I have to explain in details what my image processing and analysis is doing. My problem is that I can not rerpoduce the MathLookup Table(LUT) Power 1/X. The equivalent LUT Power X seemed to be cleared out for me.
I have found that the function that relates pixel values (x) with the new pixel values (y) is, for LUT power X, y = x^n, where n is the chosen exponent. It need to be normalised into the 1-to-255 scale (8-bit image), so the final LUT power X fonction is y = x^n * NF, where NF is the normalisation factor
In order to obtain the look up table that Mathlookup uses, I have used IMAQ array to image to create an image with one line and 255 pixels with values varying from 1 to 255 values.The vi applied the MathLookup table and, again, transformed it in number (IMAQ Image to array).
Comparing the LUT power X function y = x^n * NF with the values found using the vi is OK (see excel file). However, I could not be able to do the same for the LUT power 1/X. The best approximation was y = x^(1/n) but the curves are not the same.
I've alredy read all the concepts_manual of VISION and it's not weel explained the function for the LUT. Also, I've tried internet and nothing, even here, in this forum, I did not find this information.
Can you help me to understand what the MathLookup Power 1/X does ? Also, can you verify if my idea for Mathlookup Power X is correct ?
Thanks a lot
Attached: the comparision between the values calculated by Labview and by myself for different "n"(one sheet for Power X and another one for Power 1/X). Also the vi (Labview 8.2) to apply the mathlookup into an array.
PS. I speak also french and portuguese if you prefer.
Here are two documents that could be interesting for you and don't hesitate to look at the NI-IMAQ Function Reference Manual and the NI-IMAQ User Manual too :
I hope this helps,
Charlotte F. | CLAD
National Instruments France