Hi again!
I did the hough implementation in C as a dll. Works fast and good i added it to labview by call library function. But still it is not exactly what i want.
The problem is that my images are of different brightness, and only when i threshold them appropriately does the transform work correctly. Otherwise i cant get any useful info from it because most of the points 'vote' for lines passing through them.
I want in my VI not to enter the threshold. I cant also calculate it as % of overall brightness because i can loose the scratch (line) i am looking for.
Do you have any idea how to solve that problem?
For me it seems impossible without supplying the threshold, or improving hardware so that the images acquired are all of similar
brightness and the same threshold can be applied to all.
Thanks,
hope somebody can give some hints, but it looks hard.
Andru (CZ)
