SAFL Work Summary Index

We summarize the work to date on this page. Last updated 3/25/03. The log of the work is here.

New:

I have added the PC executables to the web site. They are listed here.
The problem with mutes is resolved. The places where the AGC images are missing data are the same places that the original data is missing. We are working on a resolution.
The hue shift "around the tenth" image is also resolved. The data had a gap, from slice 0_62 to slice 0_70. That gap, coupled with the gentle dispersal of the blue dye made a gap in the slices look like a hue shift.
Finally, I have rebuilt the AGC movie at W50P.wmv. The movie now has titles that show the various AGC levels used. The AGC levels are documented in AgcRgbHsi in the section HSI Results. The final output, listed at here uses a weight of 50 and a percentage of 50.

Old


I have added a paper on the use of FFT to enhance images. It is listed here: html and pdf.
We have processed a full section using AGC on the I (intensity) channel, and the results look good!

A test AGC Section


Here is a test slice before and after the agc:
Bright Agc on the I (intensity) channel, window 50 percentage 50:
Download the tif version.
Dark Agc on the I (intensity) channel, window 50 percentage 25:
Download the tif version.
Original
Download the tif version.
Rgb Agc
Download the tif version.
Mask of RgbAgc: We used the Rgb Agc to make a mask that could be used for correction if desired:
Download the tif version.
Click on them to get a larger picture.
Here is a movie of the test plot with the various percentages, viewable on a PC. The movie will take about 30 seconds to load on a broadband connection.
To compare, here are a before and after movie of the first 20 slices. I note that the lateral variations are fixed, but there is still a hue shift around the tenth slice that needs fixing:
Before and After AGC of W 50, P 50 on the I channel.
For a list of all the files, go to here.

The method is documented here. There also is a PDF version of the document here.
The original data is at ppmList. The section AGC stuff is at the bottem.
To see all the test sections and/or to download them for viewing, see the PpmList.
Click HERE for recent updates. For documentation of the software, look here.
We have added a new plot at the end of the above page. The agc routine is now debugged. We have applied it to the data.
Here is the original test slice:
(Click on the images to enlarge them.)
Original Image
Here is the slice with annotation describing the problem:

You can see that most of the variation is in the Red and Green channels, not the Blue channels.

Here is the plots of each color channel.


As you can see, most of the lateral noise is in the red channel. That would make sense, because if the noise was caused by a varying amount of water, the most change would be in the red channel.
Here is a movie you might like of the first dozen slices. You need to be on a PC to play it. It is 2.5Mbytes in length. Click here.

Histograms in the RGB, LHS and CMY Domain


Now that we have diagnosed the question of the "pseudo zero" in the color channels, we can go back to the AGC code.
For documentation of the software, look here.
For a log of the work so far, see HERE.

Libraries

The work done here depends on a set of C++ libraries, which are documentes here.
Send suggestions to:
Wm P. Kamp