KaydedilenlerDerginizi ekleyin

Dergiler / Türkiye Uzaktan Algılama Dergisi / 2026 / Volume: 8

Benchmarking ensemble machine learning methods for all-weather SAR land cover mapping using OpenEarthMap-SAR

Melih Altay(HACETTEPE ÜNİVERSİTESİ, MÜHENDİSLİK FAKÜLTESİ, GEOMATİK MÜHENDİSLİĞİ BÖLÜMÜ, GEOMATİK MÜHENDİSLİĞİ PR. (İNGİLİZCE))0009-0005-0902-3713
Sayfa
—
Yayın Tarihi
2026-07-03
🏛️Bu makalenin bibliyografik verileri ve tam metin bağlantısı DergiPark üzerinden sağlanmıştır.
DergiPark Kaydı ↗
✨

Yapay Zeka Hızlı Bakış & Çıkarımlar

AI Insight
⚡ TL;DR(1 Cümlede Hızlı Çıkarım)

These findings establish strong and reproducible machine learning benchmarks for the OpenEarthMap-SAR benchmark and provide valuable insights into the role of feature-based ensemble learning for high-resolution, all-weather SAR land cover mapping.

📌 3 Maddede Bu Makale

🎯
Amaç & Hipotez:The growing demand for reliable, high-resolution land cover information has increased interest in Synthetic Aperture Radar (SAR) imagery due to its ability to provide consistent observations under all-weather and day–night conditions.
📊
Temel Bulgular & Veri:Experimental results demonstrate that gradient boosting–based methods substantially outperform the classical RF benchmark.
💡
Sonuç & Katkı:These findings establish strong and reproducible machine learning benchmarks for the OpenEarthMap-SAR benchmark and provide valuable insights into the role of feature-based ensemble learning for high-resolution, all-weather SAR land cover mapping.
🔬Tür:Modelleme & Sayısal Simülasyon
📍Kapsam:Genişletilmiş Akademik Örneklem
⚙️Değişkenler:Synthetic aperture radar · Land cover mapping · Machine Learning

Abstract

The growing demand for reliable, high-resolution land cover information has increased interest in Synthetic Aperture Radar (SAR) imagery due to its ability to provide consistent observations under all-weather and day–night conditions. Despite these advantages, SAR-based land cover classification remains challenging because of speckle noise, complex scattering mechanisms, and the historical lack of large-scale, high-quality benchmark datasets. The recently introduced OpenEarthMap-SAR dataset addresses this limitation by enabling systematic and reproducible evaluation of SAR-based land cover mapping methods at submeter spatial resolution. This study presents a feature-based ensemble machine learning framework for all-weather land cover mapping using the OpenEarthMap-SAR benchmark. A comprehensive set of handcrafted features, including SAR backscatter statistics, polarization descriptors, and Gray-Level Co-occurrence Matrix (GLCM) texture measures, was extracted to characterize radiometric and spatial properties of diverse land cover types. Three widely used ensemble learning algorithms Random Forest (RF), Light Gradient Boosting Machine (LightGBM), and Extreme Gradient Boosting (XGBoost) were systematically evaluated on a representative experimental subset of the dataset comprising high-resolution SAR imagery across eight land cover classes. Experimental results demonstrate that gradient boosting–based methods substantially outperform the classical RF benchmark. XGBoost achieved the highest overall accuracy of 92.4%, corresponding to a 5.1% improvement over RF, while LightGBM offered an effective tradeoff between classification accuracy (91.7%) and computational efficiency. Class-wise analysis indicates that water and vegetation classes are consistently well discriminated, whereas confusion persists among spectrally and structurally similar urban classes such as roads and buildings. Feature importance analysis highlights the dominant role of VV polarization and texture-based features in SAR land cover discrimination. These findings establish strong and reproducible machine learning benchmarks for the OpenEarthMap-SAR benchmark and provide valuable insights into the role of feature-based ensemble learning for high-resolution, all-weather SAR land cover mapping.

🔖

Bu Makaleye Atıf Yap

Altay, M. (2026). Benchmarking ensemble machine learning methods for all-weather SAR land cover mapping using OpenEarthMap-SAR. Türkiye Uzaktan Algılama Dergisi, Volume: 8. https://doi.org/10.51489/tuzal.1844395

📄 Tam Metin PDF Okuyucu

📄

PDF görüntüleyici hazırlanıyor...

Kaynakça (25)

  1. Ban, Y., & Yousif, O. (2012). Multitemporal spaceborne SAR data for urban change detection. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 5(4), 1087–1094. https://doi.org/10.1109/JSTARS.2012.2201135
  2. Belgiu, M., & Drăguţ, L. (2016). Random forest in remote sensing. ISPRS Journal of Photogrammetry and Remote Sensing, 114, 24–31. https://doi.org/10.1016/j.isprsjprs.2016.01.011
  3. Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324
  4. Camps-Valls, G., Tuia, D., Bruzzone, L., & Benediktsson, J. A. (2014). Advances in hyperspectral image classification. IEEE Signal Processing Magazine, 31(1), 45–54. https://doi.org/10.1109/MSP.2013.2279179
  5. Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings Book of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, New York, USA, 785–794. https://doi.org/10.1145/2939672.2939785
  6. Cloude, S., & Pottier, E. (1997). An entropy-based classification scheme for land applications of polarimetric SAR. IEEE Transactions on Geoscience and Remote Sensing, 35(1), 68–78. https://doi.org/10.1109/36.551935
  7. Freeman, A., & Durden, S. L. (1998). A three-component scattering model for polarimetric SAR data. IEEE Transactions on Geoscience and Remote Sensing, 36(3), 963–973. https://doi.org/10.1109/36.673687
  8. Inglada, J., Vincent, A., Arias, M., Tardy, B., Morin, D., & Rodes, I. (2017). Operational high resolution land cover map production at the country scale using satellite image time series. Remote Sensing, 9(1), 95. https://doi.org/10.3390/rs9010095
  9. Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q., & Liu, T.-Y. (2017). LightGBM: A highly efficient gradient boosting decision tree. In Advances in Neural Information Processing Systems, 30, 3146–3154. https://dl.acm.org/doi/10.5555/3294996.3295074
  10. Maxwell, A. E., Warner, T. A., & Fang, F. (2018). Implementation of machine-learning classification in remote sensing: An applied review. International Journal of Remote Sensing, 39(9), 2784–2817. https://doi.org/10.1080/01431161.2018.1433343
  11. Mou, L., & Zhu, X. X. (2020). Learning to pay attention on spectral domain: A spectral attention module-based convolutional network for hyperspectral image classification. IEEE Transactions on Geoscience and Remote Sensing, 58(1), 110–122. https://doi.org/10.1109/TGRS.2019.2933609
  12. Mountrakis, G., Im, J., & Ogole, C. (2011). Support vector machines in remote sensing: A review. ISPRS Journal of Photogrammetry and Remote Sensing, 66(3), 247–259. https://doi.org/10.1016/j.isprsjprs.2010.11.001
  13. Oliver, C., & Quegan, S. (2004). Understanding synthetic aperture radar images. SciTech Publishing.
  14. Pelletier, C., Webb, G. I., & Petitjean, F. (2019). Temporal convolutional neural network for the classification of satellite image time series. Remote Sensing, 11(5), 523. https://doi.org/10.3390/rs11050523
  15. Pesaresi, M., Ehrlich, D., Ferri, S., Florczyk, A., Freire, S., Halkia, S., Julea, A., Kemper, T., Soille, P., & Syrris, V. (2016). Operating procedure for the production of the Global Human Settlement Layer from Landsat data of the epochs 1975, 1990, 2000, and 2014. Publications Office of the European Union. https://doi.org/10.2788/253582
  16. Richards, J. A. (2009). Remote sensing with imaging radar. Springer.
  17. Ronneberger, O., Fischer, P., & Brox, T. (2015). U-Net: Convolutional networks for biomedical image segmentation. Proceedings Book of Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015, Munich, Germany, 234–241. https://doi.org/10.1007/978-3-319-24574-4_28
  18. Şimşek, F. F. (2025a). Comparison of agricultural crop type classifications with different machine learning algorithms (RF–SVM–ANN–XGBoost) by generating ground truth data from farmer declaration parcels. International Journal of Engineering and Geosciences, 10(2), 207–220. https://doi.org/10.26833/ijeg.1552141
  19. Şimşek, F. F., & Durduran, S. S. (2023). Crop classification using multi-temporal Sentinel-2 imagery and the open-source EO-Learn library. Journal of Geodesy and Geoinformation, 10(1), 45–62. https://doi.org/10.9733/JGG.2023R0004.T
  20. Şimşek, F. F. (2025b). Determination of land use and land cover change using multi-temporal PlanetScope images and deep learning CNN model. Paddy and Water Environment, 23, 405–423. https://doi.org/10.1007/s10333-025-01024-9
  21. Touzi, R. (2007). Target scattering decomposition in terms of roll-invariant target parameters. IEEE Transactions on Geoscience and Remote Sensing, 45(1), 73–84. https://doi.org/10.1109/TGRS.2006.887989
  22. Ulaby, F. T., Moore, R. K., & Fung, A. K. (1986). Microwave remote sensing: Active and passive (Vol. 3). Artech House.
  23. Xia, J., Chen, H., Broni-Bediako, C., Wei, Y., Song, J., & Yokoya, N. (2025). OpenEarthMap-SAR: A benchmark synthetic aperture radar dataset for global high-resolution land cover mapping [software and data sets]. IEEE Geoscience and Remote Sensing Magazine, 13(4), 476–487. https://doi.org/10.1109/MGRS.2025.3599512
  24. Yokoya, N., Ghamisi, P., Xia, J., Sukhanov, S., Heremans, R., Tankoyeu, I., Bechtel, B., Le Saux, B., Moser, G., & Tuia, D. (2018). Open data for global multimodal land use classification: Outcome of the 2017 IEEE GRSS data fusion contest. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 11(5), 1363–1377. https://doi.org/10.1109/JSTARS.2018.2799698
  25. Zhu, Y., Liu, K., Myint, S. W., Du, Z., Li, Y., Cao, J., Liu, L., & Wu, Z. (2020). Integration of GF2 optical, GF3 SAR, and UAV data for estimating aboveground biomass of China's largest artificially planted mangroves. Remote Sensing, 12(12), 2039. https://doi.org/10.3390/rs12122039