低信噪比下面向时空特征融合的雷达目标智能检测方法

doi:10.12305/j.issn.1001-506X.2026.05.08

系统工程与电子技术 ›› 2026, Vol. 48 ›› Issue (5): 1523-1531.doi: 10.12305/j.issn.1001-506X.2026.05.08

低信噪比下面向时空特征融合的雷达目标智能检测方法

陈垣光¹^,*(), 刘晓凯¹, 肖思瑶¹, 张顺生¹, 崔红元², 陈晓莹², 刘莹²

1. 电子科技大学电子科学技术研究院，四川成都 611731
2. 中国科学院大学计算机科学与技术学院，北京 100190

收稿日期:2025-04-08 接受日期:2025-08-07 出版日期:2026-05-27 发布日期:2026-05-27
通讯作者: 陈垣光 E-mail:ygchen0111@163.com
作者简介:刘晓凯（2000—），女，硕士研究生，主要研究方向为雷达目标检测与识别
肖思瑶（2000—），男，硕士研究生，主要研究方向为图像去噪、目标检测与识别
张顺生（1980—），男，研究员，博士，主要研究方向为雷达探测与成像识别、智能感知与信息系统、信号与信息智能处理
崔红元（1990—），女，高级工程师，硕士，主要研究方向为人工智能、高性能计算
陈晓莹（1997—），女，博士研究生，主要研究方向为雷达信号处理、雷达抗干扰、认知雷达
刘　莹（1977—），女，教授，博士，主要研究方向为数据挖掘、人工智能、高性能计算

Radar intelligent target detection method using spatio-temporal feature fusion under low signal-to-noise ratio

Yuanguang CHEN¹^,*(), Xiaokai LIU¹, Siyao XIAO¹, Shunsheng ZHANG¹, Hongyuan CUI², Xiaoying CHEN², Ying LIU²

1. Research Institute of Electronic Science and Technology，University of Electronic Science and Technology of China，Chengdu 611731，China
2. School of Computer Science and Technology，University of Chinese Academy of Sciences，Beijing 100190，China

Received:2025-04-08 Accepted:2025-08-07 Online:2026-05-27 Published:2026-05-27
Contact: Yuanguang CHEN E-mail:ygchen0111@163.com

摘要/Abstract

摘要：

针对传统雷达目标检测方法在低信噪比条件下检测性能大幅度降低的问题，提出一种基于特征增强和特征融合策略的多帧深度学习检测方法。利用多尺度卷积层提取雷达帧的抽象表示，通过挖掘当前相干处理间隔和历史相干处理间隔回波数据的时空特征，在连续的相干处理间隔中重建和预测出运动目标。利用对空跟踪民航飞机的雷达实测数据进行实验，结果表明：当检测信噪比为6 dB时，相同虚警条件下，所提方法得到的检测概率较传统恒虚警检测方法和检测前跟踪方法提升50％以上，为雷达目标检测提供解决方案。

关键词: 深度学习, 多帧检测, 特征融合, 低信噪比

Abstract:

To address the problem of significant performance degradation of traditional radar target detection methods under low signal-to-noise ratio （SNR） conditions, a multi-frame deep learning detection method based on feature enhancement and fusion strategies is proposed. The method utilizes multi-scale convolutional layers to extract abstract representations of radar frames. By mining spatio-temporal features from both current and historical coherent processing interval （CPI） echo data, it reconstructs and predicts moving targets across consecutive CPIs. Experiments conducted on radar-measured data from civil aviation aircraft tracking demonstrate that, at a detection SNR of 6 dB, the proposed method achieves over 50% improvement in detection probability compared to traditional constant false alarm rate （CFAR） and track-before-detect （TBD） methods under the same false alarm conditions, providing a solution for radar target detection.

Key words: deep learning, multi-frame detection, feature fusion, low signal-to-noise ratio （SNR）

中图分类号:

TN 958.2

陈垣光, 刘晓凯, 肖思瑶, 张顺生, 崔红元, 陈晓莹, 刘莹. 低信噪比下面向时空特征融合的雷达目标智能检测方法[J]. 系统工程与电子技术, 2026, 48(5): 1523-1531.

Yuanguang CHEN, Xiaokai LIU, Siyao XIAO, Shunsheng ZHANG, Hongyuan CUI, Xiaoying CHEN, Ying LIU. Radar intelligent target detection method using spatio-temporal feature fusion under low signal-to-noise ratio[J]. Systems Engineering and Electronics, 2026, 48(5): 1523-1531.

图/表 12

图1

图2

图3

图4

图5

图6

图7

图8

表1

表2

图9

图10

参考文献 31

1	JIANG W, REN Y H, LIU Y, et al. A method of radar target detection based on convolutional neural network[J]. Neural Computing and Applications, 2021, 33 (16): 9835- 9847. doi: 10.1007/s00521-021-05753-w
2	DAVEY J S, RUTTEN G M, CHEUNG B. A comparison of detection performance for several track-before-detect algorithms[J]. EURASIP Journal on Advances in Signal Processing, 2008, 2008 (1): 428036- 10.
3	GANDHI P P, KASSAM S A. Analysis of CFAR processors in nonhomogeneous background[J]. IEEE Trans. on Aerospace and Electronic systems, 1988, 24 (4): 427- 445. doi: 10.1109/7.7185
4	BALL J E. Low signal-to-noise ratio radar target detection using linear support vector machines (L-SVM)[C]//Proc. of the IEEE Radar Conference, 2014: 1291−1294.
5	王治飞, 于俊朋, 杨予昊. 基于帧间多维特征的深度学习雷达目标检测技术[J]. 现代雷达, 2022, 44 (12): 48- 54.
	WANG Z F, YU J P, YANG Y H, et al. Radar target detection with multi-frame multi-dimensional features based on deep learning[J]. Modern Radar, 2022, 44 (12): 48- 54.
6	WANG L, TANG J, LIAO Q M. A study on radar target detection based on deep neural networks[J]. IEEE Sensors Letters, 2019, 3 (3): 7000504.
7	DELAMOU M, BAZZI A, CHAFII M, et al. Deep learning-based estimation for multitarget radar detection[C]//Proc. of the IEEE 97th Vehicular Technology Conference, 2023.
8	WANG Y Z, JIANG Z Y, LI Y D, et al. RODNet: a real-time radar object detection network cross-supervised by camera-radar fused object 3D localization[J]. IEEE Journal of Selected Topics in Signal Processing, 2021, 15 (4): 954- 967. doi: 10.1109/JSTSP.2021.3058895
9	JIANG T Z, ZHUANG L, AN Q, et al. T-RODNet: Transformer for vehicular millimeter-wave radar object detection[J]. IEEE Trans. on Instrumentation and Measurement, 2022, 72, 5003912.
10	VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[J]. Proc. of the 31st International Conference on Neural Information Processing Systems, 2017, 30, 6000- 6010.
11	陈佳音, 郭山红, 朱海锐, 等. 基于多特征融合的海面目标智能检测算法[J]. 现代雷达, 2024, 46 (12): 24- 30.
	CHEN J Y, GUO S H, ZHU H R, et al. Intelligent detection algorithm for sea surface targets based on multi-feature fusion[J]. Modern Radar, 2024, 46 (12): 24- 30.
12	施端阳, 林强, 胡冰, 等. 基于YOLO的航管一次雷达目标检测方法[J]. 系统工程与电子技术, 2024, 46(01): 143−151.
	DUAN S Y, LIN Q, HU B, et al. Target detection method of primary surveillance radar based on YOLO[J]. Systems Engineering and Electronics, 2023, 46(1): 143−151.
13	REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: unified, real-time object detection[C]//Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 779−788.
14	施赛楠, 姜苏桐. 基于改进1D-AlexNet的海面小目标高维特征检测[J]. 信号处理, 2024, 40 (6): 1098- 1110.
	SHI S N, JIANG S T. High-dimensional feature detection of small sea-surface targets via improved 1D-AlexNet[J]. Journal of Signal Processing, 2024, 40 (6): 1098- 1110.
15	DALBAH Y, LAHOUD J, CHOLAKKAL H. RadarFormer: lightweight and accurate real-time radar object detection model[C]//Proc. of the Scandinavian Conference on Image Analysis, 2023: 341−358.
16	XIAO Y H, LIU Y F, LUAN K, et al. Deep LiDAR-radar-visual fusion for object detection in urban environments[J]. Remote Sensing, 2023, 15 (18): 4433. doi: 10.3390/rs15184433
17	汪翔, 汪育苗, 陈星宇, 等. 基于深度学习的多特征融合海面目标检测方法[J]. 雷达学报, 2023, 13 (3): 554- 564.
	WANG X, WANG Y M, CHEN X Y, et al. Deep learning-based marine target detection method with multiple feature fusion[J]. Journal of Radars, 2023, 13 (3): 554- 564.
18	LUO D Q, ZHU J H. Radar target detection in strong clutter using spatial-temporal U-NET[C]//Proc. of the IEEE 32nd International Workshop on Machine Learning for Signal Processing, 2022.
19	RONNEBERGER O, FISCHER P, BROX T. U-net: convolutional networks for biomedical image segmentation[C]//Proc. of the 18th International Conference on Medical Image Computing and Computer-Assisted Intervention, 2015: 234−241.
20	WEN L W, DING J S, XU Z. Multiframe detection of sea-surface small target using deep convolutional neural network[J]. IEEE Trans. on Geoscience and Remote Sensing, 2021, 60, 5107116.
21	DECOURT C, VANRULLEN R, SALLE D, et al. A recurrent CNN for online object detection on raw radar frames[J]. IEEE Trans. on Intelligent Transportation Systems, 2024, 25 (10): 13432- 13441. doi: 10.1109/TITS.2024.3404076
22	WANG K Q, WANG Z Y. A triple-channel network for maritime radar targets detection based on multi-modal features[J]. Remote Sensing, 2024, 16 (24): 4662. doi: 10.3390/rs16244662
23	司凌宇, 强文文, 李港, 等. 基于对比学习的航海雷达目标检测方法[J]. 电子学报, 2023, 51 (7): 1791- 1802. doi: 10.12263/DZXB.20221088
	SI L Y, QIANG W W, LI G, et al. Marine radar object detection method based on contrastive learning[J]. Acta Electronica Sinica, 2023, 51 (7): 1791- 1802. doi: 10.12263/DZXB.20221088
24	HUANG J, YANG X, ZHOU F, et al. A deep learning framework based on improved self-supervised learning for ground-penetrating radar tunnel lining inspection[J]. Computer-Aided Civil and Infrastructure Engineering, 2024, 39 (6): 814- 833. doi: 10.1111/mice.13042
25	CHEN F L, LI Y, ZHAO B, et al. Unsupervised domain adaption on category level by pseudo label for radar target detection[J]. IEEE Geoscience and Remote Sensing Letters, 2025, 22, 3501105.
26	ZHUANG L, JIANG T Z, WANG J H, et al. Effective mmWave radar object detection pretraining based on masked image modeling[J]. IEEE Sensors Journal, 2023, 24 (3): 3999- 4010.
27	GROSSI E, LOPS M, VENTURINO L. A novel dynamic programming algorithm for track-before-detect in radar systems.[J]. IEEE Trans. Signal Processing, 2013, 61 (10): 2608- 2619. doi: 10.1109/TSP.2013.2251338
28	ZHANG Y, YE M, ZHU G Y, et al. FFCA-YOLO for small object detection in remote sensing images[J]. IEEE Trans. on Geoscience and Remote Sensing, 2024, 62, 5611215.
29	YU F, KOLTUN V. Multi-Scale context aggregation by dilated convolutions[EB/OL]. [2025-04-08]. https://arxiv.org/abs/1511.07122.
30	XIAO Y, XU T F, YU X, et al. A lightweight fusion strategy with enhanced interlayer feature correlation for small object detection[J]. IEEE Trans. on Geoscience and Remote Sensing, 2024, 62, 4708011.
31	GIRSHICK R. Fast R-CNN[C]//Proc. of the IEEE International Conference on Computer Vision, 2015: 1440−1448.

检测方法	检测概率
检测方法	SNR=6 dB	SNR=10 dB
CFAR	0.001	0.870
DP-TBD^[27]	0.348	0.925
ECE^[1]	0.005	0.033
DCNN^[20]	0.461	0.939
所提方法	0.883	0.952

检测方法	推理时间
CFAR	3.82
DP-TBD^[27]	7.02
ECE^[1]	0.12
DCNN^[20]	2.83
所提方法	0.36

[1]	黄琼丹, 刘露露, 韩洁婧, 王佳鹏, 康仕林. 基于差异特征反投影融合的图像超分辨重建[J]. 系统工程与电子技术, 2026, 48(6): 1809-1818.
[2]	袁胜智, 王少蕾, 李静, 肖任楷, 谭浩天, 赖建伟, 许江涛. 基于小波变换和改进残差神经网络的故障诊断方法[J]. 系统工程与电子技术, 2026, 48(6): 2031-2041.
[3]	王鹏昊, 何迪, 尤明懿, 郭细平. 基于随机共振和多尺度特征融合调制识别方法[J]. 系统工程与电子技术, 2026, 48(6): 2096-2104.
[4]	杨泽莹, 陈卓, 徐志明, 艾小锋, 吴其华, 王玲. 空间目标ISAR图像部件正向标记方法[J]. 系统工程与电子技术, 2026, 48(4): 1165-1173.
[5]	林炅, 高朋, 洪文兴. 物理信息神经网络在空天领域的应用与展望[J]. 系统工程与电子技术, 2026, 48(4): 1264-1282.
[6]	程鲲, 雷洪涛, 吕志轩. 改进YOLOv8的轻量级无人机跟踪方法[J]. 系统工程与电子技术, 2026, 48(3): 737-750.
[7]	张龙, 朱连宏, 杨波, 雷震, 冯轩铭. MFA-Net：一种面向复杂对抗环境的反舰导弹智能识别多模态自适应融合网络[J]. 系统工程与电子技术, 2026, 48(3): 894-907.
[8]	田弘宇, 葛松虎, 郭宇, 崔中普, 梁潇. 低干噪比下多特征融合的通信干扰识别方法[J]. 系统工程与电子技术, 2026, 48(3): 1061-1071.
[9]	王瑶, 胥辉旗, 曹司磊, 王磊. 基于深度学习的舰船关键部位检测算法[J]. 系统工程与电子技术, 2026, 48(2): 410-421.
[10]	羿昌宇. 低信噪比多径下多天线联合时间同步算法[J]. 系统工程与电子技术, 2026, 48(2): 705-710.
[11]	刘恒燕, 方君, 凌青, 闫文君, 于柯远, 张立民. 基于1D-2D-GRU-ResNet的辐射源个体识别方法[J]. 系统工程与电子技术, 2026, 48(2): 727-735.
[12]	包玉刚, 贾皓翔, 赵旦峰. 基于高阶递归网络的单幅图像去雨滴模型[J]. 系统工程与电子技术, 2026, 48(1): 12-21.
[13]	陈海青, 汪刘应, 刘顾, 王龙, 葛超群, 陈孟州. 合成孔径雷达图像去噪算法研究进展[J]. 系统工程与电子技术, 2026, 48(1): 94-105.
[14]	李开明, 张袁鹏, 罗迎, 代肖楠. 弹道导弹雷达目标识别研究进展[J]. 系统工程与电子技术, 2025, 47(9): 2870-2889.
[15]	季策, 马相宇, 牟晓宇, 赵家毅. TS-GRU-VTA：基于深度学习的车辆信道估计方案[J]. 系统工程与电子技术, 2025, 47(9): 3093-3098.

低信噪比下面向时空特征融合的雷达目标智能检测方法

Radar intelligent target detection method using spatio-temporal feature fusion under low signal-to-noise ratio

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献 31

相关文章 15

编辑推荐

Metrics

本文评价