Support Vector Machines for Classification and Regression (provided by HU Berlin)
ImageSVM is an IDL based tool for the support vector machine (SVM) classification and regression analysis of remote sensing image data. Its workflow allows a flexible and transparent use of the support vector concept for both simple and advanced classification/regression approaches. The goal of imageSVM is to advance the use of the support vector concept in the field of remote sensing image analysis.
Random Forests for Classification and Regression (provided by Uni Bonn and HU Berlin)
imageRF is an IDL based tool for the supervised classification and regression analysis of remote sensing image data. It implements the machine learning approach of Random Forests™ (RF) (Breiman, L & Cutler, A, 2011) that uses multiple self-learning decision trees to parameterize models and use them for estimating categorical or continuous variables.
Partial Least Squares Regression (provided by Uni Bonn)
The aim of the autoPLSR is to provide a software tool with automatic feature and latent variable selection for the Partial Least Squares Regression (PLSR: Wold et al. (2001)), a multivariate regression method that is widely used in chemometrics, hyperspectral remote sensing, bioinformatics and other fields.
Spectral Index Data Mining Tool (provided by Uni Trier)
SpInMine (Spectral Index Data Mining Tool) is a tool for finding the optimal index of two narrow bands for a regression problem.
Advanced Statistical Evaluator (ASE) (provided by LMU München)
The objective of the ASE is to provide for remote sensing practitioners (i.e., non-statisticians) guidance for model evaluation. An optimal set of statistical measures is proposed for the quantitative assessment of model performance in the context of vegetation biophysical variable retrieval from Earth observation (EO) data.
Analyze Spectral Integral (provided by LMU München)
The Analyze Spectral Integral (ASI) is based on the concept of continuum removal, an approach commonly applied in the chemical sciences for the determination of mixture component concentrations. This approach has been developed as an alternative to simple Vegetation Indices.
Agricultural Vegetation Indices (AVI) (provided by LMU München)
The module AVI is a collection of 66 Vegetation Indices (VI) selected from an extensive literature survey.
imageMath (provided by HU Berlin)
Calculator for spatial and spectral image processing functionality with look-and-feel of a common hand-held calculator.
synthMix-SVR (provided by HU Berlin)
synthMix-SVR is a tool for quantitative analysis of remote sensing data. It implements the concept of combining support vector regression (SVR) with synthetically mixed training data for mapping sub-pixel fractions of land cover (Okujeni et al. 2013).
Maximum Entropy Analysis (provided by Uni Bonn)
The MaxEnt-Wrapper is an IDL based wrapper for the Java based program MaxEnt written by Steven Phillips, Miro Dudik and Rob Schapire for the maximum entropy analysis of remote sensing image data. The MaxEnt-Wrapper provides users of the EnMap-Box/ENVI to use remote sensing data within MaxEnt without converting the data itself.
Import Vector Machines for Classification (provided by Uni Bonn)
imageIVM is a tool for the import vector machine (IVM) classification of remote sensing image data. IVM is a sparse realization of kernel logistic regression and was originally proposed by Zhu and Hastie in 2003. The algorithm was modified by Roscher et al. in 2012.
Phytobenthos Index (provided by Helmholtz-Zentrum Geesthacht)
The Phytobenthos Index is a modified version of the NDVI. It is specifically designed for benthic diatoms, a dominant species on estuarine intertidal flats. The index uses the reflectance at 635 nm associated with chlorophyll c, a diagnostic absorption feature for this pigment.
Spectral Mixture Analysis (provided by Uni Trier)
The iterativeSMA algorithm is based on the principle that endmember abundances predicted by linear spectral mixture analysis are most accurate when only those endmembers are used which are actually needed to describe the spectral properties of a pixel. Large abundance errors may occur when either too few or too many endmembers are used. iterativeSMA aims to remove unnecessary endmembers.