Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
143 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Reuse out-of-year data to enhance land cover mapping via feature disentanglement and contrastive learning (2404.11114v1)

Published 17 Apr 2024 in cs.LG

Abstract: Timely up-to-date land use/land cover (LULC) maps play a pivotal role in supporting agricultural territory management, environmental monitoring and facilitating well-informed and sustainable decision-making. Typically, when creating a land cover (LC) map, precise ground truth data is collected through time-consuming and expensive field campaigns. This data is then utilized in conjunction with satellite image time series (SITS) through advanced machine learning algorithms to get the final map. Unfortunately, each time this process is repeated (e.g., annually over a region to estimate agricultural production or potential biodiversity loss), new ground truth data must be collected, leading to the complete disregard of previously gathered reference data despite the substantial financial and time investment they have required. How to make value of historical data, from the same or similar study sites, to enhance the current LULC mapping process constitutes a significant challenge that could enable the financial and human-resource efforts invested in previous data campaigns to be valued again. Aiming to tackle this important challenge, we here propose a deep learning framework based on recent advances in domain adaptation and generalization to combine remote sensing and reference data coming from two different domains (e.g. historical data and fresh ones) to ameliorate the current LC mapping process. Our approach, namely REFeD (data Reuse with Effective Feature Disentanglement for land cover mapping), leverages a disentanglement strategy, based on contrastive learning, where invariant and specific per-domain features are derived to recover the intrinsic information related to the downstream LC mapping task and alleviate possible distribution shifts between domains. Additionally, REFeD is equipped with an effective supervision scheme where feature disentanglement is further enforced via multiple levels of supervision at different granularities. The experimental assessment over two study areas covering extremely diverse and contrasted landscapes, namely Koumbia (located in the West-Africa region, in Burkina Faso) and Centre Val de Loire (located in centre Europe, France), underlines the quality of our framework and the obtained findings demonstrate that out-of-year information coming from the same (or similar) study site, at different periods of time, can constitute a valuable additional source of information to enhance the LC mapping process.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (34)
  1. A theory of learning from different domains. Machine Learning, 79:151–175.
  2. Domain adaptation problems: A dasvm classification technique and a circular validation strategy. IEEE transactions on pattern analysis and machine intelligence, 32(5):770–787.
  3. Temporal-domain adaptation for satellite image time-series land-cover mapping with adversarial learning and spatially aware self-training. IEEE J. Sel. Top. Appl. Earth Obs. Remote. Sens., 16:3645–3675.
  4. Why do we need large batchsizes in contrastive learning? A gradient-bias perspective. In NeurIPS.
  5. Towards domain-specific features disentanglement for domain generalization. CoRR, abs/2310.03007.
  6. Cross-domain multi-prototypes with contradictory structure learning for semi-supervised domain adaptation segmentation of remote sensing images. Remote. Sens., 15(13):3398.
  7. Optical remotely sensed time series data for land cover classification: A review. ISPRS Journal of Photogrammetry and Remote Sensing, 116:55–72.
  8. A spectral-temporal constrained deep learning method for tree species mapping of plantation forests using time series sentinel-2 imagery. ISPRS Journal of Photogrammetry and Remote Sensing, 204:397–420.
  9. Land cover classification via multitemporal spatial data by deep recurrent neural networks. IEEE Geosc. and Rem. Sens. Letters, 14(10):1685–1689.
  10. Assessment of an operational system for crop type map production using high temporal and spatial resolution satellite optical imagery. Remote Sensing, 7(9):12356–12379.
  11. POEM: polarization of embeddings for domain-invariant representations. In Williams, B., Chen, Y., and Neville, J., editors, AAAI, pages 8150–8158.
  12. Harmonized in situ datasets for agricultural land use mapping and monitoring in tropical countries. Earth System Science Data, 13(12):5951–5967.
  13. Spatial dependence between training and test sets: another pitfall of classification accuracy assessment in remote sensing. Machine Learning.
  14. Towards delivering on the sustainable development goals using earth observations. Remote Sensing of Environment, 247:111930.
  15. In NeurIPS.
  16. Regional scale mapping of grassland mowing frequency with sentinel-2 time series. Remote Sensing, 10(8):1221.
  17. Towards inheritable models for open-set domain adaptation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 12376–12385.
  18. Conditional adversarial domain adaptation. In Bengio, S., Wallach, H., Larochelle, H., Grauman, K., Cesa-Bianchi, N., and Garnett, R., editors, Advances in Neural Information Processing Systems, volume 31. Curran Associates, Inc.
  19. Decoupled weight decay regularization. In 7th International Conference on Learning Representations, ICLR 2019, New Orleans, LA, USA, May 6-9, 2019. OpenReview.net.
  20. A bayesian-inspired, deep learning-based, semi-supervised domain adaptation technique for land cover mapping. Mach. Learn., 112(6):1941–1973.
  21. Improvement in crop mapping from satellite image time series by effectively supervising deep neural networks. ISPRS Journal of Photogrammetry and Remote Sensing, 198:272–283.
  22. Timematch: Unsupervised cross-region adaptation by temporal shift estimation. ISPRS Journal of Photogrammetry and Remote Sensing, 188:301–313.
  23. Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12:2825–2830.
  24. Temporal convolutional neural network for the classification of satellite image time series. Remote. Sens., 11(5):523.
  25. Data-centric machine learning for geospatial remote sensing data. CoRR, abs/2312.05327.
  26. EuroCrops: The Largest Harmonized Open Crop Dataset Across the European Union. Scientific Data, 10(1):612.
  27. Deep CORAL: correlation alignment for deep domain adaptation. In Hua, G. and Jégou, H., editors, Computer Vision - ECCV 2016 Workshops - Amsterdam, The Netherlands, October 8-10 and 15-16, 2016, Proceedings, Part III, volume 9915, pages 443–450.
  28. Recent advances in domain adaptation for the classification of remote sensing data. Geoscience and Remote Sensing Magazine, 4(2):41–57.
  29. Visualizing data using t-sne. Journal of Machine Learning Research, 9:2579–2605.
  30. A survey of unsupervised deep domain adaptation. ACM Trans. Intell. Syst. Technol., 11(5):51:1–51:46.
  31. Opening the archive: How free data has enabled the science and monitoring promise of landsat author links open overlay panel. Remote Sensing of Environment, 122:2–10.
  32. Semi-supervised domain adaptation with source label adaptation. In CVPR, pages 24100–24109. IEEE.
  33. Data-centric artificial intelligence: A survey. CoRR, abs/2303.10158.
  34. Deep learning based multi-temporal crop classification. Remote Sensing of Environment, 221:430–443.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now