AI Blocks › Build and ship software
Spectral image analyst
Give it a hyperspectral or satellite image. Get it cleaned, unmixed and checked for targets or anomalies.
Spectral image analyst is a free AI skill for the job "Build and ship software". Give the instructions to an AI assistant that supports the required files and tools, then review its output.
Free · No account · BSD 2 Clause licence · Made by MathWorks · Level: expert · Thinking: high
Made by MathWorks. See the original on GitHub; we copied it as it is, with its licence.
What it does
Cleaned hyperspectral or satellite images, with materials separated and targets or odd spots flagged. The skill writes MATLAB code for the full path. It reads files in formats like ENVI, GeoTIFF, Sentinel-2 and Landsat. Then it applies calibration, atmospheric correction, denoising and sharpening, and cuts bands down with PCA. From there it finds pure materials, unmixes each pixel into their shares, matches known spectra, and flags anomalies. It also computes indices like NDVI, segments scenes, and labels data for deep learning. It needs desktop MATLAB and the Hyperspectral Imaging Library add-on.
You give it
- A hyperspectral or multispectral image file
- The material or feature you want to find
- Desktop MATLAB with the Hyperspectral Imaging Library
- A reference spectral library (optional)
You get
- MATLAB code for each processing step
- Corrected and denoised image data
- Abundance maps for each material
- Target or anomaly detections
- Spectral index maps
Use it when
- You have a satellite scene and need to map vegetation or water.
- You want to find a target material hidden in a hyperspectral image.
- Your image data is noisy and needs correcting before analysis.
- You are preparing labelled data to train a classifier.
How to use it
- Add it to your AI once: download it and add the folder to Claude, ChatGPT or Codex as a skill.
- Ask for the job in your own words. Your AI opens the skill when it fits.
- It works best when your AI can run MATLAB with the Hyperspectral Imaging Library.