Image ROI

[David_CSS]

Hi,

I'm working on a project where I need to acquire images from a IDS uEye camera and display them on screen (simple enough). I am displaying the images using a LabVIEW image control and want to allow the user to make angle/linear measurements by clicking on different regions of interest on the image.

 

I want the user to select multiple ROIs but I am not able to use the overlay or similar VIs as the end user does not want any runtime costs. Does anyone know of any free/download toolkits I can use, or know of a way to pick multiple ROIs using the free VIs/properties within LabVIEW.

 

Also I would like to define where the second ROI starts based on the first ROI etc. Any thoughts?

I can do this with the old picture controls however it is not very elegant.

Any pointers would be greatly appreciated

[Jordan Kuehn]

To clarify, are you saying that you want the functionality of the vision development system, but your user doesn't want to pay for the runtime license?

[jcarmody]

[...] the end user does not want any runtime costs. [...]

 

Maybe he's asking for something that doesn't require any processor or memory use?

[fabric]

The Vision Common Resources install is FREE and allows you to use the IMAQ image control in your application. You don't get any image processing but it is fine for loading images and displaying them with all the ROI tools.

 

I believe the only bit you need a license for is IMAQdx...

[hooovahh]

The Vision Common Resources install is FREE and allows you to use the IMAQ image control in your application. You don't get any image processing but it is fine for loading images and displaying them with all the ROI tools.

 

I believe the only bit you need a license for is IMAQdx...

Very neat I never knew that existed.  So you develop using the Vision Toolkit, and if you don't use any image processing functions, then you can deploy to PCs without needing a license of the Vision Run-Time Engine.  I would need to test to see what functions do and don't work without a license.

 

EDIT:  Looks like there is a page describing the functions that are included.

 

http://digital.ni.com/public.nsf/allkb/5476BD570E87E0BA8625797D0071263C

Edited September 12, 2013 by hooovahh

[David_CSS]

Hi guys,

 

thanks for the help. I wasn't looking for any hard processing, simply overlaying some lines and making linear/angular measurements. In the end I managed to put together something quickly using picture controls. I will check out the recommended free toolkit for future use. 

 

Cheers

[hooovahh]

Hi guys,

 

thanks for the help. I wasn't looking for any hard processing, simply overlaying some lines and making linear/angular measurements. In the end I managed to put together something quickly using picture controls. I will check out the recommended free toolkit for future use. 

 

Cheers

To be clear I don't think the toolkit is free (someone correct me if I'm wrong).  What is free is the deployment of a EXE that uses a subset of the Vision toolkit components.  So you pay for the Vision Toolkit, develop code in it, then build an EXE.  If the toolkit VIs you used, included image processing functions, then you must pay for a run-time license on each machine that runs that EXE.  If you only use the other functions of the toolkit then you can get away with making an EXE and not need to pay per seat by using what was linked to.

[David_CSS]

No I think that you are right - the problem I had was that I needed some tools that were not available in the free subset. I misunderstood the post above, however I have solved it now. Once this project is out of the way I will tidy up my library and make it available on the forum.

 

I don't know what the overhead of the picture control is? It seems to work well but like i mentioned I am not doing any heavy processing

[fabric]

To be clear I don't think the toolkit is free (someone correct me if I'm wrong).  What is free is the deployment of a EXE that uses a subset of the Vision toolkit components.  So you pay for the Vision Toolkit, develop code in it, then build an EXE.  If the toolkit VIs you used, included image processing functions, then you must pay for a run-time license on each machine that runs that EXE.  If you only use the other functions of the toolkit then you can get away with making an EXE and not need to pay per seat by using what was linked to.

Actually it is a little better than you think...

 

If you install the Vision Acquisition Software (i.e. the dev toolkit) then you don't need a license at all provided you UNCHECK the IMAQdx component... Magic!

 

Of course, this will limit you to basic Image Management functions, but if that's all you need then you you don't need to pay half a year's salary for a full Vision license 

 

A handy side effect of this is that you can acquire images using IMAQ with a 3rd party CameraLink board without any license (since camera link uses plain IMAQ, rather than IMAQdx), but if you use a GigE camera then you need IMAQdx and therefore you need a license. Something to think about if you are building a vision system and have a choice of hardware interfaces!

 

I don't know what the overhead of the picture control is? It seems to work well but like i mentioned I am not doing any heavy processing

I tested this a little while back. Let's say you have a 1024x1024 RGB image:

• The IMAQ control uses 4MB to display the image, as expected since the underlying data type is u32 (4 bytes per pixel). Pretty efficent.
  
• The picture control uses 7MB which includes 4MB to display the image plus an additional 3MB for the input data. Why 3MB for the data? Well, the picture data is smart enough to know that an RGB image only has three bytes per pixel that are worth encoding. 
  

Note that the IMAQ control doesn't have the overhead for the input data since IMAQ images are by-ref.