收稿日期: 2013-05-21
网络出版日期: 2025-05-30
基金资助
国家自然科学基金(61273296;60975040)
A Method of Ballistic Recognition Based on Online Learning
Received date: 2013-05-21
Online published: 2025-05-30
用SVM机器学习算法来解决弹道识别问题极大提高了识别精度,然而在处理过程中采用批处理优化方法很难缩短识别时间。考虑到实际中雷达捕获弹道数据是以在线的方式存在的,文中提出一种基于在线学习的弹道识别方法。仿真实验结果表明,在线算法在保证识别精度相当的情形下,大大的缩短了弹道识别时间。从而认为基于在线学习的识别方法是一种值得引进的弹道识别方法。
关键词： 弹道识别; 支持向量机(SVM); 在线优化; Pegasos
章显 , 高乾坤 , 陶卿 . 一种基于在线学习的弹道识别方法[J]. 弹箭与制导学报, 2014 , 34(2) : 109 -112 . DOI: 10.15892/j.cnki.djzdxb.2014.02.029
Solution to ballistic recognition based on SVM's method can greatly improve recognition accuracy, but recognition time can hardly be shortened by batch method during processing. Considering that in fact ballistic data captured by radar exists in online mode, in this paper, an online-learning method was proposed to solve ballistic recognition problem. The final simulation results show that the method maintains the same recognition accuracy and greatly shortens recognition time, the online-learning algorithm based on online learning is worthy of being introduced into ballistic recognition.
| [1] | 陶卿, 刘欣, 唐升平, 等. 基于支持向量机的弹道识别及其在雷达弹道外推中的应用[J]. 兵工学报, 2005, 26(3): 308-311. |
| [2] | 刘欣, 丁俊松, 储德军, 等. 基于 Boosting 学习算法的雷达弹道识别[J]. 弹箭与制导学报, 2010, 30(4): 193-196. |
| [3] | Duda RO, Hart PE, Stork DG. Pattern classification[M]. 2nd ed. John Wiley Sons, 2001. |
| [4] | Vapnik V. The nature of statistical learning theory[M]. New York: Springer-Verlag, 1995. |
| [5] | Cristianini N, Schawe-Taylor J. An introduction to support vector machines[M]. Cambridge: Cambridge Univ Press, 2000. |
| [6] | 陶卿, 那健, 冯勇, 等. 支持向量机弹道识别方法的精度分析[J]. 模式识别与人工智能, 2009, 22(3): 494-498. |
| [7] | M Zinkevich. Online convex programming and generalized infinitesimal gradient ascent[C]// Proceedings of the 20th Annal International Conference on Machine Learning, 2003: 856-863. |
| [8] | Bertsekas D. Convex analysis and optimization[M]. Athena: Scientific. 2003. |
| [9] | Hazan E, Agarwal A, Kale S, et al. Logarithmic regret algorithms for online convex optimization[J]. Machine Learning, 2007, 69(2/3): 169–192. |
| [10] | Shalev-Shwartz S, Singer Y, Srebro N. Pegasos: Primal estimated sub-gradient solver for SVM[C] // Proceedings of the 24th International Conference on Machine Learning, 2007. |
| [11] | Shalev-Shwartz. Online learning and online convex optimization[J]. Foundation and Trend in Machine Learning, 2011, 4(2): 107– 194. |
/
| 〈 |
|
〉 |