Contrast Maximization Toolbox

Note: The toolbox works best when the original image in the Main Interface is monochromatic. If you see issues, please close this toolbox, return to the Main Interface and select Single Channel radiobutton.

Features:

1.      Finds the channel/wavelength for the best contrast between two selected regions of interest (ROIs).

2.      Finds the channel/wavelength for the best contrast and the corresponding bandwidths between two selected regions of interest (ROIs).

3.      Identify whether there is a difference between two ROIs and quantify the difference using Bhattacharyya distance (Db). Higher Db corresponds to better contrast between two ROIs

4.      Identifies new indexes.

Steps:

1.      Load the file and select Toolboxes → Contrast Maximization. A toolbox window with instructions will pop-up.

2.      Select Mathematics and choose an option.

3.      Select the Selection Tool. Available options are:

a.      Ellipse

b.      Polygon

c.      Freehand

d.      Rectangular

4.      Select the Band Shape. Available options are:

a.      Rectangular Band Shape

b.      Gaussian Band Shape

5.      Select the method for finding the best channel(s) for contrast. Available options are:

a.      Exhaustive Search without the bandwidths

b.      Monte Carlo Search with the bandwidths

6.      Draw two regions of interest with the cursor. The calculations will start immediately.  

If the Monte Carlo is selected, then a new dialog asks to put the minimum and maximum bandwidths as well as the number of samples. (NOTE: The number of samples depends on the size of the image. A higher number will make the calculation slower.)

The results are displayed as a new image and a contrast map. The title of the image gives the wavelengths. Toolbox display once processing has been completed. Note the title of the image gives the specific max contrast values.

Additional Information:

The algorithm implies relatively homogenous intensities within the region of interest. Adjust the size of the ROI accordingly.

References:

T Du, DK Mishra, L Shmuylovich, A Yu, H Hurbon, ST Wang, MY Berezin. Hyperspectral imaging and characterization of allergic contact dermatitis in the short‐wave infrared, Journal of Biophotonics 13 (9), e202000040, 2020

DM Kim, H Zhang, H Zhou, T Du, Q Wu, TC Mockler, MY Berezin. Highly sensitive image-derived indices of water-stressed plants using hyperspectral imaging in SWIR and histogram analysis Scientific Reports 5 (1), 1-11, 2015