Using Image Operators ADD MAX AVERAGE should maintain the image calibration info

[chris_1243]

Using the image operators ADD or MAX or AVERAGE, even if both the images in the VI are calibrated, the output image is no longer calibrated.

To get around this, I have to separately copy the calibration to the output image. This is not possible from Vision Assistant (within VBAI) so I have to instead use a Run Labview VI step to re-calibrate the image. The same is also true for these VIs in Labview VDM, so I am posting this on both forums.

[Brad]

When using image operator functions on a calibrated image and a constant, the calibration info is preserved. It's only when performing the operator on two images the the calibration is lost. It's not clear which one to use and comparing if the calibration info is exactly the same is non trivial. In VBAI, you can call the Calibrate step after executing the Vision Assistant step to apply a calibration to the resulting image of the Vision Assistant step. Let me know if there is something I'm missing and provide more details about why this doesn't work.

Thanks,

Brad

[chris_1243]

You are correct and there is a workaround. It is an annoyance and the "its not clear which one to use" seems to be clear enough for most other functions that do not have this problem. I did not see any documentation that stated the calibration info would be lost either so it took me a while to find where this was happening in my code as there also is no quick way (like using probes) to determine whether an image has maintained calibration info or not. I basically would have to put a read calibration info VI all over the place when debugging to have figured this out.

[Brad]

Good points. I will file a Corrective Action Request (CAR) to update the documentation to better explain when calibration info can be lost. We discussed always using the calibration info from Image A, and this would work for functions where it doesn't matter which image is A or B (i.e. Max, or Add, or Average), but this wouldn't work well for functions where it does matter (i.e. Subtract, Divide, etc.), and it would also break backwards compatibility. I don't think we will implement this feature request because of these issues, but I appreciate the idea, keep them coming.

Also it might make it easier to see when the calibration info is lost if you have a seperate VI that is continually polling the image of interest and checking if the calibration info is present and this way you don't need to modify your existing code and can know as soon as the calibration info is lost.

Thanks,

Brad