Carbonate Feature Extraction from Hyperspectral Image Cube

This Python script extracts and visualises carbonate features from a hyperspectral image cube using Automated Feature Extraction (AFE), as described by Murphy et al. (2017).

Installation

Install the required dependencies using:

pip install -r requirements.txt

Usage

To run the script on the example data located in the data/ directory:

python ImagePlotAFE.py

The example data located in the data/ directory was acquired from a coral sample provided by H. V. McGregor. See McGregor and Gagan (2003) for more information.

Output

The script will generate visualisations of extracted carbonate features, which will be saved as an image file and displayed interactively.

Reference

McGregor, H. V., and M. K. Gagan (2003), Diagenesis and geochemistry of porites corals from Papua New Guinea: Implications for paleoclimate reconstruction, Geochim. Cosmochim. Acta, 67(12), 2147–2156.

Murphy, R. J, Webster, J.M., Nothdurft, L., Dechnik, B, McGregor, H. V., Patterson, M. A., Sanborn, K, Webb, G. E., Kearney, L. I., Rintoul, L., Erler, D. V., 2017. High resolution hyperspectral imaging of diagenesis and clays in fossil coral reef material: a non-destructive tool for improving environmental and climate reconstructions, Geochem. Geophys. Geosyst.18, doi:10.1002/2017GC006949.