引入拒识的最小风险弹道目标识别

赵振冲; 王晓丹

doi:10.7652/xjtuxb201804019

您当前的位置：

首页 >

文章列表页 >

引入拒识的最小风险弹道目标识别

更新时间：2025-07-09

- 引入拒识的最小风险弹道目标识别
- A Minimum Risk Recognition Method of Ballistic Targets with Rejection Options
- 西安交通大学学报 2018年52卷第4期页码：132-138
- 作者机构：
  
  空军工程大学防空反导学院,西安,710051
- 作者简介：
- 基金信息：
  
  : 2017-09-04。作者简介: 赵振冲(1990—),男,博士生;王晓丹(通信作者),女,教授,博士生导师。基金项目: 国家自然科学基金资助项目(60975026,61273275)
- DOI：10.7652/xjtuxb201804019
  中图分类号： TP391
- 网络首发：2018-04-10，
  
  纸质出版：2018
- 稿件说明：
移动端阅览
赵振冲, 王晓丹. 引入拒识的最小风险弹道目标识别[J]. 西安交通大学学报, 2018,52(4):132-138.

A Minimum Risk Recognition Method of Ballistic Targets with Rejection Options[J]. 2018, 52(4): 132-138.
赵振冲, 王晓丹. 引入拒识的最小风险弹道目标识别[J]. 西安交通大学学报, 2018,52(4):132-138. DOI： 10.7652/xjtuxb201804019.

A Minimum Risk Recognition Method of Ballistic Targets with Rejection Options[J]. 2018, 52(4): 132-138. DOI： 10.7652/xjtuxb201804019.

摘要

针对弹道目标识别过程中误分类代价的不对称性

提出了一种引入拒识的最小风险弹道目标识别方法。该方法借助受试者工作特征(ROC)曲线

从概率分布的角度分析了真阳性率、虚警率与决策阈值之间的关系; 定义了代价敏感模型及引入拒识代价的模型的代价损失函数

推导了最小化分类代价所需满足的条件

结合ROC曲线的切线斜率

确定了最优拒识阈值

并利用两类目标的高分辨距离像(HRRP)数据验证了方法的有效性。实验结果表明:该方法能够唯一确定使代价最小的拒识阈值

克服了传统方法由于依赖于类先验概率及训练结果而造成的随机性和不确定性的缺点

提高了可靠性和稳定性

具有理论和实际应用价值。

Abstract

A new method to minimize the recognition risk of ballistic targets with rejection options is proposed to focus on the unsymmetry problem of different misclassification costs in ballistic targets recognition. The receiver operating characteristic(ROC)curve is applied to analyze the relationship between the true positive rate and false alarm rate and the decision threshold from the perspective of probability distribution. Loss functions of a cost-sensitive model and a cost-sensitive model with rejection options are defined

and the condition to be met for minimizing the loss function is derived. The optimal rejection thresholds are obtained via the tangent of ROC curve. The proposed method is validated on high resolution range profile(HRRP)datasets of two kinds of targets. The results show that the proposed method obtains the unique optimal rejection threshold

and minimizes the classification risk as well. The weaknesses of traditional methods are overcome and the reliability and stability are improved. It shows that the method has good values in theoretical analysis and realistic applications.

关键词

Keywords

references

胡小生, 钟勇. 基于加权聚类质心的SVM不平衡分类方法 [J]. 智能系统学报, 2013, 8(3): 261-265.

HU Xiaosheng, ZHONG Yong. Support vector machine imbalanced data classification based on weighted clustering centroid [J]. CAAI Transactions on Intelligent Systems, 2013, 8(3): 261-265.

DATTA S, DAS S. Near-Bayesian support vector machines for imbalanced data classification with equal or unequal misclassification costs [J]. Neural Networks, 2015, 70(12): 39-52.

JIANG L, LI C, WANG S S. Cost-sensitive Bayesian network classifiers [J]. Pattern Recognition Letters, 2014, 45(2): 211-216.

IBANEZ A, BIELZA C, LARRANAGA P. Cost-sensitive selective naïve Bayes classifiers for predicting the increase of the h-index for scientific journals [J]. Neurocomputing, 2014, 135(4): 42-52.

吕淑静. 基于代价敏感学习的手写邮编和地址识别 [D]. 上海: 华东师范大学, 2014: 23-45.

LI X, ZHAO H, ZHU W. A cost sensitive decision tree algorithm with two adaptive mechanisms [J]. Knowledge-Based Systems, 2015, 88(4): 24-33.

熊冰妍, 王国胤, 邓维斌. 分级式代价敏感决策树及其在手机换机预测中的应用 [J]. 山东大学学报, 2015, 45(5): 36-42.

XIONG Bingyan, WANG Guoyin, DENG Weibin. Hierarchical cost sensitive decision tree and its application in the prediction of the mobile phone replacement [J]. Journal of Shandong University, 2015, 45(5): 36-42.

SUN Y M, KARNEL M S, WONG A, et al. Cost-sensitive boosting for classification of imbalanced data [J]. Pattern Recognition, 2007, 40(12): 3358-3378.

薛一哲, 王拓. 基于代价敏感Adaboost目标跟踪 [J]. 中国图像图形学报, 2016, 21(5): 544-555.

XUE Yizhe, WANG Tuo. Object-tracking method based on improved cost-sensitive Adaboost [J]. Journal of Image and Graphics, 2016, 21(5): 544-555.

CHATELAIN C, ADAM S. A multi-model selection framework for unknown and/or evolutive misclassification cost problems [J]. Pattern Recognition, 2010, 43(3): 815-823.

BERNARD S, CHATELAIN C. The multiclass ROC front method for cost-sensitive classification [J]. Pattern Recognition, 2016, 52(4): 46-60.

TAPKANA P, ÖZBAKIRA L, KULLUKA S, et al. A cost-sensitive classification algorithm: BEE-Miner [J]. Knowledge-Based Systems, 2016, 95(2): 99-113.

JU H, YANG X, YU H, et al. Cost-sensitive rough set approach [J]. Information Sciences, 2016, 355/356(4): 282-298.

刘传武, 张智军, 毕笃彦. 雷达目标自动识别系统的拒识新算法 [J]. 系统工程与电子技术, 2009, 31(8): 1846-1850.

LIU Chuanwu, ZHANG Zhijun, BI Duyan. Novel refuse-recognition algorithm of radar automatic target recognition system [J]. Systems Engineering and Electronics, 2009, 31(8): 1846-1850.

黄垚, 刘思颂, 孔瑞. 基于支持向量机的嵌入拒识代价的手写字符识别研究 [J]. 电子质量, 2011(4): 5-7.

HUANG Yao, LIU Sisong, KONG Rui. SVM-based hand-written character recognition with reject option [J]. Electronics Quality, 2011(4): 5-7.

FUMERA G, ROLI F, GIACINTO G. Multiple reject thresholds for improving classification reliability [C]∥Proceedings of the Joint International Workshops on Advances in Pattern Recognition. Berlin, Germany: Springer-Verlag, 2000: 863-871.

FUMERA G, ROLI F, GIACINTO G. Reject option with multiple thresholds [J]. Pattern Recognition, 2000, 33(12): 2099-2101.

郑恩辉, 徐欢. 嵌入非对称拒识代价的二元分类算法 [J]. 控制与决策, 2013, 28(6): 855-860.

ZHENG Enhui, XU Huan. Binary classification algorithm with class-dependent reject cost [J]. Control and Decision, 2013, 28(6): 855-860.

TORTORELLA F. An optimal reject rule for binary classifiers [C]∥ Proceedings of the Joint International Workshops on Advances in Pattern Recognition. Berlin, Germany: Springer-Verlag, 2000: 611-620.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

采用试验设计方法的第三代CO2气体箔轴承箔片几何参数敏感性分析

张角凹坑气膜孔对双层壁气膜冷却特性的影响

面向高海拔空冷发电机防晕性能演化规律的电热耦合研究

电子级超纯氮深冷空分流程节能优化研究

超临界水异常区域热力学特性研究