Parallel tasks within Vision Assistant Scripting

[MikeEd]

Hello

I'm new to using NI vision assistant, I think I may be missing an obvious interface tool but I'm can not seem to find it all.

I need to have parallel tasks within a script for to identify a certain area.Basically I want to reuse a intermediate processing result a number of times before bring the lines back together for the answer.

ie to have Task 3 then Task4,5 in different branch.

Task 1 ->  Task 2  -> Task 3 -> Task 6 -> Answer

Task 1 ->  Task 2  -> Task 4 -> Task 5 ->Task 6 -> Answer

[BCho]

Hey MikeEd,

It is a little unclear what exactly you are trying to do. Are you working in the vision assistant from within LabVIEW or the stand alone vision assistant? When you say your are interested in having parallel tasks, are you saying that you want to analyze a single image in two different ways, and hence yield two results for the same image? If you are working with the vision assistant in LabVIEW, you could accomplish what I think you are trying to accomplish in with three separate vision assistant steps: One for tasks 1 and 2, One for tasks 3 and 6, and then one for 4,5, and 6. You could simply wire the output of your first vision assistant to the input of the other two.

Perhaps a little more explanation on what it is you are trying to do would be useful. Also, knowing if you are working in LabVIEW or just the vision assistant would be helpful in knowing what you are working with and what your options are. Versions of this software would also be helpful.

Hope this helps,

-Ben

Message Edited by BCho on 05-13-2009 02:28 PM

[MikeEd]

Hello Ben

I have both Labview and VIsion Assistant at version  8.5.

I am trying to use Vision Assistant as a stand alone application and from there generate the C code for a application.  What I need is to be able to wire the outputs like in Labview.  I could use labview but have always hated the development enviroment(it gives me RSI in the wrist and too hardcoded for my liking) and just preffered to code.

With the two branches the result of each needs to come back to a single task that preforms a logical operation comparing the two images.  

[BCho · Accepted solution]

Hey Mike,

One way you could go about your application is to use the image to buffer step. The basic idea of this method would be as follows:

1.)Acquire your image

2.)Perform your first few steps

3.)Store your image to a buffer

4.)Perform the next few steps on the image

5.) Perform the logical comparison between your current image and the image contained in the buffer.

This method isn't necessarily parallel processing, but would be one way to go about accomplishing what it sounds like you are trying to do in NI Vision Assistant. Check out the attached example for a little more clarification on the method I proposed.

Hope this helps,

-Ben

(The example script is in vision assistant 3.6. So also attached is a screen shot of the steps. Also attached is the sample image used for the script)

[MikeEd]

Awesome, I can do what I want now.

Thanks heaps

 Mike

[dinesh282]

i want to see the both original image and processed mage in the front panel can anybody help me

[muks]

Just define a seperate buffer for original image and the processed image. (Connect imaq create to img dst).

Say you want to view jpg image. Connect an imaq read and connect to imaq create(RGB 32 bits). Then if you want to extract the luminance plane then connect a seperate imaq create(Grey scale). Now you can view both of them.

Why cant you start a new thread?