EnMAP-Box 3 Documentation

The EnMAP-Box is a python plugin for QGIS, designed to process and visualise hyperspectral remote sensing data.

Get going with the Installation and the Getting Started chapter. Have a look at the Cookbook for usage examples!

Key features

• Extend your QGIS for remote sensing image analysis

• Add powerful tools to process and analyse imaging spectroscopy data

• Integrate machine learning algorithms into your image classification and regression with Random Forests, Support Vector Machines and many more

• Create and manage spectral libraries with attribute data

• Develop your own image processing algorithms using a powerful Python API

Related websites

• Environmental Mapping and Analysis Program (EnMAP)

• Source code repository

• HYPERedu on eo-college

News

FOSSGIS 2023 Berlin

The EnMAP-Box will be present at the FOSSGIS with a demosession: Visualisierung und Analyse von Satellitenbildern mit der EnMAP-Box (https://pretalx.com/fossgis2023/talk/9WAMJ9/)

Looking forward meeting you there!

UPDATE - Here is the recording of the presentation:

EnMAP-Box Version 3.11 released

Oct. 10, 2022

A new EnMAP-Box version has been released. A list of all improvements can be found in the changelog.

EnMAP-Box Version 3.10 released

June 09, 2022

A new EnMAP-Box version has been released. A list of all improvements can be found in the changelog.

EnMAP-Box Version 3.9 released

Oct. 10, 2021

A new EnMAP-Box version has been released. A list of all improvements can be found in the changelog.

EnMAP-Box Workshop 2021

June 07, 2021

The next EnMAP-Box workshop will be held online on 21-23 June, 2021. The workshop aims to demonstrate the current state of the EnMAP-Box by way of presentations, live demos with Q&A and self-paced tutorials. Registration for up to 250 participants is now open at the workshop website.

FOSSGIS 2021

June 09, 2021

Benjamin Jakimow presented how to use Spectral Libraries in QGIS using the EnMAP-Box (german).

EnMAP-Box Version 3.7 released

Oct. 27, 2020

A new EnMAP-Box version has been released. It includes product readers for EnMAP L1B, L2C and L2A, PRISMA L2D and DESIS L2A images, and a first version of Spectral Math in the QGIS Expression Builder. A list of all improvements can be found in the changelog.

FOSSGIS 2020

Andreas Rabe presented the EnMAP-Box at the FOSSGIS 2020 in Freiburg. See the full live-demo session here (german):

QGIS User Conference 2019

Two Presentations by EnMAP-Box developers Andreas Rabe and Benjamin Jakimow at the QGIS User conference in Coruña

• Earth observation data processing in QGIS with a python API

• EO Time Series Viewer - A plugin to explore Earth Observation Time Series Data in QGIS

---

General

• About
• FAQ & Troubleshooting
  • Bug report & feedback
  • Contact
  • Citation
  • FAQ
  • Known Issues
• How to contribute
  • Submit a bug report or feature request
  • Provide source code
    • 1. Create a fork on GitHub
    • 2. Clone the forked EnMAP-Box repository
    • 3. Modify the local repository
    • 4. Commit your changes
    • 5. Push changes back to your fork
    • 6. Create a pull request
  • Documentation
  • Licensing
    • Applying License Terms
    • Contributor License Agreements (CLA)
      • Individual Contributor License Agreement (ICLA)
      • Corporate Contributor License Agreement (CCLA)
      • Submitting License Agreements
      • Developer Certificate of Origin (DCO)
  • Tests and Continuous Integration
• Roadmap
• Glossary
  • Index with all Terms
  • GIS and Remote Sensing
  • Raster Metadata
  • Machine Learning

User Section

• Installation
  • 1. Install QGIS
    • MacOS
  • 2. Install required python packages
    • 2.1 Package Installer
    • 2.2 Command-line
      • Windows
      • Linux
      • Mac
  • 3. Install or update the EnMAP-Box
    • Install from QGIS plugin repository
    • Install older version
• Installation (NEW)
  • QGIS Installation on Windows
    • Option A: Install QGIS via the official Standalone/OSGeo4W Installer
      • 1. Install QGIS
      • 2. Install Python Dependencies
    • Option B: Install QGIS via the Mambaforge Package Manager
      • 1. Install Miniforge
      • 2. Install QGIS and Python Dependencies
  • QGIS Installation on Linux (Ubuntu)
    • Option A: Install QGIS via the official repository
      • 1. Install QGIS
      • 2. Install Python Dependencies
    • Option B: Install QGIS via conda/mamba
      • 1. Install Micromamba
      • 2. Install QGIS and Python Dependencies
  • QGIS Installation on MacOS
  • EnMAP-Box Plugin Installation
• Getting Started
  • 1. Launching the EnMAP-Box
  • 2. Loading data
  • 3. First steps in the GUI
  • 4. Use a Processing Algorithm
  • 5. What’s next?
• Cookbook
  • Clip a raster with a vector
  • Clustering (kmeans)
  • Filtering
    • Spatial Filter
    • Spectral Filter
  • Generic Filter (Majority)
  • Map Algebra with ImageMath
    • Calculate NDVI
    • Mask raster with vector
  • Wavelength Information
  • Transformation (PCA)
• User Manual
  • The GUI
    • Toolbar
      • Data Sources
      • Maps and Views
      • Map Tools
      • Vector Tools
      • Earth Observation for QGIS (EO4Q)
    • Panels
      • Data Sources
      • Data Views
      • Spectral Profile Sources
      • Processing Toolbox
      • Cursor Location Values
    • Views
      • Map View
        • Linking
        • Crosshair
      • Spectral Library View
        • Collect profiles
          • Adding information
        • Advanced options
          • Create / Modify profiles with the Field Calculator
          • Spectral Processing
        • Visualization settings
          • General profile plot settings
          • Visualizing profiles
          • Colorize spectra by attribute
        • Loading, Saving and Importing / Exporting
          • Load and Save Spectral Libraries
          • Exporting Profiles
          • Importing Profiles
          • Spectral Profile JSON format
      • Text View
      • Attribute Table View
  • Tools
    • Add Product
    • Add Web Map Services (WMS)
    • Band Statistics
    • Bivariate Color Raster Renderer
    • Class Fraction/Probability Renderer and Statistics
    • Classification Statistics
    • CMYK Color Raster Renderer
    • Color Space Explorer
    • Decorrelation Stretch Renderer
    • Enhanced Multiband Color Renderer
    • HSV Color Raster Renderer
    • Image Cube
    • Metadata Viewer
    • Multisource Multiband Color Raster Renderer
    • Raster Layer Styling
    • Raster Source Band Properties Editor
    • Reclassify
    • Scatter Plot
      • Plotting Raster Band vs. Raster Band
      • Plotting Raster Band vs. Vector Field
    • Virtual Raster Builder
  • Applications
    • Agricultural Applications
    • Classification Dataset Manager
      • Create a classification dataset
      • Edit categories and features
      • Split dataset randomly
    • Classification workflow
    • Classification Workflow (advanced)
      • Quick Mapping
      • Detailed Analysis
        • Dataset
        • Classifier
        • Feature Clustering
        • Feature Ranking
        • Model
        • Classification
      • Settings
      • Log
    • Classification Workflow (deprecated)
    • EO Time Series Viewer
    • EnPT (EnMAP Processing Tool)
    • GFZ EnGeoMAP
    • Image Math (deprecated)
    • Raster math
      • Single-line expressions
      • Use raster bands
      • Use vector fields
      • Use raster/vector masks
      • Blockwise vs. monolithic processing
      • Multi-line code fragments
      • Metadata handling
      • Input raster lists
    • Regression Dataset Manager
    • Regression Workflow (deprecated)
    • Spectral Index Creator
  • Earth Observation for QGIS (EO4Q)
    • GEE Time Series Explorer
      • Getting started
    • Location Browser
    • Profile Analytics
    • Raster Band Stacking
    • Sensor Product Import
  • Example Data
    • Open
    • Layers
    • Full EnMAP Example Scene
  • Processing Algorithms
    • Auxilliary
      • Build 3D Cube
      • Create default paletted raster renderer
    • Classification
      • Class fraction layer from categorized vector layer
      • Class separability report
      • Classification layer accuracy and area report (for simple random sampling)
      • Classification layer accuracy and area report (for stratified random sampling)
      • Classification layer from class probability/fraction layer
      • Classification layer from rendered image
      • Classification workflow
      • Classifier performance report
      • Fit CatBoostClassifier
      • Fit GaussianProcessClassifier
      • Fit generic classifier
      • Fit LGBMClassifier
      • Fit LinearSVC
      • Fit LogisticRegression
      • Fit RandomForestClassifier
      • Fit SpectralAngleMapper
      • Fit SVC (polynomial kernel)
      • Fit SVC (RBF kernel)
      • Fit XGBClassifier
      • Fit XGBRFClassifier
      • Predict class probability layer
      • Predict classification layer
      • Reclassify raster layer
    • Clustering
      • Fit AffinityPropagation
      • Fit Birch
      • Fit KMeans
      • Fit MeanShift
      • Predict (unsupervised) classification layer
    • Convolution, morphology and filtering
      • Spatial convolution Airy Disk filter
      • Spatial convolution Box filter
      • Spatial convolution custom filter
      • Spatial convolution Gaussian filter
      • Spatial convolution Moffat filter
      • Spatial convolution Ricker Wavelet filter
      • Spatial convolution ring filter
      • Spatial convolution Savitsky-Golay filter
      • Spatial convolution Top-Hat filter
      • Spatial convolution Trapezoid filter
      • Spatial Gaussian Gradient Magnitude filter
      • Spatial generic filter
      • Spatial Laplace filter
      • Spatial Maximum filter
      • Spatial Median filter
      • Spatial Minimum filter
      • Spatial morphological Binary Closing filter
      • Spatial morphological Binary Dilation filter
      • Spatial morphological Binary Erosion filter
      • Spatial morphological Binary Fill Holes filter
      • Spatial morphological Binary Opening filter
      • Spatial morphological Binary Propagation filter
      • Spatial morphological Black Top-Hat filter
      • Spatial morphological Gradient filter
      • Spatial morphological Grey Dilation filter
      • Spatial morphological Grey Erosion filter
      • Spatial morphological Grey Opening filter
      • Spatial morphological Laplace filter
      • Spatial morphological White Top-Hat filter
      • Spatial Percentile filter
      • Spatial Prewitt filter
      • Spatial Sobel filter
      • Spectral convolution Box filter
      • Spectral convolution Gaussian filter
      • Spectral convolution Ricker Wavelet filter
      • Spectral convolution Savitsky-Golay filter
      • Spectral convolution Trapezoid filter
    • Dataset creation
      • Create classification dataset (from categorized raster layer and feature raster)
      • Create classification dataset (from categorized spectral library)
      • Create classification dataset (from categorized vector layer and feature raster)
      • Create classification dataset (from categorized vector layer with attribute table)
      • Create classification dataset (from JSON file)
      • Create classification dataset (from Python code)
      • Create classification dataset (from table with categories and feature fields)
      • Create classification dataset (from text files)
      • Create regression dataset (from continuous-valued raster layer and feature raster)
      • Create regression dataset (from continuous-valued spectral library)
      • Create regression dataset (from continuous-valued vector layer and feature raster)
      • Create regression dataset (from continuous-valued vector layer with attribute table)
      • Create regression dataset (from JSON file)
      • Create regression dataset (from Python code)
      • Create regression dataset (from table with target and feature fields)
      • Create regression dataset (from text files)
      • Create regression dataset (SynthMix from classification dataset)
      • Create unsupervised dataset (from feature raster)
      • Create unsupervised dataset (from JSON file)
      • Create unsupervised dataset (from Python code)
      • Create unsupervised dataset (from spectral library)
      • Create unsupervised dataset (from text file)
      • Create unsupervised dataset (from vector layer with attribute table)
      • Merge classification datasets
      • Random samples from classification dataset
      • Random samples from regression dataset
      • Select features from dataset
    • Export data
      • Save spectral library as GeoJSON file
    • Feature selection
      • Classifier feature ranking (permutation importance)
      • Hierarchical feature clustering
    • Import data
      • Import DESIS L1B product
      • Import DESIS L1C product
      • Import DESIS L2A product
      • Import EMIT L2A product
      • Import EnMAP L1B product
      • Import EnMAP L1C product
      • Import EnMAP L2A product
      • Import Landsat L2 product
      • Import PRISMA L1 product
      • Import PRISMA L2B product
      • Import PRISMA L2C product
      • Import PRISMA L2D product
      • Import Sentinel-2 L2A product
    • Masking
      • Apply mask layer to raster layer
      • Create mask raster layer
      • Layer to mask raster layer
    • Raster analysis
      • Aggregate raster layer bands
      • Convex hull and continuum-removal
      • Create spectral indices
      • Raster layer zonal aggregation
      • Raster math
      • Sample raster layer values
      • VRT band math
    • Raster conversion
      • Create RGB image from class probability/fraction layer
      • Save raster layer as
      • Translate categorized raster layer
      • Translate raster layer
      • Write ENVI header
    • Raster miscellaneous
      • Create grid
      • Edit raster source band properties
      • Stack raster layers
      • Subset raster layer bands
    • Raster projections
      • Geolocate raster layer
    • Regression
      • Fit CatBoostRegressor
      • Fit GaussianProcessRegressor
      • Fit generic regressor
      • Fit KernelRidge
      • Fit LGBMRegressor
      • Fit LinearRegression
      • Fit LinearSVR
      • Fit PLSRegression
      • Fit RandomForestRegressor
      • Fit SVR (polynomial kernel)
      • Fit SVR (RBF kernel)
      • Fit XGBRegressor
      • Fit XGBRFRegressor
      • Predict regression layer
      • Receiver operating characteristic (ROC) and detection error tradeoff (DET) curves
      • Regression layer accuracy report
      • Regression workflow
      • Regressor performance report
      • Spectral Index Optimizer
    • Spectral resampling
      • Spectral resampling (to custom sensor)
      • Spectral resampling (to DESIS)
      • Spectral resampling (to EnMAP)
      • Spectral resampling (to Landsat 4/5 TM)
      • Spectral resampling (to Landsat 7 ETM+)
      • Spectral resampling (to Landsat 8/9 OLI)
      • Spectral resampling (to PRISMA)
      • Spectral resampling (to response function library)
      • Spectral resampling (to Sentinel-2A MSI)
      • Spectral resampling (to Sentinel-2B MSI)
      • Spectral resampling (to wavelength)
      • Spectral resampling (to wavelength and FWHM)
    • Transformation
      • Fit FactorAnalysis
      • Fit FastICA
      • Fit FeatureAgglomeration
      • Fit KernelPCA
      • Fit MaxAbsScaler
      • Fit MinMaxScaler
      • Fit Normalizer
      • Fit PCA
      • Fit QuantileTransformer
      • Fit RobustScaler
      • Fit StandardScaler
      • Inverse transform raster layer
      • Transform raster layer
    • Unmixing
      • Regression-based unmixing
    • Vector conversion
      • Rasterize categorized vector layer
      • Rasterize vector layer
    • Vector creation
      • Random points from categorized raster layer
      • Random points from mask raster layer
      • Random points from raster layer value-ranges
  • Metadata Guide
    • Introduction to GDAL Data Model
    • Introduction to QGIS Data Model
    • EnMAP-Box Metadata Handling
• EnMAP Data
  • EnMAP Data Access Portal
    • Step 1: Instrument Planning Portal (IPP)
    • Step 2: EOWEB ® GeoPortal
    • Step 3: Download data from the FTP Delivery Server
• Application Tutorials
  • Regression-based unmixing of urban land cover
    • Introduction
      • 1. Background
      • 2. Content
      • 3. Requirements
      • 4. Further Reading
      • 4. Data
    • Exercise A: Urban land cover
      • 1. Start the EnMAP-Box
      • 2. Load data
      • 3. Visualize raster and vector data
      • 4. Basic navigation tools
    • Exercise B: Spectral mixing
      • 1. Multiple map views
      • 2. Visualize image spectra
    • Exercise C: Urban spectral libraries
      • 1. Labeled spectral libraries
      • 2. Spectral library handling
    • Exercise D: Regression-based unmixing
      • 1. Introduction
      • 2. Start the application
      • 3. Inputs and class selection
      • 4. Mixing parameters
      • 5. Regression Algorithm
      • 6. Outputs
      • 7. Run the application
      • 8. Visualize the urban land cover fraction map
    • Exercise E: Validation of fraction maps
      • 1. Create reference fraction map
      • 2. Statistical validation of fraction maps
    • Additional Exercises
  • Regression-based mapping of forest aboveground biomass
    • Introduction
      • 1. Background
      • 2. Content
      • 3. Requirements
      • 4. Further reading
      • 5. Data
    • Exercise A: Getting started with the EnMAP-Box
      • 1. Start the EnMAP-Box
      • 2. Load data
      • 3. Visualize raster data
      • 4. Basic navigation tools
      • 5. Multiple map views
      • 6. Visualize vector data
      • 7. Extract & visualize image spectra
    • Exercise B: Regression based mapping of AGB
      • 1. Use the Regression Dataset Manager for data preparation
      • 2. Use the Regression Workflow for estimating AGB
      • 3. Assess the model performance for AGB estimation
      • 4. Visualize AGB results
      • Learning Activities
    • Exercise C: Compare AGB estimates with the NDVI
      • 1. Introduction to ImageMath
      • 2. Calculate NDVI
      • 3. Create a forest mask
      • 4. Apply the forest mask
      • 5. Visualize AGB vs. NDVI with the Scatter Plot tool
      • Learning Activities
    • Additional Exercises
  • Ocean Colour analysis with ONNS
    • Overview
      • About
      • Background
      • References
    • Data
      • Test data
      • ONNS output
    • Usage
      • ONNS GUI
    • Credits
      • Development Lead
      • Contributors
      • Funding
  • EnGeoMAP 3.1 Tutorial
    • Introduction to EnGeoMAP 3.1
    • 1.   Data Preparation and Requirements
      • Hyperspectral Data:
      • Spectral Library Files:
      • RGB Color Scheme CSV file:
      • Example / Test Data:
    • 2. How to use/run the EnGeoMAP toolbox
    • 4. The Results
    • 5. Acknowledgements
    • 6. Further Reading
    • Literature
  • Manual Retrieval of Vegetation Variables using IVVRM
  • SIO and DASF for N Estimation
  • EnPT
• Workshop Tutorials
  • 12th EARSeL Imaging Spectroscopy Workshop - Part I: Introduction into the EnMAP-Box
    • Introduction
    • Exercise A: Visualize PRISMA data inside the EnMAP-Box
    • Exercise B: Perform a land cover classification
  • 12th EARSeL Imaging Spectroscopy Workshop - Part II: Working with Spectral Libraries

Developer Section

• Installation
  • Overview
  • 1. Have Git installed
  • 2. Clone this repository
  • 3. Setup a QGIS environment
    • a) OSGeo4W (Windows)
      • 1. Install or update packages
      • 2. Setup development environment
    • b) Linux
    • c) macOS
    • d) Docker
    • e) Setup Conda (all platforms)
    • f) Setup Mambaforge (all platforms)
  • 4. Setup PyCharm
  • Other Tools
    • Qt Assistant
    • Qt Designer
    • Qt Creator
  • OSGeo4W for Devs
    • Setup Environment
    • Debug QGIS with Visual Studio
  • References
• EnMAP-Box repository
  • Structure
    • Overview
    • Scripts for Developers
    • Generated Folders
  • Install the EnMAP-Box repository
  • Install / Update EnMAP-Box Testdata
  • Get Updates from other Repositories
• Build and publish the EnMAP-Box
  • Build the EnMAP-Box Plugin
  • Build the EnMAP-Box documentation
  • Publish the EnMAP-Box
• Dev Cookbook
  • EnMAP Cookbook
    • 1. Start the EnMAP-Box
    • 2. Data Sources
      • Add Data Sources
      • List Existing Data Sources
      • Remove Data Sources
      • Data Sources are unique
      • Receive data source changes:
    • 3. Dock Windows
    • 5. Map Tools
    • 6. Spatial Points and Extents
    • 7. Map Locations and Spectral Profiles
• Create EnMAP-Box Applications
  • 1. Initialize a new EnMAP-Box Application
  • 2. Define an Menu
  • 3. Define QgsProcessingAlgorithms for the EnMAPBoxAlgorithm Provider
  • 4. Create a Graphical User Interface
  • List of environmental variables
  • Further links and sources
    • Git for Beginners
    • PyQGIS
    • Python Code Documentation
• RFC list
  • RFC 1: Enhanced GDAL PAM for QGIS Raster Layers
    • Summary
    • Motivation
    • Approach
    • Implementation
  • RFC 2: Spectral Properties
    • Summary
    • Motivation
    • Approach
    • Implementation
  • RFC 3: Temporal Properties
    • Summary
    • Motivation
    • Approach
    • Implementation
  • RFC 4: Band Properties
    • Summary
    • Motivation
    • Approach
    • Implementation
• Documentation Guidelines
  • Writing Guidelines
    • 1. reStructuredText
      • Basic inline markup
      • Numbering
    • 2. Figures & Screenshots
      • Screenshots with explanations
      • Positioning of info boxes and images

Indices and tables

• Index

• Module Index

• Search Page