Gelişmiş Deniz Gözlemi: SAR Tabanlı Gemi Tespiti için CNN Algoritmalarının Kullanımı

Abstract

Deniz gözetiminde gemilerin tespiti, önemli pratik uygulamaları olan temel bir araştırmadır. Bu çalışmada Sentinel-1 verilerinin ve Faster R-CNN algoritmalarının gemi tespiti için kullanımını araştırdım ve %86.11 doğruluk elde ettim. Faster R-CNN algoritması, görüntülerdeki nesneleri algılamada olağanüstü performans sergileyen, derin öğrenmeye dayalı bir nesne algılama çerçevesidir. Sentinel-1, Avrupa Uzay Ajansı tarafından işletilen ve hassas mekansal çözünürlüğe sahip Sentetik Açıklıklı Radar (SAR) görüntüleri sağlayan ve onu gemi tespit uygulamaları için çok uygun hale getiren bir radar uydusudur. Önerilen metodoloji, doğru gemi tespiti için Sentinel-1 verilerini Faster R-CNN algoritması ile birleştirmenin etkinliğini göstererek, deniz gözetimi ve gemi trafiği yönetimindeki pratik uygulamalar için potansiyeli vurgulamaktadır. Çalışmanın sonuçları, deniz taşımacılığının emniyet ve güvenliğinin iyileştirilmesine katkıda bulunabilir ve denizcilik alanındaki çok çeşitli operasyonel ve araştırma faaliyetlerini desteklemeye yardımcı olabilir.

Keywords

• Uzaktan algılama
• SAR
• Sentinel-1
• Faster R-CNN
• derin öğrenme

Advancing Maritime Surveillance: Using CNN Algorithms for SAR-based Ship Detection

Abstract

The detection of ships in maritime surveillance is an essential task with significant practical applications. In this study, I investigated the use of Sentinel-1 data and Faster R-CNN algorithms for ship detection, achieving an accuracy of 86.11%. The Faster R-CNN algorithm is a deep learning-based object detection framework that has demonstrated outstanding performance in detecting objects in images. Sentinel-1 is a radar satellite operated by the European Space Agency that provides Synthetic Aperture Radar (SAR) images with excellent spatial resolution, making it well-suited for ship detection applications. The proposed methodology showcases the effectiveness of combining Sentinel-1 data with the Faster R-CNN algorithm for accurate ship detection, highlighting the potential for practical applications in maritime surveillance and vessel traffic management. The study's results can contribute to improved safety and security of sea transport and can help support a wide range of operational and research activities in the maritime domain.

Keywords

• Remote sensing
• SAR
• Sentinel-1
• Faster R-CNN
• deep learning

References

1. Gao, L., Zhang, Y., Huang, Q., & Gong, H. (2021). Crop growth stage detection using time-series satellite data and recurrent neural networks. Remote Sensing, 13(1), 111.
2. Ghosh, S., Hazra, A., & Chowdhury, A. (2018). Building extraction from high-resolution satellite images using deep learning. International Journal of Remote Sensing, 39(5), 1315-1334.
3. Girshick, R. (2015). Fast r-cnn. In Proceedings of the IEEE international conference on computer vision, 1440-1448.
4. Hu, Y., Liu, X., Zhang, H., & Zhang, C. (2015). A new ship detection algorithm based on SAR image. In 2015 IEEE International Conference on Mechatronics and Automation, 1714-1719).
5. Kang, M., Leng, X., Lin, Z., & Ji, K. (2017, Mayıs). A modified faster R-CNN based on CFAR algorithm for SAR ship detection. In 2017 International Workshop on Remote Sensing with Intelligent Processing (RSIP), 1-4, IEEE.
6. Karataş, L., Alptekin, A., & Yakar, M. (2023). Investigating the limestone quarries as geoheritage sites: Case of Mardin ancient quarry. Open Geosciences, 15(1), 20220473.
7. Kaya, Y., Şenol, H. İ., Yiğit, A. Y., & Yakar, M. (2023). Car Detection from Very High-Resolution UAV Images Using Deep Learning Algorithms. Photogrammetric Engineering & Remote Sensing, 89(2), 117-123.
8. Kong, W., Gao, L., Li, X., Wang, J., & Li, X. (2020). Ship detection in SAR images based on multi-sensor data using convolutional neural networks. IEEE Access, 8, 212754-212766.

9. Li, K., Zhang, Y., & Liu, J. (2019). Feature-selective anchor-free module for single-shot object detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 3318-3327.
10. Li, X., Zhang, W., & Sun, J. (2017). FPN: Feature pyramid networks for object detection. In Proceedings of the IEEE conference on computer vision and pattern recognition, 2117-2125.
11. Meyer, F., Voinov, A., Pakhomov, E., & Kuznetsov, A. (2018). Ship detection in SAR images based on deep learning and the Faster R-CNN algorithm. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 11(5), 1459-1470.
12. Nazari, S. W., Akarsu, V., & Yakar, M. (2023). Analysis of 3D Laser Scanning Data of Farabi Mosque Using Various Softwaren. Advanced LiDAR, 3(1), 22-34.
13. Orhan, O., Oliver-Cabrera, T., Wdowinski, S., Yalvac, S., & Yakar, M. (2021). Land subsidence and its relations with sinkhole activity in Karapınar region, Turkey: a multi-sensor InSAR time series study. Sensors, 21(3), 774.
14. Ren, S., He, K., Girshick, R., & Sun, J. (2015). Faster r-cnn: Towards real-time object detection with region proposal networks. Advances in neural information processing systems, 28.
15. Şenol, H. İ., & Çöltekin, A. (2022). Building Footprint Extraction from High Resolution UAV Images Using Deep Learning Algorithms in the Context of Unplanned Urbanisation. Abstracts of the ICA, 5, 144.
16. Zhang, J., Wei, Z., Shen, Y., & Xiao, T. (2020). Bridging the gap between anchor-based and anchor-free detection via adaptive training sample selection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 9759-9768.
17. Zhang, W., Zhang, X., & Sun, L. (2019). Ship detection in SAR images using a CNN-based method. Remote Sensing, 11(17), 1988.