基于Adaboost的作战目标属性判定方法

doi:10.12305/j.issn.1001-506X.2022.04.23

摘要/Abstract

摘要：

传统作战目标属性判定主要采用指挥员现场判断的定性方法, 具有一定的主观性, 并且由于缺乏较为成熟固定的算法而难以纳入指挥平台中。针对此问题, 结合作战目标属性判定关键影响因素分析, 提出一种基于自适应提升(adaptive boosting, Adaboost)的作战目标属性判定方法。首先, 针对目标有效面积、目标配置区域面积等关键因素, 采用单层决策树算法构建弱分类器。然后, 利用Adaboost对弱分类器进行加权组合, 形成作战目标属性判定的强分类模型。最后, 进行了示例分析, 并与决策树、支持向量机和人工神经网络3种属性判定方法进行对比仿真实验, 证明了所提方法的正确性和优越性。

关键词: 作战目标, 目标分类, 自适应提升, 决策树

Abstract:

The traditional battle target attribute judgment mainly adopts the qualitative method of commander's on-site judgment, which has a certain subjectivity, and it is difficult to be included in the command platform due to the lack of mature and fixed algorithm. To solve this problem, combined with the analysis of key influencing factors of combat target attribute determination, a combat target attribute determination method based on adaptive boosting (AdaBoost) is proposed. Firstly, aiming at the key factors such as target effective area and target configuration area, a single-layer decision tree algorithm is used to construct a weak classifier. Then, the weak classifiers are weighted and combined by AdaBoost to form a strong classification model for determining the attributes of combat targets. Finally, an example is analyzed and compared with three attribute determination methods: decision tree, support vector machine and artificial neural network. Simulation experiments show that the proposed method is correct and superior.

Key words: operational target, target classification, adaptive boosting (Adaboost), decision tree

中图分类号:

李园, 史宪铭, 李亚娟, 赵美. 基于Adaboost的作战目标属性判定方法[J]. 系统工程与电子技术, 2022, 44(4): 1256-1262.

Yuan LI, Xianming SHI, Yajuan LI, Mei ZHAO. Decision method of operational target attribute based on Adaboost[J]. Systems Engineering and Electronics, 2022, 44(4): 1256-1262.

图/表 10

表1

图1

表2

图2

图3

表3

表4

表5

表6

图4

参考文献 40

1	邵诗佳. 基于智能算法的武器目标分配问题研究[D]. 哈尔滨: 哈尔滨工程大学, 2019.
	SHAO S J. Research on weapon target assignment based on intelligent algorithm[D]. Harbin: Harbin Engineering University, 2019.
2	JIANG G S, SHI X M, ZHAO M, et al. Research on weapon target allocation of synthetic detachment considering emergency target[C]//Proc. of the International Conference on Scientific and Technological Innovation and Industrial Economy, 2020.
3	薛辉, 王源, 张天鹏, 等. 随机组合约束下的联合火力打击弹药需求预测模型[J]. 兵工学报, 2019, 40 (8): 1716- 1724. doi: 10.3969/j.issn.1000-1093.2019.08.022
	XUE H , WANG Y , ZHANG T P , et al. Demand forecasting model for joint fire strike ammunition under stochastic combination constraints[J]. Acta Armamentarii, 2019, 40 (8): 1716- 1724. doi: 10.3969/j.issn.1000-1093.2019.08.022
4	LI K, SHI X M, LI Y, et al. Combined forecasting method of ammunition consumption based on wavelet network[C]//Proc. of the 3rd International Conference on Applied Mathematics, Modeling and Simulation, 2020.
5	赵汝东, 史宪铭, 苏小波, 等. 基于Bayesian体系融合的新型弹药消耗预计方法[J]. 兵器装备工程学报, 2020, 41 (2): 75- 80. doi: 10.11809/bqzbgcxb2020.02.017
	ZHAO R D , SHI X M , SU X B , et al. New ammunition consumption prediction method based on Bayesian system fusion[J]. Journal of Ordnance Equipment Engineering, 2020, 41 (2): 75- 80. doi: 10.11809/bqzbgcxb2020.02.017
6	梁勇, 赵贺伟, 王志强, 等. 基于遗传算法的空空导弹消耗规律神经网络预测方法[J]. 海军航空工程学院学报, 2019, 34 (1): 151- 162.
	LIANG Y , ZHAO H W , WANG Z Q , et al. Neural network prediction method for the law of air to air missile consumption based on genetic algorithm[J]. Journal of Naval Aeronautical and Astronautical University, 2019, 34 (1): 151- 162.
7	DELEN D , TOPUZ K , ERYARSOY E . Development of a Bayesian belief network-based DSS for predicting and understanding freshmen student attrition[J]. European Journal of Operational Research, 2020, 281 (3): 575- 587. doi: 10.1016/j.ejor.2019.03.037
8	RAJ N , PHILIPOSE A B , DOMINIC D , et al. Exploration of twitter sentiments and classification by using deep CNN and naive Bayes[J]. International Journal of Recent Technology and Engineering, 2020, 9 (2): 1100- 1105.
9	ZHANG Y , JATOWT A . Estimating a one-class naive Bayes text classifier[J]. Intelligent Data Analysis, 2020, 24 (3): 567- 579. doi: 10.3233/IDA-194669
10	RAMESH N , DEVI G L , RAO K S . A frame work for classification of multi class medical data based on deep learning and naive Bayes classification model[J]. International Journal of Information Engineering and Electronic Business, 2020, 12 (1): 37- 43. doi: 10.5815/ijieeb.2020.01.05
11	张俊玉, 胡家豪, 黄嵩. CART决策树方法在煤电厂节能降耗中的应用[J]. 控制与决策, 2021, 36 (5): 1232- 1238.
	ZHANG J Y , HU J H , HUANG S . Application of CART decision tree model in reducing coal consumption in coal power plant[J]. Control and Decision, 2021, 36 (5): 1232- 1238.
12	BEHROUZ M , MOGHADAS N F , HAMZEH Z . Pavement maintenance and rehabilitation optimization based on cloud decision tree[J]. International Journal of Pavement Research and Technology, 2021, 14 (6): 740- 750. doi: 10.1007/s42947-020-0306-7
13	KIM D Y . Adopting and implementation of decision tree classification method for image interpolation[J]. Journal of the Korea Society of Digital Industry and Information Management, 2020, 16 (1): 55- 65.
14	AGGARWA H , ARORA H D , KUMAR V . An entropy based decision making problem using fuzzy decision tree classification of data mining[J]. Journal of Critical Reviews, 2020, 7 (18): 1801- 1811.
15	LORENZ C L , SPAETH A B , SOUZA C B D , et al. Artificial neural networks for parametric daylight design[J]. Architectural Science Review, 2020, 63 (2): 210- 221. doi: 10.1080/00038628.2019.1700901
16	RANA A , KIM K K . Comparison of artificial neural networks for low-power ECG-classification system[J]. Journal of Sensor Science and Technology, 2020, 29 (1): 19- 26. doi: 10.5369/JSST.2019.29.1.19
17	SAKAA B , CHAFFAI H , HANI A . ANNs approach to identify water demand drivers for Saf-Saf river basin[J]. Journal of Applied Water Engineering and Research, 2020, 8 (1): 44- 54. doi: 10.1080/23249676.2020.1719220
18	GULTEKIN A , SIRAC O M . Displacement prediction of precast concrete under vibration using artificial neural networks[J]. Structural Engineering and Mechanics, 2020, 74 (4): 559- 565.
19	WANG H , ZHENG B , YOON S W , et al. A support vector machine-based ensemble algorithm for breast cancer diagnosis[J]. European Journal of Operational Research, 2018, 267 (2): 687- 699. doi: 10.1016/j.ejor.2017.12.001
20	LIU M , CYGLER J E , VANDERVOORT E . Patient-specific PTV margins for liver stereotactic body radiation therapy determined using support vector classification with an early warning system for margin adaptation[J]. Medical Physics, 2020, 47 (10): 5172- 5182. doi: 10.1002/mp.14419
21	ASHFAQ A , JIN Y , ZHU C G , et al. Photovoltaic cell defect classification using convolutional neural network and support vector machine[J]. IET Renewable Power Generation, 2020, 14 (14): 2693- 2702. doi: 10.1049/iet-rpg.2019.1342
22	WANG H R , LU S J , ZHOU Z J . Ramp loss for twin multi-class support vector classification[J]. International Journal of Systems Science, 2020, 51 (8): 1448- 1463. doi: 10.1080/00207721.2020.1765047
23	FREUND Y , SCHAPIRE R E . A decision theoretic generalization of on-line learning and an application to boosting[J]. Journal of Computer and System Sciences, 1997, 55 (1): 119- 139. doi: 10.1006/jcss.1997.1504
24	LIU M . Fingerprint classification based on Adaboost learning from singularity features[J]. Pattern Recognition, 2010, 43 (1): 1062- 1070.
25	INHO C , DAIJIN K . A variety of local structure patterns and their hybridization for accurate eye detection[J]. Pattern Re-cognition, 2017, 61 (C): 417- 432.
26	HOWE N R, RATH T M, MANMATHA R. Boosted decision trees for word recognition in handwritten document retrieval[C]//Proc. of the 28th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 2005: 377-383.
27	SCHAPIRE R E , SINGER Y . A boosting-based system for text categorization[J]. Machine Learning, 2000, 39 (23): 135- 168.
28	PENG X J , SETLUR S , GOVINDARAJU V , et al. Using a boosted tree classifier for text segmentation in hand-annotated documents[J]. Pattern Recognition Letters, 2012, 33 (7): 943- 950. doi: 10.1016/j.patrec.2011.09.007
29	CHEN S Q , GUO Z F , ZHAO X L . Predicting mortgage early delinquency with machine learning methods[J]. European Journal of Operational Research, 2021, 290 (1): 358- 372. doi: 10.1016/j.ejor.2020.07.058
30	SHI X M, LI G N, LI K, et al. Customer classification method of logistics enterprises based on BP-AdaBoost[C]//Proc. of the 3rd International Conference on Applied Mathematics, Modeling and Simulation, 2020.
31	LIU C , CAO Y . Task re-pricing model based on density-based spatial clustering of applications[J]. Applied Soft Computing, 2020, 96, 106608. doi: 10.1016/j.asoc.2020.106608
32	VALENTINI G , DIETTERICH T G . Bias-variance analysis of support vector machines for the development of SVM-based ensemble methods[J]. Journal of Machine Learning Research, 2004, 5, 725- 775.
33	ZHANG X D , LI A , PAN R . Stock trend prediction based on a new status box method and AdaBoost probabilistic support vector machine[J]. Applied Soft Computing, 2016, 49, 385- 398. doi: 10.1016/j.asoc.2016.08.026
34	梅比, 汪旭光, 杨仁树. 基于Adaboost-SVM组合算法的爆破振动强度预测研究[J]. 振动与冲击, 2019, 38 (18): 231- 235.
	MEI B , WANG X G , YANG R S . Prediction of blasting vibration intensity based on Adaboost-SVM combination algorithm[J]. Journal of Vibration and Shock, 2019, 38 (18): 231- 235.
35	MARSLAND S . Machine learning[M]. Florida: Chapman and Hall/CRC, 2009.
36	FREUND Y, SCHAPIRE R E. Experiments with a new boosting algorithm[C]//Proc. of the 13th International Conference on Machine Learning, 1996.
37	ZHANG H B , QIU D D , WU R Z , et al. Novel framework for image attribute annotation with gene selection XGBoost algorithm and relative attribute model[J]. Applied Soft Computing, 2019, 80 (1): 57- 79.
38	SAMUEL W P , MAHENDRA N , DHANASEKHARA N , et al. Investigation of severity level of diabetic retinopathy using adaboost classifier algorithm[J]. Materials Today: Procee-dings, 2020, 33 (7): 3037- 3042.
39	DOU L J , LI X L , ZHANG L C , et al. Identification of lysine glutarylation using the adaboost classifier[J]. Journal of Proteome Research, 2020, 20 (1): 191- 201.
40	STEHMAN S V . Selecting and interpreting measures of the matic classification accuracy[J]. Remote Sensing of Environment, 1997, 62 (1): 77- 89. doi: 10.1016/S0034-4257(97)00083-7

代码	变量名称	变量类型	备注
0	β₁	连续型	目标面积, 由目标基本尺寸和数量决定
1	β₂	连续型	目标有效面积, 即暴露的目标面积
2	β₃	连续型	目标配置区域面积, 目标配置区域外接圆面积

编号	类别	β_k/m²
编号	类别	β₁	β₂	β₃
1	1	2	0.84	11.00
2	1	2	0.34	15.00
3	1	2	0.80	26.69
4	1	6	1.80	35.00
5	1	6	0.60	55.80
6	1	6	1.80	93.76
7	1	20	6.80	119.00
8	1	20	4.00	209.00
9	1	20	6.40	378.37
10	1	45	13.50	320.00
11	1	45	6.75	514.96
12	1	135	21.60	1 015.00
13	1	135	39.15	1 621.81
14	1	405.0	109.35	3 745.000
15	1	405.0	137.70	9 949.090
16	1	1 500.0	180.00	16 800.000
17	1	1 500.0	420.00	26 690.000
18	1	27.0	5.40	256.000
19	1	90.0	27.00	880.000
20	-1	0.2	0.20	0.200
21	-1	15.0	9.00	15.000
22	-1	45.0	38.25	145.000
23	-1	135.0	108.00	535.000
24	-1	405.0	291.60	1 705.000
25	-1	1 500.0	945.00	6 450.000
26	-1	9.0	5.67	9.000
27	-1	27.0	22.68	87.000
28	-1	90.0	72.90	360.000
29	-1	100.0	90.00	100.000
30	-1	300.0	210.00	400.000
31	-1	300.0	192.00	981.250
32	-1	300.0	255.00	625.000
33	-1	500.0	400.00	700.000
34	-1	500.0	325.00	1 746.625
35	-1	500.0	395.00	1 000.000

弱分类器	权重	特征	阈值/m²	符号
1	0.506	1	22.295	<
2	0.368	1	199.053	<
3	0.555	2	1 404.926	>
4	0.351	1	44.389	<
5	0.318	2	2 809.653	>
6	0.378	0	0.200	>
7	0.272	2	2 809.653	>
8	0.300	1	44.389	<
9	0.271	0	79.137	>

β_k/m²			真实结果	测试结果
β₁	β₂	β₃	真实结果	测试结果
20	6.40	378.37	1	1
45	13.50	320.00	1	1
45	6.75	514.96	1	1
405	291.60	1 705.00	-1	-1
27	22.68	87.00	-1	-1

目标	分类为面目标	分类为点目标
实际为面目标	TP	FN
实际为点目标	FP	TN