IMAQ classify VI concept needed

[aveo]

Hi,

One probable reason as to why you are getting more particles than expected is because you are sending a grayscale image through particle analysis

rather than a binary image. I had pointed this out in the previous reply only.

Just because I said put a probe and check doesnt mean you check at that single point only. You need to verify the results at every point by using probes, if required.

No solution is ready-made. So you need to try and test a few times/techniques. 

I am sorry I have not used the NI particle training interface. 

Regards

Aveo

[LearningLV]

Hello I am learning LV right now and my instructor has LV 2010 version.

I coudn't open the VI (provided in this topic discussion) with this version.

Can someone provide me screenshot of the custom classify.VI block diagram?

This way I can learn how to create a simple classification algorithm using LV.

[aveo]

Hi,

 I can send a 2010 version. but i'm unable to understand which program yor want? there are so many vis attached/mentioned in the posts. can you specify the name of the vi?

Regards

Aveo 

Edit: ok i saw you mentioned custom classify.vi. I am attaching a 2010 version of it. please check it out and let me know if it works.

[LearningLV]

Hello aveo,

Thank You very much for the old version file.

I just run the NI updates and found that we have access to LV Machine Vision runtime module 2013.

I hope this will make some positive changes.

Another question:

Can I run Vision assistant five times simultaneously?

I am afraid that ''Classifier session not closed'' error will come up if I do it.

If I can't do it, Is there any other way to work that out?

Thanks again

[aveo]

Hi,

The classifier session not closed error comes up when the classifier file has not been disposed off. It is essential to dispose the classifier file after every use. I dont understand your problem regarding vision assistant, however I can assure you that if you use classification in LabVIEW and use IMAQ dispose classifier file session then you will not be prompted with the error.

Try to create a labview version of everything that you do in vision assistant(it should be possible), and then make sure that the classifier file is disposed.

Regards

Aveo 

[LearningLV]

Thank You for the answer aveo.

[htoolatt]

---

@Klemen wrote:

Hello aveo,

I got around doing some classification testing and maybe this can help you in your project somehow.

I have followed the paper algorithm to create a feature vector:

http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=5376498&url=http%3A%2F%2Fieeexplore.ieee.org%2...

or

http://hal.archives-ouvertes.fr/docs/00/75/25/09/PDF/118.pdf

I am attaching this vi.

Then I created custom classifier using two examples of built feature vectors, trained the nearest-neighbour classifier and classified the sample (see the attached vi also).

Best regards,

K

---

---

@Klemen wrote:

Hello aveo,

I got around doing some classification testing and maybe this can help you in your project somehow.

I have followed the paper algorithm to create a feature vector:

http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=5376498&url=http%3A%2F%2Fieeexplore.ieee.org%2...

or

http://hal.archives-ouvertes.fr/docs/00/75/25/09/PDF/118.pdf

I am attaching this vi.

Then I created custom classifier using two examples of built feature vectors, trained the nearest-neighbour classifier and classified the sample (see the attached vi also).

Best regards,

K

---

hello Klemen,

i want to know how to classify animals for real time image recognition with labview and how to train classifier image .i have some problem for my project . Please reply to me.

Best regards,

HL