Anterior|Siguiente

On orbit extraction method of ship target in SAR images based on ultra-lightweight network

  • role： First author第一作者
  • Affiliation：

    School of Information and Electronics, Beijing Institute of Technology, Beijing 100081, China

    Institute of Spacecraft System Engineering, China Academy of Space Technology, Beijing 100094, China

  • Email：leezl0519@163.com
  • Introduction：李宗凌,1985年生, 男, 高级工程师, 研究方向为遥感图像目标检测识别。E-mail: leezl0519@163.com
LI Zongling12,  
  • Affiliation：

    Institute of Spacecraft System Engineering, China Academy of Space Technology, Beijing 100094, China

WANG Luyuan2,  
  • Affiliation：

    Institute of Spacecraft System Engineering, China Academy of Space Technology, Beijing 100094, China

JIANG Shuai2,  
  • Affiliation：

    Institute of Spacecraft System Engineering, China Academy of Space Technology, Beijing 100094, China

WU Yuhang2,  
  • role： Corresponding author通信作者
  • Affiliation：

    Institute of Spacecraft System Engineering, China Academy of Space Technology, Beijing 100094, China

  • Email：ztzhangqj@163.com
  • Introduction：张庆君，1969年生，男,研究员，研究方向为航天器系统与总体技术、航天遥感。E-mail: ztzhangqj@163.com
ZHANG Qingjun2*

resumen

The low efficiency of orbit data processing and poor effectiveness of remote sensing satellite information acquisition are problems requiring great attention. Meeting the requirements of satellites in orbit in real time and efficient extraction of the targets of interest are challenging endeavors because of their heavy dependence on data transmission and communication bandwidth.In this paper, an efficient ship target detection algorithm based on global statistics and an ultra-lightweight suspected target identification network is designed to achieve the rapid extraction of ship targets. The traditional image processing method is used to detect the target quickly, obtain the suspected target slice, and reduce the amount of data significantly. A self-designed and improved ultra-lightweight identification network based on deep learning is then used to realize a second cycle of suspected target screening and improve the accuracy of target extraction. Much work has been done to simplify the real-time implementation of image-processing methods, such as OTSU threshold calculation, connected domain labeling, and target identification network based on deep learning. The real-time processing accuracy, speed, hardware scale, and heat consumption of the proposed method are well balanced by reasonable optimization of the algorithm flow and calculation method and establishment of an appropriate error analysis model. The calculation cost of the algorithm is low, and the dependence of the real-time implementation of the algorithm on the performance of the hardware platform is reduced greatly, thereby ensuring the excellent overall performance of the algorithm. In particular, network pruning, weight parameter sharing, and quantization are used to reduce the network weight parameter and forward reasoning calculation storage requirements, thus improving the design of the target identification network.GF-3 satellite data were used to test the algorithm, and experimental results showed that the accuracy of ship target extraction could be improved by 20%—98%, the computational complexity could be reduced by 90%, and the real-time performance could be improved by 50%. When the ship target extraction algorithm model was implemented in a low-power embedded circuit, the power consumption of the whole calculation circuit was less than 13 W, and the standby power consumption was less than 1.5 W; these values meet the requirements of full-time operations in orbit.The proposed method takes into account the effectiveness of the algorithm and feasibility of in-orbit real-time processing to improve the energy efficiency ratio and target extraction performance of the system effectively. The proposed algorithm was implemented in the current satellite embedded circuit and the orbit of a new radar test satellite, and the results obtained reflect good application prospects.

palabra clave

ultra-lightweight network;SAR image;target detection;target identify;real time processing on orbit

References

  1. 1.
    Bi F K, Gao L N, Long T and Yang J. 2011. Remote sensing target detection based on visual saliency guidance and classifier fusion. Infrared and Laser Engineering, 40(10): 2058-2064
  2. 2.
    Chen W T, Xing X W and Ji K F. 2012. A survey of ship target recognition in SAR images. Modern Radar, 34(11): 53-58
  3. 3.
    Chen Z, Shi Y P and Ji S P. 2012. Improved image threshold segmentation algorithm based on OTSU method. Laser and Infrared, 42(5): 584-588
  4. 4.
    Cumming I G and Wong F H. 2012. Digital Processing of Synthetic Aperture Radar Data: Algorithms and Implementation. Hong W and Hu D H, trans. Beijing: Publishing House of Electronics Industry: 1-13
  5. 5.
    Cui B, Zhang Y H, Yan L and Wei J J.2020. Dual-thresholds change detection in GF-3 SAR images. Journal of Remote Sensing, 24(1):1-10
  6. 6.
    Fang M X, Bi D P, Shen A G and Liu Q. 2017. Jamming technique of multiple false targets against CFAR detection in SAR images. Journal of Electronics and Information Technology, 39(4): 973-980
  7. 7.
    He Z Y, Sun L N and Chen L G. 2013. Fast computation of threshold based on Otsu criterion. Acta Electronica Sinica, 41(2): 267-272
  8. 8.
    LeCun Y, Bottou L, Bengio Y and Haffner P. 1998. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11): 2278-2324
  9. 9.
    Li J W, Qu C W, Peng S J and Deng B. 2018. Ship detection in SAR images based on convolutional neural network. Systems Engineering and Electronics, 40(9): 1953-1959
  10. 10.
    Li J W, Qu C W, Peng S J and Jiang Y. 2019. Ship detection in SAR images based on generative adversarial network and online hard examples mining. Journal of Electronics and Information Technology, 41(1): 143-149
  11. 11.
    Li Z L, Wang L Y, Yu J Y, Cheng B W and Hao L. 2020. Remote sensing ship target detection and recognition method. Remote Sensing Information, 35(1): 64-72
  12. 12.
    Li Z L, Wang L Y, Yu J Y, Hao L and Cheng B W. 2018. Space-borne SAR on-orbit imaging and ship target detection method. Spacecraft Engineering, 27(6): 41-47
  13. 13.
    Mark S, Andrew H, Zhu M L, Andrey Z and Chen L C . 2018. MobileNetV2: Inverted Residuals and Linear Bottlenecks, IEEE/CVF Conference on Computer Vision and Pattern Recognition. Proceedings, 4510-4520
  14. 14.
    Novak L M, Owirka G J, Brower W S and Weaver A L. 1997. The automatic target-recognition system in SAIP. The Lincoln Laboratory Journal, 10(2): 187-202
  15. 15.
    Song W Q, Wang Y H and Liu H W. 2016. An automatic block-to-block censoring target detector for high resolution SAR image. Journal of Electronics and Information Technology, 38(5): 1017-1025
  16. 16.
    Wang C, Zhang H, Wu F and Zhang B. 2013. Ship Target Detection and Classification in High Resolution SAR Images. Beijing: Science Press: 14-15
  17. 17.
    Xie Y Z, Tan X B and Chen H. 2012. A new algorithm for connected components labeling. Transactions of Beijing Institute of Technology, 32(12): 1273-1278
  18. 18.
    Xu Y F, Tan Y J, He R J and Li J F. 2010. System analysis and research overview of space-based maritime surveillance. Journal of Astronautics, 31(3): 628-640
  19. 19.
    Zhai L, Li Y and Su Y. 2017. A method for ship detection in SAR images based on anomaly detection and two-layer censoring mechanism. Computer Engineering, 43(4): 244-250
  20. 20.
    Zhou Y, Shen A L and Bi D P. 2020. A fast detection algorithm of vibration targets for Dual-Channel SAR. Journal of Remote Sensing, 24(9): 1143-1156
  21. 21.
    Zhang G H, Xu K, Duan G D, Zhao C and Liang F. 2018. A fast labeling algorithm of connected components applicable for hardware implementation. Journal of Xi'an Jiaotong University, 52(8): 95-101
  22. 22.
    Zhang Q J, Han X L and Liu J. 2017. Technology progress and development trend of spaceborne synthetic aperture radar remote sensing. Spacecraft Engineering, 26(6): 1-8

Leer el texto completo

The above content is generated by Large Model Translation. The translated content is for reference only. We do not assume any commercial or legal responsibilty for any consequences arising from the use of our website