bikekowal Posted September 2, 2012 Report Posted September 2, 2012 Hi, I am wondering if anyone did or found a performance test of IMAQ functions. What I am interesting in is what is the performance of IMAQ functions compared to functions written in .net or C++ (OpenCV)? Can we improve or deteriorate the performance of IMAQ based applications? regards, Marcin Quote
Jordan Kuehn Posted September 3, 2012 Report Posted September 3, 2012 (edited) In the Vision Assistant there is a 'Performance Meter' that will give you some timing information regarding each step in the process. That should at least get you started. If you pursue this, I would love to know what you come up with. http://i.imgur.com/R5DN1.png Edited September 3, 2012 by Jordan Kuehn Quote
JamesMc86 Posted September 3, 2012 Report Posted September 3, 2012 Hi Marcin, The only reference I can find to specific image processing benchmarks online are in http://www.ni.com/white-paper/2957/en#toc5. Unfortunately I have no more details than on there. I do believe a lot of effort goes into the performance of the library for example the algorithms are multicore optimised. If there are specific benchmarks you need post it here and I'm sure we could whip something up. James Quote
crelf Posted September 5, 2012 Report Posted September 5, 2012 It's difficult to compare, as we don't know exactly what's going on under the hood, so we might be comparing apples to oranges without knowing it (eg: a c++ function might do a poorer job of finding an edge, but the IMAQ one might be slower). I guess I need to know where you're coming from to better explore your question. Are you trying to decide on whether to use the Vision toolkit or something else based solely on execution speed? What's your target? PS: The functions that are called in the Vision toolkit are invariably DLLs anyway, so perhaps the comparison is moot? Quote
Wouter Posted September 5, 2012 Report Posted September 5, 2012 (edited) Maybe this may interest the topic starter, http://halide-lang.org/. Paper: http://people.csail.mit.edu/jrk/halide12/halide12.pdf Edited September 5, 2012 by Wouter Quote
JamesMc86 Posted September 5, 2012 Report Posted September 5, 2012 I agree, the comparison between LabVIEW and C++ is moot. You can write bad code in either. It really depends on the implementation of the algorithms and that is likely to vary between algorithm. I know a lot of effort is put into the performance of our Vision libraries (it is one of the few modules support in 64 bit to deal with large data better). If you have specific algorithms to use or benchmarks your interested in it is best to compare these directly. Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.