screen coordinates of edge detection in real time

[iZACHdx]

Hello,

Things might be a little clearer if you change the stacked sequence structure to a flat one by right clicking and then selecting Replace»Replace with Flat Sequence. The tick count is obtained in order to determine the start time. This is then compared with the time it takes to display the image and then it waits the additional amount of time in order to achieve a particular frame rate. You might want to try slowing this loop down by making the wait a constant value in order to make sure everything is getting processed. It is possible that the processing demands of your algorithm cannot be reliably done every iteration of the loop. This will be something you need to investigate further and will also depend on your processing power of the computer you are using. One other thing would be to try turning off debugging to see if that will give you some better processing times. You can do this by clicking File»VI Properties»Execution and then unchecking Allow Debugging.

-Zach

[Sniepsy]

HI Zach,

Thanks for the ideas.  I think I have prooved concept so the optomisation is not a big worry, as long as I know it can be done.  Becasue the feed will eventually become a live feed rather than an avi I wanted to try and understand how close to "real time" I can get my code based on the image processing and the feed.  Eventually I would get rid of any unnecessary video displays on the front screen and a dedicated computer setup to run the program.  I'm sure with all of this it will be possible to get the program to run fast enough to drive a simple control system.

Thanks!

Wes 

[Vishwarath1990]

Hi iZachdx

I am working on a similar problem. I am trying to get the images processing done after getting them from the video. However I am getting an error as the image is probably getting overwritten. Any suggestions please?

https://forums.ni.com/t5/Machine-Vision/Using-Vision-Assistant-calibration-for-post-processing/m-p/3...