| 基于支持向量回归与多核集成的红外成像导引头抗干扰性能评估方法 |
| |
| 引用本文： | 葛辰杰,陆志沣,洪泽华,马潮,余海鸣,赖鹏,乔宇,杨杰.基于支持向量回归与多核集成的红外成像导引头抗干扰性能评估方法[J].上海航天,2019,36(5):94-98. |
| |
| 作者姓名： | 葛辰杰 陆志沣 洪泽华 马潮 余海鸣 赖鹏 乔宇 杨杰 |
| |
| 作者单位： | 上海交通大学电子信息与电气工程学院;上海机电工程研究所 |
| |
| 基金项目： | 航天先进技术联合研究中心技术创新项目基金(USCAST2015-13,USCAST2016-23, USCAST2016-8)；上海航天科技创新基金(SAST2016008, SAST2017100,SAST2016085) |
| |
| 摘 要： | 面对着日益复杂的对抗环境,红外成像导引头的抗干扰性能需要不断提高。如何全面、客观、准确地对红外成像导引头的抗干扰性能进行评估,是一项急需解决的难题。针对传统基于支持向量机的评估方法中单核学习能力的不足,提出了一种基于支持向量回归与多核集成的评估方法,该方法在抗干扰评估指标体系下得到了综合的抗干扰性能值,为红外成像导引头抗干扰性能评估提供了新的思路。该方法能够训练多个支持向量回归机并融合多个核函数的优势,充分利用了特征的多样性,进一步降低了回归误差。实验结果表明:该算法能够实现高效可靠的红外成像导引头抗干扰性能评估。
|
| 关 键 词： | 红外成像导引头 性能评估 支持向量回归 多核学习 |
| 收稿时间： | 2018/7/24 0:00:00 |
| 修稿时间： | 2019/9/2 0:00:00 |
Anti-interference Capability Evaluation Based on SVM Regression and Multi-kernel Boosting |
| |
| GE Chenjie,LU Zhifeng,HONG Zehu,MA Chao,YU Haiming,LAI Peng,QIAO Yu and YANG Jie.Anti-interference Capability Evaluation Based on SVM Regression and Multi-kernel Boosting[J].Aerospace Shanghai,2019,36(5):94-98. |
| |
| Authors: | GE Chenjie LU Zhifeng HONG Zehu MA Chao YU Haiming LAI Peng QIAO Yu and YANG Jie |
| |
| Affiliation: | School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai200240, China,Shanghai Electro-Mechanical Engineering Institute, Shanghai 201109, China,Shanghai Electro-Mechanical Engineering Institute, Shanghai 201109, China,School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai200240, China,Shanghai Electro-Mechanical Engineering Institute, Shanghai 201109, China,Shanghai Electro-Mechanical Engineering Institute, Shanghai 201109, China,School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai200240, China and School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai200240, China |
| |
| Abstract: | Due to the increasingly complex fighting environment, there is a great need to improve the anti-interference capability. Therefore, how to evaluate the anti-interference capability of infrared guidance system precisely, objectively and efficiently has currently become an urgent issue. Since traditional SVM-based methods are unable to learn the regression effectively with single kernel machine, we propose an automatic method to evaluate the anti-interference capability based on SVM regression and multi-kernel boosting, using the manual index system and empirical evaluation results as training samples. The proposed method relies on multiple kernel learning to take full advantage of multiple features, which reduces the overall regression error. Experimental results show the effectiveness and efficiency of the proposed method on anti-interference capability evaluation. |
| |
| Keywords: | infrared guidance system capability evaluation SVM regression multi-kernel boosting |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《上海航天》浏览原始摘要信息 |
|
点击此处可从《上海航天》下载全文 |
|
