave (GMC) penalty is deseribed in this paper. The penalty of the algorithm is different from that of the L1 norm regularization. The penalty function not only maintains the convexity of the least squares cost function to be minimized but also avoids the systematic underestimation characteristic of the L1 norm regularization. This work illustrates the amplitude preservation characteristics of GMC algorithm in ISAR imaging by simulation experiments and imaging results of real data of Yak-42 aircraft. The results show that GMC algorithm has obvious advantages in imaging accuracy and has better imaging effect. |
| |
| Keywords: | ISAR generalized minimax concave penalty L1 norm regularization" target="_blank">L1 norm regularization')">L1 norm regularization |
|
| 点击此处可从《》浏览原始摘要信息 |
| 点击此处可从《》下载免费的PDF全文 |
