Spectral Correlation

Features:

Generates comparison between any two regions of interest interactively selected on the image using spectral correlation algorithm.

Steps:

1.       Load the file. Select Spectra Mathematics and select Spectral Correlation from the dropdown menu.

2.       Click Select 1st Spectrum. Draw a region of interest. A spectrum corresponding to the average spectrum of the selected region will appear in the SPECTRAL ANALYSIS panel.

3.       Click Select 2nd Spectrum. Draw a region of interest. A spectrum corresponding to the average spectrum of the selected region will appear in the SPECTRAL ANALYSIS panel.

4.       Press Generate to measure the spectral correlation between two objects. Correlation coefficient scores can range from –1 to +1. A score of –1 indicates a perfect negative correlation, while a score of +1 indicates a perfect positive correlation. Higher scores correspond to stronger spectral correlation. A score of 0 indicates no correlation between the two selected areas.

In the shown case where two areas (purple boxes) were selected, the spectral correlation score is 0.964. For comparison, the correlation score between the roof and the grass is 0.672.

Press Clear ROI/Spectra to start another correlation analysis or any other spectral tasks.

The function is based on a spectral cross-correlation algorithm that computes the p-scores for Pearson's correlation using a Student's t distribution for a transformation of the correlation.

References:

Gibbons, J.D. Nonparametric Statistical Inference. 2nd ed. M. Dekker, 1985.

Hollander, M., and D.A. Wolfe. Nonparametric Statistical Methods. Wiley, 1973.