08-07-2018 03:04 AM
The image has different sizes. It can be very small (140*168) but also bigger (520*640). It comes from an infrared cameras so the size is quite small compare to other cameras.
I have function on my DLL which give the size of the image so if LabVIEW is able to get a pointer, and through a "MoveFunction" or something like this recreate the image, I can deal with it. LabVIEW has only to display the result so I just have to recreate the image from the pointer.
Is other side, it is the same, I will have an image read on LabVIEW and want to send it to the DLL once for now.
For 32/64bits I agree but don't have the time for this, I have to do something working as quicker I can.
08-07-2018 04:31 AM
@Mathieu33800 wrote:
The image has different sizes. It can be very small (140*168) but also bigger (520*640). It comes from an infrared cameras so the size is quite small compare to other cameras.
I have function on my DLL which give the size of the image so if LabVIEW is able to get a pointer, and through a "MoveFunction" or something like this recreate the image, I can deal with it. LabVIEW has only to display the result so I just have to recreate the image from the pointer.
Is other side, it is the same, I will have an image read on LabVIEW and want to send it to the DLL once for now.
For 32/64bits I agree but don't have the time for this, I have to do something working as quicker I can.
No disrespect to you, but "As quick as you can" may turn out to be never. If time is an issue it may be a good idea to hire someone for this aspect of the software, who has a sound understanding of both C programming and how to interface such things to LabVIEW.
08-07-2018 04:56 AM
My company doesn't think like you and me... Which was funny when I was told I has to do this work. I will try to find a way with all your explanation.
08-08-2018 01:51 AM
Just like this, do you know if transfer a matrix of double instead of uint8/16 will be in the same way or more complex? ^^
08-08-2018 02:10 AM
@Mathieu33800 wrote:
Just like this, do you know if transfer a matrix of double instead of uint8/16 will be in the same way or more complex? ^^
The only difference is the size of the array element. So yes the principle is the same. But if you want to support multiple variants you will end up writing a VI for each datatype.