ArcGIS Blog

Imagery & Remote Sensing

ArcGIS Pro

ArcGIS Pro 3.7: New Hyperspectral Imagery Tools for GIS and Image Analysis

By Hong Xu and Shengan Zhan and Jeff Liedtke

Advances in remote sensing have enabled global, multi-decadal observation of the Earth. Today, that capability extends further into the spectral dimension with hyperspectral imagery, which captures detailed material characteristics through their spectral signatures across dense, continuous spectral bands.  For GIS and image analysts, this means more precise material identification, improved classification, and deeper insight into land surface processes.

Building on the major hyperspectral release in 2025, ArcGIS Pro 3.7 introduces a set of enhancements designed to make these capabilities more accessible and efficient in day-to-day workflows. From easier data access to improved spectral analysis tools, this release focuses on helping you move more quickly from raw data to meaningful results.

Explore Spectral Features More Effectively

Continuum Removal in the Spectral Signature Viewer

Understanding subtle spectral characteristics is critical for tasks like mineral identification or vegetation analysis. The Spectral Signature Viewer now includes continuum removal, a technique that normalizes spectra and highlights absorption features.

Previously available only within geoprocessing tools, this capability is now part of your interactive analysis workflow.

continuum removal tool

In the Spectral Signature Pane, highlight the material and click the Remove Continuum tool, the spectral signature with the continuum removed will be added to the pane and viewer. In practice, this allows you to quickly inspect and analyze the absorption features.

continuum removed signature

Calculate Mean Spectra

When working with multiple samples of the same material or land cover type, generating a representative signature is often necessary. The new Calculate Mean tool lets you quickly average selected spectra and add the result back into your analysis.

Calculate mean signature

To use, highlight the signatures and click the Calculate Mean tool, where the calculated signature will be added to the viewer and pane. This is especially useful for creating cleaner inputs for classification, spectral unmixing, or target detection workflows.

Mean signatures

Reduce Data Complexity Without Losing Insight

Hyperspectral datasets are information-rich—but also computationally demanding. ArcGIS Pro 3.7 introduces the Reduce Spectral Bands tool to help you manage this complexity. It supports two commonly used approaches:

  • PCA (Principal Component Analysis) for general dimensionality reduction
  • MNF (Minimum Noise Fraction) for noise reduction and improved signal separation
MNF and PCA tool

A typical workflow that involves reducing a high dimensional dataset is classification. After applying MNF or PCA, the resulting components can be used as input to machine learning or deep learning-based classification workflows.

AVIRIS hyperspectral image classification

Derive Spectral Signatures Directly from Imagery

Spectral signatures used in analysis may come from libraries such as the USGS spectral library or they can be collected in the field.

The new Extract Spectra From Image tool allows you to generate spectral libraries directly from your imagery using field observations. You can input point or polygon feature classes repressing known materials- such as tree species or land cover types and extract their corresponding spectral signatures.

In a vegetation analysis scenario, field points representing trees, shrubs, and grassland are used to extract class-specific spectra.

calculated signatures

These signatures can then feed directly into tools like Linear Spectral Unmixing to generate abundance maps for each vegetation type. This creates a seamless workflow from field data to analytical output.

unmixing result

Simplified Spectral Index Calculation with Wavelengths

Spectral indices are widely used in analysis but applying them to hyperspectral data can be challenging due to inconsistent band naming across sensors.

ArcGIS Pro 3.7 addresses this challenge with wavelength-enabled expressions in the Band Arithmetic raster function.  For example, the Plant Senescence Reflectance Index (PSRI): (n678.0−n500.0)/n750.0, where n represents wavelength in nanometers, can be input directly into the Band Arithmetic function.

indices calculation

ArcGIS Pro automatically selects the closest matching bands, eliminating the need to manually identify band indices. Below is the output of PSRI.

PSRI map

Expanded Support for Hyperspectral data

New Tanager and Dragonette raster types

ArcGIS Pro 3.7 expands support for hyperspectral data with new raster types for Tanager and Dragonette imagery from Planet and Wyvern data providers.

For analysts, this means less time spent on data preparation. You can directly load a hyperspectral image product and the associated products from a single HDF5 file using predefined templates or build mosaic datasets to manage multiple scenes across a study area.

Hyperspectral image product

Direct Access to Hyperspectral Data STAC Catalogs

Many hyperspectral datasets are now distributed through STAC catalogs. With ArcGIS Pro 3.7, you can directly connect to Tanager and Wyvern STAC catalogs and start working with the data immediately. This makes it easier to discover datasets and quickly evaluate spectral signatures.

access HSI STAC catalog

Summary

Hyperspectral imagery provides an unprecedented level of detail about the Earth’s surface. With ArcGIS Pro 3.7, new tools and enhancements make it easier than ever to extract meaningful information and turn complex spectral data into actionable insights.

We’d love to hear how you’re using these new capabilities in your workflows.

Share this article

Leave a Reply