(no subject)

Posted by Eric BADEL on Jan 10, 2008; 4:27pm
URL: http://imagej.273.s1.nabble.com/no-subject-tp3697624p3697625.html

for 1. That is a typical mechanical problem in the field of "deformation
measurements".  Several methods are possibles. Try this: simply draw a
virtual regular cell using a drawing software (a perfect round or a
perfect hexagon for example). Apply a well-known strain (uniform for
example) and then apply your process with these 2 theoretical images.
You will get an evaluation of the accuracy of your imaging process.
If you want to go further in the characterization, add reasonable random
noise to your 2 images and do the same.

If you work with experimental images :  apply your process using 2
images  that were recorded at the same state (2 consecutive images
without stress for example) => you will get an approximation of the
accuracy of your complete measurement process.

Eric

Wenhao She wrote:

> Hi, Dear ImageJ group,
>  
> Thank you for your patience in advance!
>  
> We are planning to detect the strain of the structure within the cells. We use this plugin at first. http://biocomp.cnb.uam.es/~iarganda/bUnwarpJ/  Shortly about this that this devides the source and target pictures to many cubic then tries to warp them to be same as each other. We try to use it to detect the deformation of the cell and input this coefficient into the MatLab to use the spline toolbox to know the stretch of the X , Y direction and also about stretch on one point.
>  
> Question about it:
>  
> 1.Do you guys have any opinion to test this way to know the precision of this way??
>