基于DCNN的弹道中段目标HRRP图像识别

doi:10.3969/j.issn.1001-506X.2020.11.03

摘要/Abstract

摘要：

针对弹道中段目标识别问题,现有的基于高分辨距离像(high resolution range profile, HRRP)的识别方法直接提取一维HRRP(1-dimension HRRP, 1D-HRRP)的整体特征,对局部特征提取能力较弱,且由1D-HRRP数据提取的特征的表达能力有限,为此提出了一种基于深度卷积神经网络(deep convolutional neural network, DCNN)的弹道中段目标HRRP图像识别方法。首先,将1D-HRRP转化为0-1二值图像,从而把数值变化特征转化为图像结构特征;然后,构建DCNN逐层提取图像的局部特征和共性特征并进行识别;最后,结合Dropout和L2正则化双重机制缓解DCNN过拟合问题,利用AdaBound算法提高DCNN训练收敛速度和识别正确率。实验结果表明,所提出的基于HRRP图像的弹道中段目标识别方法比其他12种基于1D-HRRP或二维HRRP(2-dimension HRRP, 2D-HRRP)的识别方法正确率更高,在测试数据集上达到了96.28%,实验结果验证了该方法的有效性。

关键词: 弹道导弹, 目标识别, 高分辨距离像, 深度卷积神经网络, AdaBound算法

Abstract:

To solve the problem of midcourse ballistic target recognition, the existing recognition methods based on high resolution range profile (HRRP) directly extract the overall features of 1-dimension HRRP (1D-HRRP), which has a weak ability to extract local features, and the expression ability of features extracted from 1D-HRRP is limited. Therefore, a recognition method for midcourse ballistic targets of HRRP images based on the deep convolutional neural network (DCNN) is proposed. Firstly, 1D-HRRP is transformed into 0-1 binary images, so that the numerical variation features are transformed into the graph structure features. Then, DCNN is constructed to extract the local and common features of HRRP images layer by layer and recognize them. Finally, Dropout and L2 regularization are combined to alleviate the over-fitting problem of DCNN, and AdaBound algorithm is used to improve the convergence speed and accuracy of DCNN training. The experimental results show that the recognition method based on HRRP images is more accurate than other 12 recognition methods based on 1D-HRRP or 2-dimension HRRP (2D-HRRP), and achieves an accuracy of 96.28% in the test dataset, verifying the effectiveness of the proposed method.

Key words: ballistic missile, target recognition, high resolution range profile (HRRP), deep convolutional neural network (DCNN), AdaBound algorithm

中图分类号:

TP391.4

向前, 王晓丹, 李睿, 来杰, 张国令. 基于DCNN的弹道中段目标HRRP图像识别[J]. 系统工程与电子技术, 2020, 42(11): 2426-2433.

Qian XIANG, Xiaodan WANG, Rui LI, Jie LAI, Guoling ZHANG. HRRP image recognition of midcourse ballistic targets based on DCNN[J]. Systems Engineering and Electronics, 2020, 42(11): 2426-2433.

图/表 9

图1

图2

表1

图3

图4

表2

表3

图5

图6

参考文献 20

1	CHEN J , XU S , CHEN Z . Convolutional neural network for classifying space target of the same shape by using RCS time series[J]. IET Radar, Sonar & Navigation, 2018, 12 (11): 1268- 1275.
2	MA Y , HU M F , LU H Z , et al. Recurrent neural networks for discrimination of exo-atmospheric targets based on infrared radiation signature[J]. Infrared Physics & Technology, 2019, 96 (1): 123- 132.
3	CHOI I O , PARK S H , KIM M , et al. Efficient discrimination of ballistic targets with micro-motions[J]. IEEE Trans.on Aerospace and Electronic Systems, 2019, 22 (3): 1- 17.
4	PERSICO A R , ILIOUDIS C V , CLEMENTE C , et al. Novel classification algorithm for ballistic target based on HRRP frame[J]. IEEE Trans.on Aerospace and Electronic Systems, 2019, 55 (6): 3168- 3189. doi: 10.1109/TAES.2019.2905281
5	柴晶, 刘宏伟, 保铮. 加权KNN分类器在HRRP库外目标拒判中的应用[J]. 系统工程与电子技术, 2010, 32 (4): 718- 723.
	CHAI J , LIU H W , BAO Z . Application of a weighted KNN classifier for HRRP out-of-database target rejection[J]. Systems Engineering and Electronics, 2010, 32 (4): 718- 723.
6	翟夕阳, 王晓丹, 李睿, 等. 基于二叉树直觉模糊SVM的弹道目标HRRP识别[J]. 火力与指挥控制, 2017, 42 (10): 64- 68.
	ZHAI X Y , WANG X D , LI R , et al. Ballistic target recognition of HRRP based on intuitionistic fuzzy binary tree SVM[J]. Fire Control & Command Control, 2017, 42 (10): 64- 68.
7	李睿, 王晓丹, 雷蕾, 等. 结合多分类RVM和DS的弹道目标HRRP融合识别方法[J]. 信息与控制, 2017, 46 (1): 65- 71, 102.
	LI R , WANG X D , LEI L , et al. Ballistic target HRRP fusion recognition combining multi-class relevance vector machine and DS[J]. Information and Control, 2017, 46 (1): 65- 71, 102.
8	WANG Y H , BI X J , CHEN W , et al. Deep forest for radar HRRP recognition[J]. The Journal of Engineering, 2019, 2019 (21): 8018- 8021. doi: 10.1049/joe.2019.0723
9	WANG L Y, CUI Z Y, CAO Z J, et al. Fine-grained gesture recognition based on high resolution range profiles of terahertz radar[C]//Proc.of the IEEE International Geoscience and Remote Sensing Symposium, 2019: 1470-1473.
10	WANG S X, LI J B, WANG Y H, et al. Radar HRRP target recognition based on gradient boosting decision tree[C]//Proc.of the 9th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics, 2016: 1013-1017.
11	LIU W B , YUAN J W , ZHANG G , et al. HRRP target recognition based on kernel joint discriminant analysis[J]. Journal of Systems Engineering and Electronics, 2019, 30 (4): 703- 708.
12	YAN H Q, ZHANG Z H, XIONG G, et al. Radar HRRP recognition based on sparse denoising autoencoder and multi-layer perceptron deep model[C]//Proc.of the 4th International Conference on Ubiquitous Positioning, Indoor Navigation and Location Based Services, 2016: 283-288.
13	ZHAO F X , LIU Y X , HUO K , et al. Radar HRRP target recognition based on stacked autoencoder and extreme learning machine[J]. Sensors, 2018, 18 (1): 173.
14	PAN M , JIANG J , KONG Q P , et al. Radar HRRP target recognition based on t-SNE segmentation and discriminant deep belief network[J]. IEEE Geoscience and Remote Sensing Letters, 2017, 14 (9): 1609- 1613. doi: 10.1109/LGRS.2017.2726098
15	KRIZHEVSKY A, SUTSKEVER I, HINTON G E. Imagenet classification with deep convolutional neural networks[C]//Proc.of the Advances in Neural Information Processing Systems, 2012: 1097-1105.
16	董志鹏, 王密, 李德仁, 等. 遥感影像目标的尺度特征卷积神经网络识别法[J]. 测绘学报, 2019, 48 (10): 1285- 1295.
	DONG Z P , WANG M , LI D R , et al. Object detection in remote sensing imagery based on convolutional neural networks with suitable scale features[J]. Acta Geodaetica Et Cartographica Sinica, 2019, 48 (10): 1285- 1295.
17	SRIVASTAVA N , HINTON G , KRIZHEVSKY A , et al. Dropout: a simple way to prevent neural networks from overfitting[J]. The Journal of Machine Learning Research, 2014, 15 (1): 1929- 1958.
18	LUO L C, XIONG Y H, LIU Y, et al. Adaptive gradient methods with dynamic bound of learning rate[EB/OL].[2020-04-22]. https://arxiv.org/abs/1902.09843.
19	RUDER S. An overview of gradient descent optimization algorithms[EB/OL].[2020-04-22]. http://arxiv.org/abs/1609.04747.
20	DUCHI J , HAZAN E , SINGER Y . Adaptive subgradient methods for online learning and stochastic optimization[J]. Journal of Machine Learning Research, 2011, 12 (7): 2121- 2159.

目标	物理特征
弹头	平底锥,长度1.6 m,底面半径0.3 m,锥顶半径0.05 m
高仿诱饵	曲面底锥,长度1.6 m,底面半径为0.3 m,锥顶半径为0.05 m,曲面底为以0.5 m为半径的球的部分
简单诱饵	平底锥柱,长度1.6 m,底面半径0.3 m,锥顶半径0.05 m
球形诱饵	球体,半径0.5 m
母舱	圆柱体,长度0.4 m,半径0.75 m

训练算法	Fill	Nofill	Plot	Area
SGD	94.73	91.45	94.78	94.73
SGD(Nesterov)	94.78	92.17	95.39	94.67
AdaGrad	95.61	92.23	95.56	95.45
AdaBound	96.28	93.56	96.17	95.95

算法	球形诱饵	高仿诱饵	母舱	简单诱饵	弹头	总体	平均时间/μs
DT	100.00	90.77	96.67	93.01	85.94	93.23	＜100
RF	100.00	88.29	96.90	88.44	84.21	91.56	＜100
GB	100.00	81.19	97.98	93.27	84.53	91.17	＜100
MLP	100.00	92.84	98.86	96.62	87.19	94.95	140
ELM	100.00	89.24	98.82	96.00	87.37	94.11	＜100
1D-DCNN	100.00	91.76	98.87	99.42	88.63	95.61	＜100
2D-DCNN	100.00	89.68	99.14	92.18	91.76	94.56	274
IMG-DCNN (Nofill)	100.00	88.64	94.64	93.87	90.40	93.56	2 281
IMG-DCNN (Area)	100.00	94.31	97.77	95.36	92.35	95.95	2 285
IMG-DCNN (Plot)	100.00	94.63	96.73	96.70	92.74	96.17	2 968
IMG-DCNN (Fill)	100.00	94.93	99.15	96.92	90.86	96.28	2 290

[1]	王彩云, 吴钇达, 王佳宁, 马璐, 赵焕玥. 基于改进的CNN和数据增强的SAR目标识别[J]. 系统工程与电子技术, 2022, 44(8): 2483-2487.
[2]	孙晶明, 虞盛康, 孙俊. 基于元学习的雷达小样本目标识别方法及改进[J]. 系统工程与电子技术, 2022, 44(6): 1839-1845.
[3]	吉瑞萍, 张程祎, 梁彦, 王跃东. 基于LSTM的弹道导弹主动段轨迹预报[J]. 系统工程与电子技术, 2022, 44(6): 1968-1976.
[4]	周晓玲, 张朝霞, 鲁雅, 王倩, 王琨琨. 基于改进R-FCN的SAR图像识别[J]. 系统工程与电子技术, 2022, 44(4): 1202-1209.
[5]	孙晶明, 虞盛康, 孙俊. 基于深度学习的HRRP识别姿态敏感性分析[J]. 系统工程与电子技术, 2022, 44(3): 802-807.
[6]	雷禹, 冷祥光, 周晓艳, 孙忠镇, 计科峰. 基于改进ResNet网络的复数SAR图像舰船目标识别方法[J]. 系统工程与电子技术, 2022, 44(12): 3652-3660.
[7]	刘旗, 张新禹, 刘永祥. 基于门控多尺度匹配网络的小样本SAR目标识别[J]. 系统工程与电子技术, 2022, 44(11): 3346-3356.
[8]	王彩云, 姚晨, 吴钇达, 王佳宁, 李晓飞, 黄盼盼. 基于改进Dijkstra算法与时频域滤波的雷达目标识别[J]. 系统工程与电子技术, 2022, 44(10): 3090-3095.
[9]	但波, 付哲泉, 高山, 简涛. 基于卷积神经网络的海面目标全极化高分辨距离像识别技术[J]. 系统工程与电子技术, 2022, 44(1): 108-116.
[10]	关欣, 于昊天, 衣晓. 证据独立性及其对目标识别的影响研究[J]. 系统工程与电子技术, 2022, 44(1): 192-198.
[11]	李永祯, 刘业民, 庞晨, 黄大通. 基于分层极化特性的箔条云识别方法[J]. 系统工程与电子技术, 2021, 43(8): 2099-2107.
[12]	谢春思, 刘志赢, 桑雨. 基于特征匹配的舰载对陆导弹目标识别模型[J]. 系统工程与电子技术, 2021, 43(8): 2244-2253.
[13]	晏媛, 孙俊, 孙晶明, 于俊朋. 雷达小样本目标识别方法及应用分析[J]. 系统工程与电子技术, 2021, 43(3): 684-692.
[14]	徐嘉华, 张小宽, 郑舒予, 宗彬锋, 郑书昌. 基于改进3D-ESPRIT算法的GTD模型参数估计与目标识别[J]. 系统工程与电子技术, 2021, 43(2): 336-342.
[15]	韦娟, 丁智恺, 宁方立. 基于神经网络的声场景数据声谱图提取方法[J]. 系统工程与电子技术, 2021, 43(12): 3462-3469.