pixel

[unisol]

Hi, I have a problem. I have 4 display of 320 columns by 240 rows used to form a larger one. Each display is controlled by the driver via a USB interface to display the images obtained by software belonging. Now suppose you want to view an image of 640x480, it means that the image should be divided into the 4 parts, each of which must then be sent to one of the 4 display that view completely. Is possible via a labview program that I split it in 4 the original image and that the program automatically recognizes the display to which each of these sub-images to be sent? Of course the 4 parts of the total should reach about 4 displays at once.

thanks

m

[PJS]

Dividing the image - YES
Insuring that they reach all the displays simultaneously -  USB/OS does not guarantee latency.

[PJS]

Duplicate post

[simply_me]

Hi unisol,

Basically there are two approaches to split an image (that I can think of now):

The first is using IMAQ where you get the dimensions and redefine an area of interest (see attached vi, if you have IMAQ package).

The second way is converting the image to an array and then it's just a matter of splitting the array (which I think you know how to do) but it's using IMAQ too.

See the following link which is a thread I've started when I did something similar:

http://forums.ni.com/t5/Machine-Vision/Split-Image-amp-Summation-Problem/m-p/1235688

hope this helps,

[unisol]

really thank you for answering the first part of the problem. The second part consists of the software to automatically understand the correspondence between the various subparts of the image and the 4 USB ports connected to the 4 drivers

[simply_me]

Well, I can tell you that when I used it on ONE screen all the images where simultaneous (at least to the naked eye), but as the other member said I think it's a matter of hardware. Hence, there is only one way to find out, and theoretically it should work. And, as mentioned in your other post, once you divide the original image, you should have a USB interface to send the images out (probably both parts in a while loop, for example).

[unisol]

Ok, fine, but the USB port of a PC are marked with labview? drivers should also be implemented according to specific controller? you send me more examples on image processing (split, change size) products from you?

[simply_me]

In the link that I've sent you and the vi, you can do all of the split image processing (split to which ever size you want, dump part of the image), now to change the original # of pixels I think is impossible. Invest some time in learning the mechanism of those vi's. Are you going to acquire image and split or use an .avi file (or similar)? Which hardware are you using? Also, in your original post you wrote:  Each display is controlled by the driver via a USB interface to display the images obtained by software belonging. Does that mean that you already have the USB interface? The USB drivers are given by the manufacturer, typically, and NI. Have you checked your MAX (Measurement & Automation explorer)?

[unisol]

they each display has a controller with USB connection. The image for each display is sent through proprietary software. Try it out and give you new information
thanks for the help

[simply_me]

What should I try? I didn't understand your message.... Each display should be recognized in Measurements & Automation in order to be able to be sent out....