Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
139 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

A Satellite Band Selection Framework for Amazon Forest Deforestation Detection Task (2404.02659v1)

Published 3 Apr 2024 in cs.CV and cs.NE

Abstract: The conservation of tropical forests is a topic of significant social and ecological relevance due to their crucial role in the global ecosystem. Unfortunately, deforestation and degradation impact millions of hectares annually, necessitating government or private initiatives for effective forest monitoring. This study introduces a novel framework that employs the Univariate Marginal Distribution Algorithm (UMDA) to select spectral bands from Landsat-8 satellite, optimizing the representation of deforested areas. This selection guides a semantic segmentation architecture, DeepLabv3+, enhancing its performance. Experimental results revealed several band compositions that achieved superior balanced accuracy compared to commonly adopted combinations for deforestation detection, utilizing segment classification via a Support Vector Machine (SVM). Moreover, the optimal band compositions identified by the UMDA-based approach improved the performance of the DeepLabv3+ architecture, surpassing state-of-the-art approaches compared in this study. The observation that a few selected bands outperform the total contradicts the data-driven paradigm prevalent in the deep learning field. Therefore, this suggests an exception to the conventional wisdom that 'more is always better'.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (41)
  1. SLIC superpixels compared to state-of-the-art superpixel methods. IEEE transactions on pattern analysis and machine intelligence 34, 11 (2012), 2274–2282.
  2. Eduardo Barreto Alexandre. 2017. IFT-SLIC: geração de superpixels com base em agrupamento iterativo linear simples e transformada imagem-floresta. Master’s thesis. Universidade de São Paulo.
  3. Mass detection in mammograms using ga based PCA and Haralick features selection. In 2011 19th Iranian Conference on Electrical Engineering. 1–4.
  4. Evaluation of semantic segmentation methods for deforestation detection in the amazon. ISPRS Archives; 43, B3 43, B3 (2020), 1497–1505.
  5. A training algorithm for optimal margin classifiers. In Workshop on Computational Learning Theory (Pittsburgh, Pennsylvania, United States) (COLT ’92). 144–152.
  6. Generative Semantic Segmentation. arXiv:2303.11316 [cs.CV]
  7. Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation. arXiv:1802.02611 [cs.CV]
  8. ForestEyes Project - Citizen Science and Machine Learning to detect deforested areas in tropical forests. In Anais Estendidos do XXXIV Conference on Graphics, Patterns and Images (Online). SBC, Porto Alegre, RS, Brasil, 14–20. https://doi.org/10.5753/sibgrapi.est.2021.20008
  9. ForestEyes Project: Can Citizen Scientists Help Rainforests?. In IEEE 15th International Conference on eScience. IEEE, 18–27.
  10. TERRABRASILIS: A SPATIAL DATA INFRASTRUCTURE FOR DISSEMINATING DEFORESTATION DATA FROM BRAZIL. Proceedings of the XIX Brazilian Symposium on Remote Sensing, Santos, Brazil (2019), 14–17.
  11. Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition. Ieee, 248–255.
  12. On power Jaccard losses for semantic segmentation. In VISAPP 2021 : 16th International Conference on Computer Vision Theory and Applications. Vienne (on line), Austria. https://hal.science/hal-03139997
  13. The drivers and impacts of Amazon forest degradation. Science 379, 6630 (2023), eabp8622. https://doi.org/10.1126/science.abp8622 arXiv:https://www.science.org/doi/pdf/10.1126/science.abp8622
  14. A New Metric for Evaluating Semantic Segmentation: Leveraging Global and Contour Accuracy. In 2018 IEEE Intelligent Vehicles Symposium (IV). 1051–1056. https://doi.org/10.1109/IVS.2018.8500497
  15. High-resolution global maps of 21st-century forest cover change. science 342, 6160 (2013), 850–853.
  16. Textural features for image classification. IEEE Transactions on systems, man, and cybernetics 6 (1973), 610–621.
  17. Mark Hauschild and Martin Pelikan. 2011. An introduction and survey of estimation of distribution algorithms. Swarm and Evolutionary Computation 1, 3 (2011), 111–128.
  18. INPE. 2022. Estimativa de desmatamento na Amazônia Legal para 2022202220222022 é de 11.56811.56811.56811.568 k⁢m2𝑘superscript𝑚2km^{2}italic_k italic_m start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT. https://encurtador.com.br/aBDP2. Accessed: 2023-08-01.
  19. The next Landsat satellite: The Landsat Data Continuity Mission. Remote Sensing of Environment 122 (2012), 11–21. https://doi.org/10.1016/j.rse.2011.08.026 Landsat Legacy Special Issue.
  20. Ian Jolliffe. 2011. Principal component analysis. Springer.
  21. Diederik P. Kingma and Jimmy Ba. 2017. Adam: A Method for Stochastic Optimization. arXiv:1412.6980 [cs.LG]
  22. Landsat Science 2017. Landsat Science. https://landsat.gsfc.nasa.gov/. Accessed:2017-12-15.
  23. Fully convolutional networks for semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern recognition. 3431–3440.
  24. Eduardo F.P. Luz et al. 2014. The ForestWatchers: A Citizen Cyberscience Project for Deforestation Monitoring in the Tropics. Human Computation 1 (2014), 137–145.
  25. An application of genetic algorithms on band selection for hyperspectral image classification. In Proceedings of the 2003 international conference on machine learning and cybernetics (IEEE Cat. No. 03EX693), Vol. 5. IEEE, 2810–2813.
  26. Adeline Marinho Maciel and Lubia Vinhas. 2019. Reasoning about deforestation trajectories in Pará state, Brazil. In Anais do XIX Simpósio Brasileiro de Sensoriamento Remoto.
  27. Spatio-temporal deep learning approach to map deforestation in amazon rainforest. IEEE Geoscience and Remote Sensing Letters 18, 5 (2020), 771–775.
  28. Claude Martin. 2015. On the Edge: The State and Fate of the World’s Tropical Rainforests. Greystone Books Ltd.
  29. Heinz Mühlenbein and Gerhard Paass. 1996. From recombination of genes to the estimation of distributions I. Binary parameters. In International conference on parallel problem solving from nature. Springer, 178–187.
  30. Hyperspectral band selection using genetic algorithm and support vector machines for early identification of charcoal rot disease in soybean stems. Plant methods 14 (2018), 1–13.
  31. Sebastian Nowozin. 2014. Optimal Decisions from Probabilistic Models: The Intersection-over-Union Case. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
  32. Evaluation of deep learning techniques for deforestation detection in the Amazon forest. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences 4 (2019), 121–128.
  33. Feature selection for splice site prediction: a new method using EDA-based feature ranking. BMC bioinformatics 5 (2004), 1–11.
  34. Metodologia Utilizada nos Projetos PRODES e DETER. São José dos Campos: INPE (2019).
  35. A Deep-Learning Ensemble Method to Detect Atmospheric Rivers and Its Application to Projected Changes in Precipitation Regime. Journal of Geophysical Research: Atmospheres 128, 12 (2023), e2022JD037041. https://doi.org/10.1029/2022JD037041 arXiv:https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/2022JD037041 e2022JD037041 2022JD037041.
  36. Deforestation Detection with Fully Convolutional Networks in the Amazon Forest from Landsat-8 and Sentinel-2 Images. Remote Sensing 13, 24 (2021).
  37. Tropical deforestation.
  38. Large-area classification and mapping of forest and land cover in the Brazilian Amazon: A comparative analysis of ALOS/PALSAR and Landsat data sources. IEEE Journal of selected topics in applied earth observations and remote sensing 3, 4 (2010), 594–604.
  39. Feature Selection on the Basis of Rough Set Theory and Univariate Marginal Distribution Algorithm. In Proceedings of the 2017 International Conference on Applied Mathematics, Modelling and Statistics Application (AMMSA 2017). Atlantis Press, 369–372. https://doi.org/10.2991/ammsa-17.2017.83
  40. Selection of landsat 8 OLI band combinations for land use and land cover classification. In 2019 8th International Conference on Agro-Geoinformatics (Agro-Geoinformatics). IEEE, 1–5.
  41. Optimal band selection for high dimensional remote sensing data using genetic algorithm. In Second International Conference on Earth Observation for Global Changes, Vol. 7471. SPIE, 522–528.
Citations (1)
View on Semantic Scholar

Summary

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

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