Using high-resolution imagery and deep learning to classify land-use following deforestation: a case study in Ethiopia

Simple item page
cg.authorship.typesCGIAR and developing country instituteen_US
cg.authorship.typesCGIAR and advanced research instituteen_US
cg.contributor.affiliationWageningen University & Researchen_US
cg.contributor.affiliationUniversity of Münsteren_US
cg.contributor.affiliationBioversity International and the International Center for Tropical Agricultureen_US
cg.contributor.affiliationEnvironment, Forest and Climate Change Commission, Ethiopiaen_US
cg.contributor.affiliationCenter for International Forestry Researchen_US
cg.contributor.affiliationInternational Center for Tropical Agricultureen_US
cg.contributor.crpForests, Trees and Agroforestryen_US
cg.contributor.donorFederal Ministry for the Environment, Nature Conservation, Nuclear Safety and Consumer Protection, Germanyen_US
cg.contributor.donorCGIAR Trust Funden_US
cg.contributor.donorNorwegian Agency for Development Cooperationen_US
cg.contributor.donorEuropean Unionen_US
cg.contributor.initiativeLow-Emission Food Systemsen_US
cg.coverage.countryEthiopiaen_US
cg.coverage.iso3166-alpha2ETen_US
cg.coverage.regionAfricaen_US
cg.coverage.regionSub-Saharan Africaen_US
cg.coverage.regionEastern Africaen_US
cg.howPublishedFormally Publisheden_US
cg.identifier.doihttps://doi.org/10.1080/15481603.2022.2115619en_US
cg.isijournalISI Journalen_US
cg.issn1548-1603en_US
cg.issue1en_US
cg.journalGIScience & Remote Sensingen_US
cg.reviewStatusPeer Reviewen_US
cg.subject.actionAreaSystems Transformationen_US
cg.volume58en_US
dc.contributor.authorMasolele, Robert N.en_US
dc.contributor.authorSy, Veronique deen_US
dc.contributor.authorMarcos, Diegoen_US
dc.contributor.authorVerbesselt, Janen_US
dc.contributor.authorGieseke, Fabianen_US
dc.contributor.authorMulatu, Kalkidan Ayeleen_US
dc.contributor.authorMoges, Yitebituen_US
dc.contributor.authorSebrala, Heiruen_US
dc.contributor.authorMartius, Christopheren_US
dc.contributor.authorHerold, Martinen_US
dc.date.accessioned2022-10-13T14:25:33Zen_US
dc.date.available2022-10-13T14:25:33Zen_US
dc.identifier.urihttps://hdl.handle.net/10568/125031en_US
dc.titleUsing high-resolution imagery and deep learning to classify land-use following deforestation: a case study in Ethiopiaen_US
dcterms.abstractNational-scale assessments of post-deforestation land-use are crucial for decreasing deforestation and forest degradation-related emissions. In this research, we assess the potential of different satellite data modalities (single-date, multi-date, multi-resolution, and an ensemble of multi-sensor images) for classifying land-use following deforestation in Ethiopia using the U-Net deep neural network architecture enhanced with attention. We performed the analysis on satellite image data retrieved across Ethiopia from freely available Landsat-8, Sentinel-2 and Planet-NICFI satellite data. The experiments aimed at an analysis of (a) single-date images from individual sensors to account for the differences in spatial resolution between image sensors in detecting land-uses, (b) ensembles of multiple images from different sensors (Planet-NICFI/Sentinel-2/Landsat-8) with different spatial resolutions, (c) the use of multi-date data to account for the contribution of temporal information in detecting land-uses, and, finally, (d) the identification of regional differences in terms of land-use following deforestation in Ethiopia. We hypothesize that choosing the right satellite imagery (sensor) type is crucial for the task. Based on a comprehensive visually interpreted reference dataset of 11 types of post-deforestation land-uses, we find that either detailed spatial patterns (single-date Planet-NICFI) or detailed temporal patterns (multi-date Sentinel-2, Landsat-8) are required for identifying land-use following deforestation, while medium-resolution single-date imagery is not sufficient to achieve high classification accuracy. We also find that adding soft-attention to the standard U-Net improved the classification accuracy, especially for small-scale land-uses. The models and products presented in this work can be used as a powerful data resource for governmental and forest monitoring agencies to design and monitor deforestation mitigation measures and data-driven land-use policy.en_US
dcterms.accessRightsOpen Accessen_US
dcterms.audienceAcademicsen_US
dcterms.audienceCGIARen_US
dcterms.audienceScientistsen_US
dcterms.available2022-09-07en_US
dcterms.bibliographicCitationMasolele, R. N., De Sy, V., Marcos, D., Verbesselt, J., Gieseke, F., Mulatu, K. A., Moges, Y., Sebrala, H., Martius, C., & Herold, M. (2022). Using high-resolution imagery and deep learning to classify land-use following deforestation: a case study in Ethiopia. In GIScience & Remote Sensing (Vol. 59, Issue 1, pp. 1446–1472). Informa UK Limited. https://doi.org/10.1080/15481603.2022.2115619en_US
dcterms.extent1446-1472en_US
dcterms.issued2022-12-31en_US
dcterms.languageenen_US
dcterms.licenseCC-BY-4.0en_US
dcterms.publisherInforma UK Limiteden_US
dcterms.replaceshttps://hdl.handle.net/10568/128140en_US
dcterms.subjectdeforestationen_US
dcterms.subjectforestryen_US
dcterms.subjectgeographical information systemsen_US
dcterms.subjectremote sensingen_US
dcterms.subjectclimate changeen_US
dcterms.subjectsatellite imageryen_US
dcterms.typeJournal Articleen_US

Files

License bundle

Now showing 1 - 1 of 1
Thumbnail Image
Name:
license.txt
Size:
1.75 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

CGIAR Initiative on Low-Emission Food Systems
Alliance Bioversity CIAT Journal Articles
Alliance Research Lever 2: Multifunctional Landscapes
CIFOR publications
FTA outputs