Vision Assistant/Lab View - Changing bit depth

[John242424]

I need to detect very small differences in luminance. With Vision Assistant or Vision Builder the machine vision functions only output bit-depth values in 8 bit resolution (0-255) I have cameras with 10 & 12 bit depths and I might require even higher. If I create a VI from a Vision Assistant script, is there a way to change the bit depth to one of these higher resolutions once I open the VI in LabView?

Message Edited by John242424 on 08-29-2007 07:12 PM

[DOC-doc]

Most of the time the camera is noisy and in this case, having more bits will not reduce your noise level.

[McCarty]

Hello John242424,

The Image Type input on the IMAQ Create.vi allows you to specify the buffer size that will be created in memory for your image.  

For most Machine Vision VIs, the bit-depth of the output image will be the same as the bit-depth of the input image.  If you would like to know the specifics of which Machine Vision VIs support which image types, I encourage you to look at the Supported Image Types document in the NI Vision for LabVIEW Help.  You can access this document in LabVIEW by using the following procedure:

First, right-click on the Block Diagram to bring up the Functions Palette.
Second, navigate to the Machine Vision Palette, right-click on a VI, and select Help.
Third, click on the Supported Image Types document that appears after 'Using Help' in the Contents window.

If this response does not fully address your question, please inform me of the specific Machine Vision VI you are experiencing difficulty with.

Best Regards,
Travis M.

[John242424]

Thanks Travis.

   My problem stems mostly from my lack of experience with LabView. While I've been working with NI Vision for quite some time but, I've done so via straight C++ dll calls. I've only recently started using LabView. That being said, I figured out easy enough how to add and set an Image Property to the IMAQ create VI. I did so and set it to a constant 16-bit grayscale. But, when I run my VI, it still spits out a 32-bit color image?!?!?! Next I tried to extract the luminance plane but that only spits out 8-bit grayscale.

I've seen some discussions concerning image conversions and am looking into those threads but in the interim I thought I would throw out one of my own that gets right to my problem.

So, again I assume there is a way to do what I want I just need to dig a little deeper. Any other insight you might be able to add would be welcome. And, as I said, I will continue to look through the other threads to try to find a solution.

[McCarty]

Hello John242424,

If your camera is configured to acquire 32-bit color images, then you should expect to extract the luminance plane as an 8-bit grayscale image.  In order to extract the luminance plane as a 16-bit grayscale image, you would need to have your camera configured to acquire 64-bit color images.  Please refer to KBs 25TBKAOA and 3QPG40QT for additional information.

If you are only trying to extract the luminance plane because you need to use the Machine Vision VIs on a black and white image, check to see if you can just change some settings on your camera.  If you can configure your camera to acquire 16-bit grayscale images, then you can avoid the need to extract the luminance plane.  Sometimes this can be done by changing the video mode on a firewire camera or the camera attributes on a CameraLink camera.

If this still does not answer your question, please let me know the manufacturer and model number of the camera you are using.

Best Regards,
Travis M.