determining area based on varing clor intensity

[amrishchawla]

I am attaching an image from my experiment on diffusion of ozone gas in agar gels. The gels are spread out in a grid similar to a microarray. i have tried using the microarray plug in but it does not give me the area of the region i am measuring. only an average intensity of gray scale values in the region enclosed by the ROI.

The bottom row is the control and other three rows contain my treatment samples. due to the diffusion of ozone into the gels the dye looses color and becomes colorless. i need to measure this zone of color loss in the gel slices. i use your microarray plugin to get an average value for all the slices but it is not helping me measure this zone. is there a way to draw concentric circles using the plug in and measure only the annulus region. or some way to determine the area in a slice based on difference in the intensity of the dye.

i am a newbie at using image analysis. any help is appreciated.


http://picasaweb.google.com/achawl1/Gels?feat=directlink

Amrish

[Gabriel Landini]

On Thursday 01 April 2010, you wrote:
> I am attaching an image from my experiment on diffusion of ozone gas in
> agar gels.

It might be difficult to get reliable results from images like that.
Most obvious: lighting is not even, The background in the centre of the image
is brighter than at the boundaries.
To be able to compare across plates, you need images with a flat background.
See the wiki for several methods.

The plates seem to have different diameters (so presumably they will capture
different amounts of whatever one is measuring?

Probably the volumes of agar are different too, (some had massive bubbles or
appear to have empty parts while the others do not. So the samples are not
comparable.

And most curiously, the image has some dithering artifacts that are quite
obvious. Only 1 every 3 rgb bins have data in them.
Cheers

G.