02-01-2008 02:16 AM
Hello Brad, thanks for your suggestion of the GigE. We already have IEEE and other scientific grade cameras as part of our system. The point of the current work is to implement a very inexpensive image acquisition paradigm so that we can use the cameras ubiquitously. USB cameras, like the Creatives, are cheap enough to allow us to use them everywhere in the system. This would not be possible with the type of cameras you suggested. In other words, I have no choice on the hardware. I need to write a software to acquire from the USB cams, and it would be significantly better if I could acquire simultaneously from many. I understand that this technology is not a priority for National Instruments, but perhaps you could direct me towards a method of modifying the libraries myself? Any help you could give would be appreciated. Thanks.
Nicholas
02-01-2008 03:23 AM - edited 02-01-2008 03:26 AM
I don't think there is anything you could do to the driver to make it work with multple cams. For that you would need to change the underlying DLL itself to support acqusition sessions much like the IMAQ Session for the NI-IMAQ and NI-IMAQdx and VIs.
@nicmatlis wrote:
Hello Brad, thanks for your suggestion of the GigE. We already have IEEE and other scientific grade cameras as part of our system. The point of the current work is to implement a very inexpensive image acquisition paradigm so that we can use the cameras ubiquitously. USB cameras, like the Creatives, are cheap enough to allow us to use them everywhere in the system. This would not be possible with the type of cameras you suggested. In other words, I have no choice on the hardware. I need to write a software to acquire from the USB cams, and it would be significantly better if I could acquire simultaneously from many. I understand that this technology is not a priority for National Instruments, but perhaps you could direct me towards a method of modifying the libraries myself? Any help you could give would be appreciated. Thanks.
Nicholas
02-02-2008 11:36 AM
02-03-2008 12:14 PM
That's probably not a possibility even with access to the C source code. Closing and opening DirectX resources is not something that is usually gonna happen in ms. There is a lot behind the scene that is going on when DirectX queries the device capabilities and decides on what filter chain to build up and that is gonna require some time. A lot of COM objects are involved in that and they all need to be initilialized and then initiliaze other depending COM objects too. And on closing that is all deallocated again. So I think that 1s or more for closing and opening a camera are not really a lot of time.
@nicmatlis wrote:
If it is not possible to acquire continuously from more than one USB camera, is it at least possible to switch quickly between cameras? At the moment the switch is taking about 2.2 seconds to implement. If this could be done much faster, it would at least be something.
02-15-2008 10:19 AM
02-15-2008 01:02 PM
Italian has already been added. Use one of the latest DLLs and it should work. Let m know if you hav any problems.
Thanks,
Brad
02-18-2008 03:48 AM
No it doesn't work! I've tried the 2 or 3 version of the last DLL but without any positive result. I have Windows XP SP2 Italian, Labview 8.0 and the camera is an analog bw camera from "Lafayette". I acquire the immage throught an USB acquisition tool named "VS-USB 2800". In the far past I succeed to obtain an acquisition but I don't remember what version of LabView or XP was. Please HELP ME
Thanks.
Orlando
02-18-2008 09:45 AM
I suspect it is your camera. A way to verify this is the problem and not the OS would be to test the camera on an English OS. Another way would be to test a Logitech or Creative webcam (about $30) on English/Italian OS since these work well with the DirectShow filter driver.
Hope this helps,
Brad
02-18-2008 11:09 AM
02-18-2008 12:49 PM