Deep convolutional neural network for meteorology target detection in airborne weather radar images

doi:10.23919/JSEE.2023.000142

Journal of Systems Engineering and Electronics ›› 2023, Vol. 34 ›› Issue (5): 1147-1157.doi: 10.23919/JSEE.2023.000142

• Advanced Radar Imaging and Intelligent Processing • Previous Articles Next Articles

Deep convolutional neural network for meteorology target detection in airborne weather radar images

Chaopeng YU¹(), Wei XIONG¹^,*(), Xiaoqing LI¹(), Lei DONG²

¹ Aviation Industry Corporation of China Leihua Electronic Technology Institute, Wuxi 214063, China
² Key Laboratory of Civil Aircraft Airworthiness Technology, Civil Aviation University of China, Tianjin 300300, China

Received:2022-07-01 Online:2023-10-18 Published:2023-10-30
Contact: Wei XIONG E-mail:yuchaopeng@raa.org.cn;xiongweiwhumath@sina.com;1032332623@qq.com
About author:
YU Chaopeng was born in 1977. He is now a researcher-level senior engineer in Aviation Industry Corporation of China (AVIC) Leihua Electronic Technology Institute. His research interests are airborne radar system and flight environment integrated surveillance system (ISS). E-mail: yuchaopeng@raa.org.cn

XIONG Wei was born in 1984. He received his Ph.D. degree from School of Electronic and Information Engineering, Nanjing University of Aeronautics and Astronautics. Supported by China Scholarship Council (CSC), he studied abroad as a visiting scholar in the Department of Communication at the University of Pisa in Italy from October 2016 to September 2017. He is now a senior engineer in Aviation Industry Corporation of China Leihua Electronic Technology Institute. His research interests are airborne meteorological anti-collision radar system and integrated collision avoidance system of flight environment. E-mail: xiongweiwhumath@sina.com

LI Xiaoqing was born in 1995. He received his B.S. degree from Xiangtan University in 2018 and M.S. degree from Beihang University in 2021. He is now an engineer in Aviation Industry Corporation of China Leihua Electronic Technology Institute. His research interests are radar target recognition and signal processing. E-mail: 1032332623@qq.com

DONG Lei was born in 1983. He received his B.S. and M.S. degrees from Jiangsu University, in 2005 and 2008, respectively, and Ph.D. degree from Beihang University, in 2013. Since 2013, he has been a teacher with the Airworthiness College, Civil Aviation University of China. He is currently an associate professor. His research interests include model-based systems engineering, model-based safety assessment, and integrated modular avionics system. E-mail: dlcauc@126.com
First author contact:
Co-first author
Supported by:
This work was supported by the China Ministry of Industry and Information Technology Foundation and Aeronautical Science Foundation of China (ASFC-201920007002), the National Key Research and Development Plan (2021YFB1600603), and the Open Fund of Key Laboratory of Civil Aircraft Airworthiness Technology, Civil Aviation University of China.

Abstract

Abstract:

Considering the problem that the scattering echo images of airborne Doppler weather radar are often reduced by ground clutters, the accuracy and confidence of meteorology target detection are reduced. In this paper, a deep convolutional neural network (DCNN) is proposed for meteorology target detection and ground clutter suppression with a large collection of airborne weather radar images as network input. For each weather radar image, the corresponding digital elevation model (DEM) image is extracted on basis of the radar antenna scanning parameters and plane position, and is further fed to the network as a supplement for ground clutter suppression. The features of actual meteorology targets are learned in each bottleneck module of the proposed network and convolved into deeper iterations in the forward propagation process. Then the network parameters are updated by the back propagation iteration of the training error. Experimental results on the real measured images show that our proposed DCNN outperforms the counterparts in terms of six evaluation factors. Meanwhile, the network outputs are in good agreement with the expected meteorology detection results (labels). It is demonstrated that the proposed network would have a promising meteorology observation application with minimal effort on network variables or parameter changes.

Key words: meteorology target detection, ground clutter suppression, weather radar images, convolutional neural network (CNN)

Chaopeng YU, Wei XIONG, Xiaoqing LI, Lei DONG. Deep convolutional neural network for meteorology target detection in airborne weather radar images[J]. Journal of Systems Engineering and Electronics, 2023, 34(5): 1147-1157.

Figures/Tables 16

Fig 1

Fig 2

Fig 3

Fig 5

Fig 4

Fig 6

Fig 7

Fig 8

Table 1

Table 2

Table 3

Fig 9

Fig 10

Fig 11

Fig 12

Fig 13

References 36

1	WANG H J, SHAO N, RAN Y B Identification of precipitation-clouds based on the dual-polarization Doppler weather radar echoes using deep-learning method. IEEE Access, 2019, 7, 12822- 12831. doi: 10.1109/ACCESS.2018.2867546
2	HAN L, SUN J Z, ZHANG W Convolutional neural network for convective storm nowcasting using 3-D Doppler weather radar data. IEEE Trans. on Geoscience and Remote Sensing, 2020, 58 (2): 1487- 1495. doi: 10.1109/TGRS.2019.2948070
3	GUANG Y, ZONG X L, XU F G. Prediction of weather radar images via a deep LSTM for nowcasting. Proc. of the International Joint Conference on Neural Networks, 2020. DOI: 10.1109/IJCNN48605.2020.9206889.
4	GURUNG I, PENG C, MASKEY M Deep feature extraction and its application for hailstorm detection in a large collection of radar images. Signal, Image and Video Processing, 2019, 13, 541- 549. doi: 10.1007/s11760-018-1380-z
5	YANG H Y, WANG F Y Meteorological radar noise image semantic segmentation method based on deep convolutional neural network. Journal of Electronics and Information Technology, 2019, 41 (10): 2373- 2381.
6	SZEGEDY C, VANHOUCKE V, LOFFE S. Rethinking the inception architecture for computer vision. Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 2818−2826.
7	PRADHAN R, AYGUN R, MASKEY M Tropical cyclone intensity estimation using a deep convolutional neural network. IEEE Trans. on Image Processing, 2018, 27 (2): 692- 702. doi: 10.1109/TIP.2017.2766358
8	TRAN Q K, SONG S K Computer vision in precipitation nowcasting: applying image quality assessment metrics for training deep neural networks. Atmosphere, 2019, 10, 244. doi: 10.3390/atmos10050244
9	KRIZHEVSKY A, SUTSKEVER I, HINTON G E ImageNet classification with deep convolutional neural networks. Communications of ACM, 2017, 60, 84- 90. doi: 10.1145/3065386
10	HE K M, ZHANG X Y, REN S Q. Deep residual learning for image recognition. Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770−778.
11	XIONG W, LUO J S, YU C P Power line detection in millimeter-wave radar images applying convolutional neural networks. IET Radar, Sonar and Navigation, 2021, 15 (9): 1083- 1095. doi: 10.1049/rsn2.12090
12	NIU S R, QIU X L, LEI B Parameter extraction based on deep neural network for SAR target simulation. IEEE Trans. on Geoscience and Remote Sensing, 2020, 58 (7): 4901- 4914. doi: 10.1109/TGRS.2020.2968493
13	DENG Z P, SUN H, ZHOU S L Learning deep ship detector in SAR images from scratch. IEEE Trans. on Geoscience and Remote Sensing, 2019, 57 (6): 4021- 4039. doi: 10.1109/TGRS.2018.2889353
14	ZHOU L, WEI S Y, CUI Z M Multi objective detection of complex background radar image based on deep learning. Systems Engineering and Electronics, 2019, 41 (6): 1258- 1264.
15	SHI S N, DONG Z Y, YANG J Sea-surface small target detection based on autonomic learning of time-frequency graph. Systems Engineering and Electronics, 2021, 43 (1): 33- 41.
16	TAO Y, GAO X, HSU K A two-stage deep neural network framework for precipitation estimation from Bispectral satellite information. Hydrometeorology, 2018, 19 (2): 393- 408.
17	NI X, LIU C T, CECIL D J On the detection of hail using satellite passive microwave radiometers and precipitation radar. Applied Meteorology and Climate, 2017, 56 (10): 2693- 2709. doi: 10.1175/JAMC-D-17-0065.1
18	WANG H J, RAN Y B, DENG Y Y Study on deep-learning-based identification of hydrometeors observed by dual polarization Doppler weather radars. EURASIP Journal on Wireless Communication Networking, 2017, 2017, 173.
19	TAO Y M, GAO X G, IHLER A Precipitation identification with bispectral satellite information using deep learning approaches. Hydrometeorology, 2017, 18 (5): 1271- 1283. doi: 10.1175/JHM-D-16-0176.1
20	HAM Y G, KIM J H, LUO J J Deep learning for multi-year ENSO forecasts. Nature, 2019, 573, 568- 572. doi: 10.1038/s41586-019-1559-7
21	RAVURI S, LENC K, WILLSON M Skilful precipitation nowcasting using deep generative models of radar. Nature, 2021, 597, 672- 677. doi: 10.1038/s41586-021-03854-z
22	SHELHAMER E, LONG J, DARRELL T Fully convolutional networks for semantic segmentation. IEEE Trans. on Pattern Analysis and Machine Intelligence, 2017, 39 (4): 640- 651. doi: 10.1109/TPAMI.2016.2572683
23	LI F X, LONG Z R, HE P Fully convolutional pyramidal networks for semantic segmentation. IEEE Access, 2020, 8, 229132- 229140. doi: 10.1109/ACCESS.2020.3045280
24	CHEN L C, PAPANDREOU G, KOKKINOS I DeepLab: semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs. IEEE Trans. on Pattern Analysis and Machine Intelligence, 2018, 40 (4): 834- 848. doi: 10.1109/TPAMI.2017.2699184
25	DONG C, JU Y W Ship object detection SAR images based on semantic segmentation. Systems Engineering and Electronics, 2022, 44 (4): 1195- 1201.
26	LIU Z Y, XIE C S, LI J J Smoke region segmentation recognition algorithm based on improved Deeplabv3+. Systems Engineering and Electronics, 2021, 43 (2): 328- 335.
27	LECUN Y, BENGIO Y, HINTON G Deep learning. Nature, 2015, 521, 436- 444. doi: 10.1038/nature14539
28	SCHMIDHUBER J Deep learning in neural networks: an overview. Neural Networks, 2014, 61, 85- 117.
29	SKOLNIK M I. Radar handbook. 3rd ed. New York: McGraw-Hill Press, 2008.
30	KUREKIN A, RADFORD D, LEVER K New method for generating site-specific clutter map for land-based radar by using multimodal remote-sensing images and digital terrain data. IET Radar, Sonar and Navigation, 2011, 5, 374- 388. doi: 10.1049/iet-rsn.2010.0036
31	KUREKIN A, SHARK L K, LEVER K. Site-specific land clutter modelling based on radar remote sensing images and digital terrain data. Proc. of the IEEE Conference on Image and Signal Processing for Remote Sensing XVI, 2010: 25−36.
32	PASZKE A, CHAURASIA A, KIM S. ENet: a deep neural network architecture for real-time semantic segmentation. https://arxiv.org/pdf/1606.02147.pdf.
33	BADRINARAYANAN V, KENDALL A, CIPOLLA R Segnet: a deep convolutional encoder-decoder architecture for image segmentation. IEEE Trans. on Pattern Analysis and Machine Intelligence, 2017, 39 (12): 2481- 2495. doi: 10.1109/TPAMI.2016.2644615
34	IOFFE S, SZEGEDY C. Batch normalization: accelerating deep network training by reducing internal covariate shift. Proc. of the International Conference on Machine Learning, 2015: 448−456.
35	HE K M, ZHANG X Y, REN S Q. Delving deep into rectifiers: surpassing human-level performance on imagenet classification. Proc. of the IEEE International Conference on Computer Vision, 2015: 1026−1034.
36	ACHILLE A, SOATTO S Information dropout: learning optimal representations through noisy computation. IEEE Trans. on Pattern Analysis and Machine Intelligence, 2018, 40 (12): 2897- 2905. doi: 10.1109/TPAMI.2017.2784440

Parameter	Value
Learning rate $\mu $	0.0005
Decline factor of learning rate	0.1
Epoch	300
k of PReLU in (4)	0.25
Batch size	10
Optimizer parameter ( ${\beta _1}$ , ${\beta _2}$ )	(0.9, 0.999)

Network	Precision	POD	FAR	CSI	Corr	IOU
FCN [18]	0.73	0.59	0.27	0.48	0.66	0.30
SegNet [27]	0.84	0.80	0.16	0.69	0.82	0.30
Our proposed network	0.89	0.96	0.11	0.86	0.93	0.43

Network	Precision	POD	FAR	CSI	Corr	IOU
FCN [18]	0.69	0.56	0.31	0.45	0.62	0.29
SegNet [27]	0.80	0.72	0.20	0.61	0.76	0.30
Our proposed network	0.88	0.91	0.12	0.81	0.90	0.41

[1]	Qihai YAO, Yong WANG, Yixin YANG. Range estimation of few-shot underwater sound source in shallow water based on transfer learning and residual CNN [J]. Journal of Systems Engineering and Electronics, 2023, 34(4): 839-850.
[2]	Hao DU, Wei WANG, Xuerao WANG, Yuanda WANG. Autonomous landing scene recognition based on transfer learning for drones [J]. Journal of Systems Engineering and Electronics, 2023, 34(1): 28-35.
[3]	Zhengliang ZHU, Degui YANG, Junchao ZHANG, Feng TONG. Dataset of human motion status using IR-UWB through-wall radar [J]. Journal of Systems Engineering and Electronics, 2021, 32(5): 1083-1096.
[4]	Wantian WANG, Ziyue TANG, Yichang CHEN, Yongjian SUN. Parity recognition of blade number and manoeuvre intention classification algorithm of rotor target based on micro-Doppler features using CNN [J]. Journal of Systems Engineering and Electronics, 2020, 31(5): 884-889.
[5]	Binquan LI, Xiaohui HU. Effective distributed convolutional neural network architecture for remote sensing images target classification with a pre-training approach [J]. Journal of Systems Engineering and Electronics, 2019, 30(2): 238-244.
[6]	Baojun ZHAO, Boya ZHAO, Linbo TANG, Wenzheng WANG, Chen WU. Multi-scale object detection by top-down and bottom-up feature pyramid network [J]. Journal of Systems Engineering and Electronics, 2019, 30(1): 1-12.
[7]	Liangkui LIN, Shaoyou WANG, Zhongxing TANG. Using deep learning to detect small targets in infrared oversampling images [J]. Journal of Systems Engineering and Electronics, 2018, 29(5): 947-952.
[8]	Bendong Zhao, Huanzhang Lu, Shangfeng Chen, Junliang Liu, and Dongya Wu. Convolutional neural networks for time series classification [J]. Systems Engineering and Electronics, 2017, 28(1): 162-.

Deep convolutional neural network for meteorology target detection in airborne weather radar images

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

Share this article

Figures/Tables 16

References 36

Related Articles 8

Recommended Articles

Metrics

Comments