Decorrelation Stretch

Overview

Enhances visual separation between correlated image channels by applying a decorrelation stretch.

Current IDCubePro Implementation

For hyperspectral cubes containing more than three bands, IDCubePro processes consecutive three-band blocks using MATLAB decorrstretch.

For three-band data, the operation is applied directly to the

Purpose

• Increase visual color separation between correlated channels.
• Reveal subtle spatial variation that may be difficult to see directly.
• Improve exploratory visualization of multispectral data.

Important

Decorrelation Stretch is primarily a visualization/enhancement operation.

The transformed values should not automatically be interpreted as original quantitative spectral intensities.

The enhanced colors are mathematical transformations and do not directly represent material identity, concentration, or molecular composition.

Workflow

1. Load a hyperspectral dataset.

2. Open Filtering & Enhancement.

3. Select Decorrelation Stretch.

4. Allow processing to complete.

5. Compare enhanced spatial contrast with the original image.

6. Use Reset Filters to restore the original working cube if needed.