COMPARATIVE ANALYSIS OF TARGET LOCALIZATION ACCURACY IN RADAR PPI IMAGES USING GEOMETRIC ENCLOSURES

Abstract

Maritime surveillance system requires sensor fusion of radar Plan Position Indicator (PPI) imagery with Automatic Identification System (AIS) data. However, systematic azimuth errors (0°–5°) and irregular echo shapes reduce ship localization accuracy significantly. This study compares three geometric enclosure methods (bounding boxes, convex hulls, and concave hulls) using the publicly available DLR Baltic Sea dataset (X-band, ~10 m resolution, 781 synchronized scans of 4 ships). A 4.5° azimuth offset compensation is applied to minimize target position deviation. Target positions are estimated from calculated bounding box centers and hull centroids from radar images. Validation against AIS ground truth data achieved the lowest average localization errors (58–70 m) for concave hull centroids. It demonstrates up to 12.7% improvement over bounding boxes solution and a 2.1% improvement over convex hulls solution. The results show the accuracy of the concave hulls method in target detection and in preserving echo concavities caused by sea clutter, and noise etc. The concave hull method offers a lightweight and robust solution for real-time Vessel Traffic Services (VTS) and autonomous surface vehicles (ASVs).

Keywords

• AIS fusion
• Autonomous surface vehicles
• Computer vision
• Geometric enclosures
• Radar PPI
• Target localization.

Supporting Institution

The author declares that no funding was used in the study.

Ethical Statement

The authors declare that they comply with all ethical standards.

Thanks

The author declares that no funding was used in the study.

GEOMETRİK ÇEVİRMELER KULLANILARAK RADAR PPI GÖRÜNTÜLERİNDE HEDEF KONUMU BELİRLEME DOĞRULUĞUNUN KARŞILAŞTIRMALI ANALİZİ

Abstract

Deniz gözetleme sistemi, radar Plan Pozisyon Göstergesi (PPG) görüntülerinin Otomatik Tanımlama Sistemi (OTS) verileriyle algılayıcı füzyonunu gerektirir. Ancak, sistematik azimut hataları (0°–5°) ve düzensiz yankı şekilleri, gemi konumlandırma doğruluğunu önemli ölçüde azaltır. Bu çalışma, halka açık DLR Baltık Denizi veri seti (X-bant, ~10 m çözünürlük, 4 geminin 781 senkronize taraması) kullanılarak üç geometrik çevreleme yöntemini (sınırlayıcı kutular, dışbükey gövde ve içbükey gövde) karşılaştırmaktadır. Hedef konum sapmasını en aza indirmek için 4,5° azimut ofset telafisi uygulanmıştır. Hedef konumları, radar görüntülerinden hesaplanan sınırlayıcı kutu merkezlerinden ve gövde merkezlerinden tahmin edilmektedir. AIS referans verilerine karşı doğrulama, içbükey gövde merkezleri için en düşük ortalama konumlandırma hatalarını (58–70 m) elde etmiştir. Bu, sınırlayıcı kutular çözümüne göre %12,7'ye varan iyileşme ve dışbükey gövde çözümüne göre %2,1 iyileşme göstermektedir. Sonuçlar, içbükey gövde yönteminin hedef tespitinde ve gürültü vb. nedenlerle oluşan yankı bozulmalarını korumada başarısını göstermektedir. İçbükey gövde yöntemi, gerçek zamanlı Gemi Trafik Hizmetleri (GTH) ve otonom yüzey araçları (OYA) için ağır hesaplama gerektirmeyen ve sağlam bir çözüm sunar.

Keywords

• AIS füzyonu
• Otonom yüzey araçları
• Bilgisayarlı görü
• Geometrik çevirmeler
• Radar PPI
• Hedef konum tespiti.

Supporting Institution

Yazar, çalışmada herhangi bir fon kullanılmadığını beyan etmektedir.

Ethical Statement

Yazarlar tüm etik standartlara uyduklarını beyan etmektedirler.

Thanks

Yazar, çalışmada herhangi bir fon kullanılmadığını beyan etmektedir.

References

1. J. S. Fowdur, M. Baum, and F. Heymann, “Real-world marine radar datasets for evaluating target tracking methods,” Sensors, vol. 21, no. 14, p. 4641, Jul. 2021, doi: 10.3390/s21144641.
2. S. Külcü, “Radar image-based object detection and tracking for autonomous surface vehicles,” Turk. J. Marit. Mar. Sci., vol. 12, no. 1, pp. 35–47, 2026, doi: 10.52998/trjmms.1743670.
3. A. P. Nemirko and J. H. Dulá, “Machine learning algorithm based on convex hull analysis,” Proc. Comput. Sci., vol. 186, pp. 381–386, 2021, doi: 10.1016/j.procs.2021.04.160.
4. X. Xu, B. Wu, L. Xie, Â. P. Teixeira, and X. Yan, “A novel ship speed and heading estimation approach using radar sequential images,” IEEE Trans. Intell. Transp. Syst., vol. 24, no. 10, pp. 11107–11120, Oct. 2023, doi: 10.1109/TITS.2023.3281547.
5. X. Xu, B. Wu, Â. P. Teixeira, X. Yan, and C. Guedes Soares, “Integration of radar sequential images and AIS for ship speed and heading estimation under uncertainty,” IEEE Trans. Intell. Transp. Syst., vol. 25, no. 6, pp. 5688–5702, Jun. 2024, doi: 10.1109/TITS.2023.3338293.
6. X. Xu, B. Wu, Â. P. Teixeira, and C. Guedes Soares, “Ship collision risk assessment and visualization using radar sequential images for intelligent maritime supervision,” IEEE Trans. Intell. Transp. Syst., vol. 26, no. 6, pp. 9004–9017, Jun. 2025, doi: 10.1109/TITS.2025.3539011.
7. Z. Guo, X. Zhang, C. Liu, X. Ji, J. Jiao, and Q. Ye, “Convex-hull feature adaptation for oriented and densely packed object detection,” IEEE Trans. Circuits Syst. Video Technol., vol. 32, no. 8, pp. 5252–5266, Aug. 2022, doi: 10.1109/TCSVT.2022.3140248.
8. H. Liang et al., “Deephullnet: A deep learning approach for solving the convex hull and concave hull problems with transformer,” Int. J. Digit. Earth, vol. 17, no. 1, p. 2358843, 2024, doi: 10.1080/17538947.2024.2358843.

9. J. Guan et al., “Detection of small targets on sea surface based on 3-D concave hull learning algorithm,” J. Electron. Inf. Technol., vol. 45, no. 5, pp. 1602–1610, May. 2023, doi: 10.11999/JEIT220448.
10. Y. Shi and Y. Hu, “mRMR-Tri-ConcaveHull detector for floating small targets in sea clutter,” IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens., vol. 16, pp. 6799–6811, 2023, doi: 10.1109/JSTARS.2023.3297278.
11. S. Shi, J. Wang, J. Wang, and T. Li, “False-alarm-controllable detection of marine small targets via improved concave hull classifier,” Remote Sens., vol. 17, no. 11, p. 1808, May. 2025, doi: 10.3390/rs17111808.
12. X. Gao et al., “GCT-YOLOv5: A lightweight and efficient object detection model of real-time side-scan sonar image,” Signal, Image Video Process., vol. 18, no. Suppl 1, pp. S565–S574, Apr. 2024, doi: 10.1007/s11760-024-03174-5.
13. X. Yang et al., “Ship instance segmentation based on rotated bounding boxes for SAR images,” Remote Sens., vol. 15, no. 5, p. 1324, Feb. 2023, doi: 10.3390/rs15051324.
14. J. Xu, Y. Suo, Y. Jiang, and Q. Yang, “A fast algorithm for matching AIS trajectories with radar point data in complex environments,” Remote Sens., vol. 16, no. 23, p. 4360, Nov. 2024, doi: 10.3390/rs16234360.
15. M. Li, Q. Liu, and S. Peng, “Point cloud registration based on concave hull association,” Remote Sens. Lett., vol. 14, no. 3, pp. 313–322, Mar. 2023, doi: 10.1080/2150704X.2023.2195569.
16. D. Korban et al., “Radar-based detection and recognition methodology of autonomous surface vehicles in challenging marine environment,” Sci. J. Silesian Univ. Technol. Ser. Transp., vol. 122, pp. 111–127, 2024, doi: 10.20858/sjsutst.2024.122.7.
17. M. Feng, T. Zhang, S. Li, G. Jin, and Y. Xia, “An improved minimum bounding rectangle algorithm for regularized building boundary extraction from aerial LiDAR point clouds with partial occlusions,” Int. J. Remote Sens., vol. 41, no. 1, pp. 300–319, Jan. 2020, doi: 10.1080/01431161.2019.1641245.
18. X. Zhang, Z. Tang, J. Yu, M. Guo, and L. Jiang, “Convex hull properties and algorithms,” Appl. Math. Comput., vol. 216, no. 11, pp. 3209–3218, Aug. 2010, doi: 10.1016/j.amc.2010.04.044.
19. S. Asaeedi, F. Didehvar, and A. Mohades, “α-Concave hull, a generalization of convex hull,” Theor. Comput. Sci., vol. 702, pp. 48–59, Nov. 2017, doi: 10.1016/j.tcs.2017.08.014.