1. ERDC Knowledge Core
  2. Engineer Research and Development Center (ERDC)
  3. Geospatial Research Laboratory (GRL) (formerly Topographic Engineering Center (TEC))
  4. Publication
  5. Technical Report
Please use this identifier to cite or link to this item: https://hdl.handle.net/11681/42402
Full metadata record
DC FieldValueLanguage
dc.contributor.authorLasko, Kristofer D.-
dc.contributor.authorSava, Elena.-
dc.date.accessioned2021-11-19T16:50:10Z-
dc.date.available2021-11-19T16:50:10Z-
dc.date.issued2021-11-
dc.identifier.govdocERDC/GRL TR-21-7-
dc.identifier.urihttps://hdl.handle.net/11681/42402-
dc.identifier.urihttp://dx.doi.org/10.21079/11681/42402-
dc.descriptionTechnical Reporten_US
dc.description.abstractLand cover type is a fundamental remote sensing-derived variable for terrain analysis and environmental mapping applications. The currently available products are produced only for a single season or a specific year. Some of these products have a coarse resolution and quickly become outdated, as land cover type can undergo significant change over a short time period. In order to enable on-demand generation of timely and accurate land cover type products, we developed a sensor-agnostic framework leveraging pre-trained machine learning models. We also generated land cover models for Sentinel-2 (20m) and Landsat 8 imagery (30m) using either a single date of imagery or two dates of imagery for mapping land cover type. The two-date model includes 11 land cover type classes, whereas the single-date model contains 6 classes. The models’ overall accuracies were 84% (Sentinel-2 single date), 82% (Sentinel-2 two date), and 86% (Landsat 8 two date) across the continental United States. The three different models were built into an ArcGIS Pro Python toolbox to enable a semi-automated workflow for end users to generate their own land cover type maps on demand. The toolboxes were built using parallel processing and image-splitting techniques to enable faster computation and for use on less-powerful machines.en_US
dc.description.sponsorshipGeospatial Research Laboratory (U.S.)en_US
dc.description.tableofcontentsAbstract ............................................................................................................................ ii Figures and Table ............................................................................................................ iv Preface ............................................................................................................................. vi Acronyms and Abbreviations .......................................................................................... vii 1 Introduction ............................................................................................................... 1 1.1 Background ....................................................................................................... 1 1.2 Image classification .......................................................................................... 3 1.2.1 Support vector machines ................................................................................................... 4 1.2.2 Classifier training ............................................................................................................... 4 1.2.3 Classification accuracy assessment ................................................................................. 5 1.3 Objectives .......................................................................................................... 6 2 Data and Training Sites ............................................................................................ 8 2.1 Landsat 8 and Sentinel-2 imagery .................................................................. 8 2.2 Description of training sites ............................................................................. 9 2.3 Land cover datasets ...................................................................................... 10 3 Methodology ............................................................................................................ 12 3.1 Methodology .................................................................................................. 12 3.2 Python toolbox design ................................................................................... 16 3.2.1 ArcGIS Pro Python toolbox background .......................................................................... 16 3.2.2 ETP Land Cover mapping Python toolboxes ................................................................... 17 3.3 Training data .................................................................................................. 20 3.4 Model training ................................................................................................ 21 4 Example Output and Model Evaluation .................................................................. 22 4.1 Example output .............................................................................................. 22 4.2 Accuracy assessment .................................................................................... 26 4.3 Advancements with synthetic training data for sensor agnostic mapping 30 4.4 Installation and usage of ArcGIS Pro Tools .................................................. 34 Conclusion ..................................................................................................................... 35 References .................................................................................................................... 36 Report Documentation Page-
dc.format.extent47 pages / 5.84 MB-
dc.format.mediumPDF-
dc.language.isoen_USen_US
dc.publisherGeospatial Research Laboratory (U.S.)en_US
dc.publisherEngineer Research and Development Center (U.S.)-
dc.relation.ispartofseriesTechnical Report (Engineer Research and Development Center (U.S.));no. ERDC/GRL TR-21-7-
dc.rightsApproved for Public Release; Distribution is Unlimited-
dc.sourceThis Digital Resource was created in Microsoft Word and Adobe Acrobat-
dc.subjectLand cover--Remote sensingen_US
dc.subjectMachine learningen_US
dc.subjectPython (Computer program language)en_US
dc.subjectArcGISen_US
dc.subjectGeographic information systemsen_US
dc.titleSemi-automated land cover mapping using an ensemble of support vector machines with moderate resolution imagery integrated into a custom decision support toolen_US
dc.typeReporten_US
Appears in Collections:Technical Report

Files in This Item:
File Description SizeFormat 
ERDC-GRL TR-21-7.pdf5.84 MBAdobe PDFThumbnail
View/Open
Show simple item record