Data Analysis Tools#
This section helps you understand how to use WISER’s data analysis tools — the algorithms for transforming, classifying, detecting, unmixing, and visualizing hyperspectral imagery. Each page explains what the tool does, how it works, and how to run it from the interface.
Transforms & Dimensionality Reduction#
Principal Component Analysis (PCA) — reduce a cube to a few bands ordered by variance.
Minimum Noise Fraction (MNF) — reduce a cube to bands ordered by signal-to-noise ratio.
Classification & Clustering#
K-Means Clustering — group pixels into clusters by spectral similarity.
Target Detection & Spectral Matching#
Mixture-Tuned Matched Filter (MTMF) — detect a target material against an unknown background.
Spectral Angle Mapper (SAM) — match pixels to reference spectra by spectral angle.
Spectral Feature Fitting (SFF) — match pixels to reference spectra by their absorption features.
Unmixing#
Linear Unmixing — estimate per-pixel abundances of a set of endmember spectra.
Visualization#
Interactive Scatter Plot — explore two bands in feature space and link selections back to the image.