| Use IDLanROI or not [message #84418] |
Mon, 03 June 2013 06:48 |
Helder Marchetto
Messages: 520
Registered: November 2011
|
Senior Member |
|
|
Hi,
I'm confronted with a philosophical question that has probably been answered in this group already, but I wanted to check on it again.
I have my own piece of software that shows images, manages the scaling, the cross-sections and integration over stacks.
To analyze my images I have been defining my own ROIs as squares/rectangles, circles, hexagons and free hand via xROI. Now in xROI I store the objects IDLanROI, whereas for the other types I use my own structures.
I just had the illumination, that I could use IDLanROI for all of the objects above. For instance, if I want a hexagon I create the points and then manage moving the hexagon by using the translate method and for rotation and scaling I would have to replace the points after computation.
Well, here comes the question... Are there big drawbacks or advantages using IDLanROI? Are there reasons why I shouldn't be using them (I read that a bug has just been repaired in 8.2.2)?
Any general advice would be helpful before I change the whole code to accommodate only objects and not my good old structures.
The reasons why I would switch to IDLanROI objects are two:
1) The DRAW_ROI (or cgDraw_ROI) is much faster than using PLOTS (when having many points or filled regions)
2) If I switch to IDLanROI objects, the code looks nicer because I manage only one type of objects and don't have to distinguish between squares, circles, hexagons,...
Cheers,
Helder
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
