Novel method for extraction of ship target with overlaps in SAR image via EM algorithm

doi:10.23919/JSEE.2023.000170

Abstract

Abstract:

The quality of synthetic aperture radar (SAR) image degrades in the case of multiple imaging projection planes (IPPs) and multiple overlapping ship targets, and then the performance of target classification and recognition can be influenced. For addressing this issue, a method for extracting ship targets with overlaps via the expectation maximization (EM) algorithm is proposed. First, the scatterers of ship targets are obtained via the target detection technique. Then, the EM algorithm is applied to extract the scatterers of a single ship target with a single IPP. Afterwards, a novel image amplitude estimation approach is proposed, with which the radar image of a single target with a single IPP can be generated. The proposed method can accomplish IPP selection and targets separation in the image domain, which can improve the image quality and reserve the target information most possibly. Results of simulated and real measured data demonstrate the effectiveness of the proposed method.

Key words: expectation maximization (EM) algorithm, image processing, imaging projection plane (IPP), overlapping ship target, synthetic aperture radar (SAR)

Rui CAO, Yong WANG. Novel method for extraction of ship target with overlaps in SAR image via EM algorithm[J]. Journal of Systems Engineering and Electronics, 2024, 35(4): 874-887.

Figures/Tables 39

Fig 1

Fig 2

Fig 3

Fig 4

Fig 5

Table 1

Fig 6

Fig 7

Fig 8

Fig 9

Fig 10

Table 2

Fig 11

Fig 12

Fig 13

Fig 14

Fig 15

Fig 16

Fig 17

Table 3

Fig 18

Fig 19

Fig 20

Table 4

Fig 21

Fig 22

Fig 23

Fig 24

Table 5

Fig 25

Fig 26

Fig 27

Fig 28

Fig 29

Fig 30

Table 6

Fig 31

Fig 32

Fig 33

References 29

1	LI N, XING M D, HOU Y X, et al Ship focusing and positioning based on 2D ambiguity resolving for single-channel SAR mounted on high-speed maneuvering platforms with small aperture. IEEE Trans. on Geoscience and Remote Sensing, 2022, 60, 5221213.
2	WANG X Q, LI G, PLAZA A, et al Ship detection in SAR images by aggregating densities of Fisher vectors: extension to a global perspective. IEEE Trans. on Geoscience and Remote Sensing, 2022, 60, 2506613.
3	ZHANG T Y, LIU S J, DING Z G, et al A motion state judgment and radar imaging algorithm selection method for ship. IEEE Trans. on Geoscience and Remote Sensing, 2022, 60, 5236318.
4	ZHANG J Q, WANG Y, LU X F Distributed inverse synthetic aperture radar imaging of ship target with complex motion. Journal of Systems Engineering and Electronics, 2021, 32 (6): 1325- 1337.
5	CHEN S Y, WANG Y, CAO R A high frequency vibration compensation approach for ultra-high resolution SAR imaging based on sinusoidal frequency modulation Fourier-Bessel transform. Journal of Systems Engineering and Electronics, 2023, 34 (4): 894- 905.
6	HUANG P H, XIA X G, ZHAN M Y, et al ISAR imaging of a maneuvering target based on parameter estimation of multicomponent cubic phase signals. IEEE Trans. on Geoscience and Remote Sensing, 2022, 60, 5103918.
7	MARTORELLA M, PASTINA D, BERIZZI F, et al Spaceborne radar imaging of maritime moving targets with the Cosmo-SkyMed SAR system. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7 (7): 2797- 2810.
8	SHAO S, LIU H W, ZHANG L, et al Integration of super-resolution ISAR imaging and fine motion compensation for complex maneuvering ship targets under high sea state. IEEE Trans. on Geoscience and Remote Sensing, 2022, 60, 5222820.
9	MARTORELLA M Optimal sensor positioning for inverse synthetic aperture radar. IEEE Trans. on Aerospace and Electronic Systems, 2013, 49 (1): 648- 658. doi: 10.1109/TAES.2013.6404129
10	CHEN Y M, ZHOU P, DAI Y S. Application of an existing approach to refocusing maritime moving targets on Radarsat-2 SLC images. Proc. of the IEEE 5th Asia-Pacific Conference on Synthetic Aperture Radar, 2015: 502−506.
11	SHAO S, ZHANG L, LIU H W An optimal imaging time interval selection technique for marine targets ISAR imaging based on sea dynamic prior information. IEEE Sensors Journal, 2019, 19 (13): 4940- 4953. doi: 10.1109/JSEN.2019.2903399
12	PASTINA D, MONTANARI A, APRILE A. Motion estimation and optimum time selection for ship ISAR imaging. Proc. of the IEEE Radar Conference, 2003: 7−14.
13	PASTINA D, SPINA C, APRILE A. A slope-based technique for motion estimation and optimum time selection for ISAR imaging of ship targets. Proc. of the 14th European Signal Processing Conference, 2006: 1−5.
14	LI N, SHEN Q Y, WANG L, et al Optimal time selection for ISAR imaging of ship targets based on time-frequency analysis of multiple scatterers. IEEE Geoscience and Remote Sensing Letters, 2022, 19, 4017505.
15	CAO R, WANG Y, YEH C M, et al A novel optimal time window determination approach for ISAR imaging of ship targets. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2022, 15, 3475- 3503. doi: 10.1109/JSTARS.2021.3139714
16	ZHOU P, ZHANG X, SUN W F, et al Time-frequency analysis-based time-windowing algorithm for the inverse synthetic aperture radar imaging of ships. Journal of Applied Remote Sensing, 2018, 12 (1): 1- 20.
17	LI N, WANG L, ZHU D Y. Optimal ISAR imaging time selection of ship targets using real data. Proc. of the IET International Radar Conference, 2013. DOI: 10.1049/cp.20130166.
18	ZHOU P, ZHANG X, DAI Y S, et al Time window selection algorithm for ISAR ship imaging based on instantaneous Doppler frequency estimations of multiple scatterers. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2019, 12 (10): 3799- 3812. doi: 10.1109/JSTARS.2019.2943349
19	GAO F, HUO Y Y, WANG J, et al Anchor-free SAR ship instance segmentation with centroid-distance based loss. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14, 11352- 11371. doi: 10.1109/JSTARS.2021.3123784
20	PASTINA F, FICO F, LOMBARDO P. Detection of ship targets in COSMO-SkyMed SAR images. Proc. of the IEEE Radar Conference, 2011: 928−933.
21	OTSU N A threshold selection method from gray-level histograms. IEEE Trans. on Systems, Man, and Cybernetics, 1979, 9 (1): 62- 66. doi: 10.1109/TSMC.1979.4310076
22	ZHANG T W, ZHANG X L, SHI J, et al Balance scene learning mechanism for offshore and inshore ship detection in SAR images. IEEE Geoscience and Remote Sensing Letters, 2022, 19, 4004905.
23	LANG H T, XI Y Y, ZHANG X Ship detection in high-resolution SAR images by clustering spatially enhanced pixel descriptor. IEEE Trans. on Geoscience and Remote Sensing, 2019, 57 (8): 5407- 5423. doi: 10.1109/TGRS.2019.2899337
24	CAO R, WANG Y, ZHAO B, et al Ship target imaging in airborne SAR system based on automatic image segmentation and ISAR technique. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14, 1985- 2000. doi: 10.1109/JSTARS.2021.3050108
25	KARAMI J, ALIMOHAMMADI A, MODABBERI S Analysis of the spatio-temporal patterns of water pollution and source contribution using the MODIS sensor products and multivariate statistical techniques. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2012, 5 (4): 1243- 1255. doi: 10.1109/JSTARS.2012.2187273
26	KIM G, YANG S, SIM J Y Saliency-based initialisation of Gaussian mixture models for fully-automatic object segmentation. Electronics Letters, 2017, 53 (25): 1648- 1649. doi: 10.1049/el.2017.3877
27	MCLACHLAN G J. The EM algorithm and extensions. Hoboken: Wiley, 2008.
28	RANEY R K, RUNGE H, BAMLER R, et al Precision SAR processing using chirp scaling. IEEE Trans. on Geoscience and Remote Sensing, 1994, 32 (4): 786- 799. doi: 10.1109/36.298008
29	ISA D, LEE L H, KALLIMANI V P, et al Text document preprocessing with the Bayes formula for classification using the support vector machine. IEEE Trans. on Knowledge and Data Engineering, 2008, 20 (9): 1264- 1272. doi: 10.1109/TKDE.2008.76

Parameter	Value
Sample frequency/MHz	200
Bandwidth/MHz	60
Pulse repetition frequency/MHz	200
Pulse width/μs	10
Carrier frequency/GHz	5.3
Aircraft height/km	10
Aircraft velocity/(m/s)	50

Cluster	$ {{\boldsymbol{\mu}} }_k^{{\mathrm{opt}}} $	$ {\boldsymbol{\varSigma }}_k^{{\mathrm{opt}}} $	Eigenvalue		Axis length of ellipse		Eigenvector
Cluster	$ {{\boldsymbol{\mu}} }_k^{{\mathrm{opt}}} $	$ {\boldsymbol{\varSigma }}_k^{{\mathrm{opt}}} $	${\lambda _{k,1}}$	${\lambda _{k,2}}$	${a_k}$	${b_k}$	${{v}_{k,1}}$	${{v}_{k,2}}$
1	${\left[ {\begin{array}{*{20}{c}} {1\;009.5}&{933.9} \end{array}} \right]^{\text{T}}}$	$\left[ {\begin{array}{*{20}{c}} {1\;341.5}&{ - 2\;971.3} \\ { - 2\;971.3}&{9\;830.2} \end{array}} \right]$	405.0	10767.0	49.25	253.98	$ {\left[ {\begin{array}{*{20}{c}} { - 0.95}&{ - 0.30} \end{array}} \right]^{\text{T}}} $	$ {\left[ {\begin{array}{*{20}{c}} { - 0.30}&{0.95} \end{array}} \right]^{\text{T}}} $
2	$ {\left[ {\begin{array}{*{20}{c}} {1\;002.0}&{1\;021.0} \end{array}} \right]^{\text{T}}} $	$\left[ {\begin{array}{*{20}{c}} {1\;300.0}&{3\;495.0} \\ {3\;495.0}&{2\;162.1} \end{array}} \right]$	716.0	22205.0	65.50	364.73	$ {\left[ {\begin{array}{*{20}{c}} { - 0.99}&{0.16} \end{array}} \right]^{\text{T}}} $	$ {\left[ {\begin{array}{*{20}{c}} {0.16}&{0.99} \end{array}} \right]^{\text{T}}} $

Cluster	$ {{\boldsymbol{\mu}} }_k^{{\mathrm{opt}}} $	$ {\boldsymbol{\varSigma }}_k^{{\mathrm{opt}}} $	Eigenvalue (×10³)		Axis length of ellipse		Eigenvector
Cluster	$ {{\boldsymbol{\mu}} }_k^{{\mathrm{opt}}} $	$ {\boldsymbol{\varSigma }}_k^{{\mathrm{opt}}} $	${\lambda _{k,1}}$	${\lambda _{k,2}}$	${a_k}$	${b_k}$	${{v}_{k,1}}$	${{v}_{k,2}}$
1	${\left[ {\begin{array}{*{20}{c}} {805.38}&{940.40} \end{array}} \right]^{\text{T}}}$	$\left[ {\begin{array}{*{20}{c}} {1\;094.4}&{ - 369.1} \\ { - 369.1}&{2\;834.9} \end{array}} \right]$	1.02	2.91	78.15	132.04	$ {\left[ {\begin{array}{*{20}{c}} { - 0.98}&{ - 0.20} \end{array}} \right]^{\text{T}}} $	$ {\left[ {\begin{array}{*{20}{c}} {0.20}&{ - 0.98} \end{array}} \right]^{\text{T}}} $
2	${\left[ {\begin{array}{*{20}{c}} {855.23}&{732.14} \end{array}} \right]^{\text{T}}}$	$\left[ {\begin{array}{*{20}{c}} {1\;175.9}&{ - 1\;448.4} \\ { - 1\;448.4}&{6\;463.2} \end{array}} \right]$	0.81	6.83	69.45	202.34	$ {\left[ {\begin{array}{*{20}{c}} { - 0.97}&{ - 0.25} \end{array}} \right]^{\text{T}}} $	$ {\left[ { \begin{array}{*{20}{c}} {-0.25}&{0.97} \end{array}} \right]^{\text{T}}} $

Parameter	Value
Sample frequency/MHz	100
Bandwidth/MHz	60
Pulse repetition frequency/Hz	500
Pulse width/μs	10
Carrier frequency	C band
Range center/km	4.175
Aircraft velocity/(m/s)	50

Cluster	$ {{\boldsymbol{\mu}} }_k^{{\mathrm{opt}}} $	$ {\boldsymbol{\varSigma }}_k^{{\mathrm{opt}}} $	Eigenvalue (×10³)		Axis length of ellipse		Eigenvector
Cluster	$ {{\boldsymbol{\mu}} }_k^{{\mathrm{opt}}} $	$ {\boldsymbol{\varSigma }}_k^{{\mathrm{opt}}} $	${\lambda _{k,1}}$	${\lambda _{k,2}}$	${a_k}$	${b_k}$	${{v}_{k,1}}$	${{v}_{k,2}}$
1	${\left[ {\begin{array}{*{20}{c}} {288.50}&{139.44} \end{array}} \right]^{\text{T}}}$	$\left[ {\begin{array}{*{20}{c}} {441.84}&{95.72} \\ {95.92}&{38.18} \end{array}} \right]$	16.63	463.39	9.98	52.69	$ {\left[ {\begin{array}{*{20}{c}} {0.22}&{ - 0.98} \end{array}} \right]^{\text{T}}} $	$ {\left[ {\begin{array}{*{20}{c}} { - 0.98}&{ - 0.22} \end{array}} \right]^{\text{T}}} $
2	${\left[ {\begin{array}{*{20}{c}} {303.19}&{134.67} \end{array}} \right]^{\text{T}}}$	$\left[ {\begin{array}{*{20}{c}} {300.82}&{ - 83.41} \\ { - 83.41}&{30.22} \end{array}} \right]$	6.58	324.47	6.28	44.09	$ {\left[ {\begin{array}{*{20}{c}} { - 0.27}&{ - 0.96} \end{array}} \right]^{\text{T}}} $	$ {\left[ {\begin{array}{*{20}{c}} { - 0.96}&{0.27} \end{array}} \right]^{\text{T}}} $