Convolutional neural network-based semantic segmentation model for land cover classification in páramo ecosystems
Submitted: 2024-06-07
|Accepted: 2024-12-11
|Published: 2025-01-15
Copyright (c) 2024 Marcela Reyes Quintana, Iván Lizarazo
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Downloads
Keywords:
remote sensing, páramo, convolutional neural networks, semantic segmentation
Supporting agencies:
Universidad Nacional de Colombia, Bogotá
Abstract:
Páramo ecosystems are essential for water regulation and biodiversity conservation in mountainous areas. However, they face significant threats due to climate change and human activities such as agriculture, livestock farming, and mining. The absence of clear boundaries and continuous monitoring systems for their land cover hinders effective protection highlighting the need to employ advanced digital techniques that provide highly accurate, up-to-date information. Convolutional neural networks (CNNs) have emerged as promising tools for semantic segmentation of satellite images. This research aimed to evaluate the performance of two CNNs architectures U-Net++ and DeepLabV3+ for land cover classification in the Tota-Bijagual-Mamapacha (TBM) páramo complex in Colombia, using Landsat 8 imagery from 2017 to 2019 and land cover labels from 1:100.000, national coverage map produced by IDEAM in 2018. The results showed U-Net++ achieved a kappa of 0.60, while DeepLabV3+ obtained a kappa of 0.59. In páramo covers, U-Net++ achieved an F1 of 78.43% for Herbazal and 79.22% for Forests, while DeepLabV3+ achieved F1 of 75% and 74.27%, respectively, confirming the potential of CNNs for land cover classification in these ecosystems. Although both models presented similar processing times, class imbalance and reliance on consistent labels affected their performance in heterogeneous covers. This research establishes a methodological foundation for future studies and suggests addressing these limitations to
improve efficiency and thematic accuracy in the classification and monitoring in páramo ecosystems.
References:
Alzubaidi, L., Zhang, J., Humaidi, A., Al-Dujaili, A., Duan, Y., Al-Shamma, O., Santamaría, J., Fadhel, M., Al-Amidie, M., Farhan, L. 2021. Review of deep learning: concepts, CNN architectures, challenges, applications, future directions. Journal of Big Data, 8(1). https://doi.org/10.1186/s40537-021-00444-8
Ayala, J. E., Márquez, C. O., García, V. J., Recalde-Moreno, C. G., Rodríguez-Llerena, M. V., & Damián-Carrión, D. A. (2017). Land cover classification in an ecuadorian mountain geosystem using a random forest classifier, spectral vegetation indices, and ancillary geographic data. Geosciences (Switzerland), 7(2). https://doi.org/10.3390/geosciences7020034
Ayhan, B., Kwan, C., Budavari, B., Kwan, L., Lu, Y., Perez, D., Li, J., Skarlatos, D., & Vlachos, M. (2020). Vegetation detection using deep learning and conventional methods. Remote Sensing, 12(15). https://doi.org/10.3390/rs12152502
Belyadi, H., & Haghighat, A. (2021). Model evaluation. In Machine Learning Guide for Oil and Gas Using Python. https://doi.org/10.1016/b978-0-12-821929-4.00009-3
Brück, S. A., Torres, B. D. M., & de Moraes Polizeli, M. de L. T. (2023). The Ecuadorian paramo in danger: What we know and what might be learned from northern wetlands. In Global Ecology and Conservation (Vol. 47). Elsevier B.V. https://doi.org/10.1016/j.gecco.2023.e02639
Buda, M., Maki, A., Mazurowski, M.A. 2018. A systematic study of the class imbalance problem in convolutional neural networks. Neural Networks, 106, 249-259. https://doi.org/10.1016/j.neunet.2018.07.011
Buytaert, W., Célleri, R., De Bièvre, B., Cisneros, F., Wyseure, G., Deckers, J., Hofstede, R. 2006. Human impact on the hydrology of the Andean páramos. Earth-Science Reviews, 79(1-2), 53-72. https://doi.org/10.1016/j.earscirev.2006.06.002
Cabrera, M., Ramírez, W. 2014. Restauración Ecológica de los páramos de Colombia: Transformación y herramientas para su conservación. Instituto de Investigación de Recursos Biológicos Alexander von Humboldt (IAvH), 296.
Castelo, M., Piedra, J. A., Ayala, R. 2022. Deep learning for land use and land cover classification from the Ecuadorian Paramo. International Journal of Digital Earth, 15(1), 1001-1017. https://doi.org/10.1080/17538947.2022.2088872
Castellanos, H., Gómez, W.F., Mayorga, N. 2021. Mapa nacional de coberturas de la tierra, escala 1:100.000, periodo 2018. Metodología Corine Land Cover adaptada para Colombia. Memoria técnica y resultados. Instituto de Hidrología, Meteorología y Estudios Ambientales (Ideam). Bogotá, D. C., Colombia.
Chen, L.C., Papandreou, G., Schroff, F., & Adam, H. 2017. Rethinking Atrous Convolution for Semantic Image Segmentation. http://arxiv.org/abs/1706.05587
Chen, Z., Duan, J., Kang, L., & Qiu, G. 2022. Class-Imbalanced Deep Learning via a Class-Balanced Ensemble. IEEE Transactions on Neural Networks and Learning Systems, 33(10), 5626-5640. https://doi.org/10.1109/TNNLS.2021.3071122
Coca, A., Zaraza, M. A., Benavides, Y. T., Montilla, Y. M., Posada, H. B., Avendaño, A. L., Hernández, H. A., Garzón, S. C., & Franco, C. A. 2021. Evaluación de algoritmos de clasificación en la plataforma Google Earth Engine para la identificación y detección de cambios de construcciones rurales y periurbanas a partir de imágenes de alta resolución. Revista de Teledetección, 58, 71. https://doi.org/10.4995/raet.2021.15026
Deepan, P., Sudha, L.R. 2020. Object Classification of Remote Sensing Image Using Deep Convolutional Neural Network. In The Cognitive Approach in Cloud Computing and Internet of Things Technologies for Surveillance Tracking Systems (pp. 107-120). Elsevier
https://doi.org/10.1016/B978-0-12-816385-6.00008-8
Earth Engine Data Catalog. 2022. USGS Landsat 8 Level 2, Collection 2, Tier 2 | Earth Engine Data Catalog | Google Developers. https://developers.google.com/earth-engine/datasets/catalog/LANDSAT_LC08_C01_T2_SR
Géron, A. 2019. Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow: concepts, tools, and techniques to build intelligent systems. O'Reilly Media, Inc.
Instituto de Hidrología, M. y E.A. (IDEAM). 2010. Leyenda nacional de coberturas de la tierra. Metodología CORINE Land Cover adaptada para Colombia, escala 1:100.000. In Area: Vol. TH-62-04-1 (Issue 257).
Instituto de Hidrología, M. y E.A. (IDEAM), Instituto de Investigación de Recursos Biológicos Alexander von Humboldt (IAVH), Ministerio de Ambiente y Desarrollo Sostenible (MADS), Ministro de Hacienda y Crédito Público (MinHacienda). 2014. Informe final componente coberturas de la tierra en los páramos priorizados a escala 1:25.000 (metodología corine land cover adaptada para Colombia) (Vol. 000, Issue 13).
Iyer, P., A, S., & Lal, S. 2021. Deep learning ensemble method for classification of satellite hyperspectral images. Remote Sensing Applications: Society and Environment, 23, 100580. https://doi.org/10.1016/j.rsase.2021.100580
Krizhevsky, B.A., Sutskever, I., Hinton, G.E. 2012. ImageNet Classification with Deep Convolutional Neural Networks. Communications of the ACM, 60(6), 84-90. https://doi.org/10.1145/3065386
Kussul, N., Lavreniuk, M., Skakun, S., & Shelestov, A. 2017. Deep Learning Classification of Land Cover and Crop Types Using Remote Sensing Data. IEEE Geoscience and Remote Sensing Letters, 14(5), 778-782. https://doi.org/10.1109/LGRS.2017.2681128
Liu, M., Fu, B., Xie, S., He, H., Lan, F., Li, Y., Lou, P., Fan, D. 2021. Comparison of multi-source satellite images for classifying marsh vegetation using DeepLabV3 Plus deep learning algorithm. Ecological Indicators, 125, 107562. https://doi.org/10.1016/j.ecolind.2021.107562
Ma, L., Liu, Y., Zhang, X., Ye, Y., Yin, G., Johnson, B.A. 2019. Deep learning in remote sensing applications: A meta-analysis and review. In ISPRS Journal of Photogrammetry and Remote Sensing (Vol. 152, pp. 166-177). Elsevier B.V. https://doi.org/10.1016/j.isprsjprs.2019.04.015
Martin-Abadal, M., Guerrero-Font, E., Bonin-Font, F., & Gonzalez-Cid, Y. 2018. Deep Semantic Segmentation in an AUV for Online Posidonia Oceanica Meadows Identification. IEEE Access, 6, 60956-60967. https://doi.org/10.1109/ACCESS.2018.2875412
Maxwell, A.E., Warner, T.A., Fang, F. 2018. Implementation of machine-learning classification in remote sensing: An applied review. In International Journal of Remote Sensing (Vol. 39, Issue 9, pp. 2784-2817). Taylor and Francis Ltd. https://doi.org/10.1080/01431161.2018.1433343
Ministerio de Ambiente y Desarrollo Sostenible (MADS). 2016. Resolución No 1771 de 2016. In Diario Oficial de Colombia. Ministerio de Ambiente y Desarrollo Sostenible, Bogotá, Colombia. https://www.minambiente.gov.co/documentonormativa/resolucion-1771-de-2016/
Morales, M., Otero, J., Van der Hammen, T., Torres, A., Cadena, C., Pedraza, C., Rodríguez, N., Franco, C., Betancourth, J.C., Olaya, E., Posada, E., Cárdenas, L. 2007. Atlas de páramos de Colombia. In Instituto de Investigación de Recursos Biológicos Alexander von Humboldt (IAVH). (Vol. 16, Issue 1).
Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. 2015. U-Net: Convolutional Networks for Biomedical Image Segmentation. Lecture Notes in Computer Science, 9351(Cvd), 12-20.
Poveda Sotelo, Y., Bermúdez Cella, M. A., & Gil Leguizamón, P. 2022. Evaluation of supervised classification methods for the estimation of spatiotemporal changes in the Merchán and Telecom paramos, Colombia. Boletin de Geologia, 44(2), 51-72. https://doi.org/10.18273/revbol.v44n2-2022002
Suárez, A. S., Jiménez, A. F., Castro, M., & Cruz, A. A. 2017. Clasificación y mapeo automático de coberturas del suelo en imágenes satelitales utilizando Redes Neuronales Convolucionales. Orinoquia, 21(1 Sup), 64-75. https://doi.org/10.22579/20112629.432
Talukdar, S., Singha, P., Mahato, S., Shahfahad, Pal, S., Liou, Y. A., & Rahman, A. 2020. Land-use land-cover classification by machine learning classifiers for satellite observations-A review. In Remote Sensing (Vol. 12, Issue 7). MDPI AG. https://doi.org/10.3390/rs12071135
Vásquez, A., Buitrago, A. C. 2011. El gran libro de los páramos. 2011. Instituto de Investigación de Recursos Biológicos Alexander von Humboldt. Proyecto Páramo Andino, 208.
Vergara, P.A. 2020. Estrategias implementadas por el Sistema Nacional de Áreas Protegidas de Colombia para conservar los páramos. Revista de Ciencias Ambientales, 54(1), 167-176. https://doi.org/10.15359/rca.54-1.9
Yuan, X., Shi, J., & Gu, L. 2021. A review of deep learning methods for semantic segmentation of remote sensing imagery. In Expert Systems with Applications (Vol. 169). Elsevier Ltd. https://doi.org/10.1016/j.eswa.2020.114417
Zhou, Z., Siddiquee, M.M., Tajbakhsh, N., Liang, J. 2018. UNet++: A Nested U-Net Architecture for Medical Image Segmentation. Lecture Notes in Computer Science. http://arxiv.org/abs/1807.10165 https://doi.org/10.1007/978-3-030-00889-5_1
