This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
Okay, I see what you're doing. Not sure why you didn't just do 1.3*iCol+3.5, but whatever.
You're seeing the effects of a couple of rounds and the interpolation. You get better results with INTER_LANCOSZ4. warpAffine is for images, not math.
If you want to have mathematical precision, use the function transform(). It takes a small amount of setup, but you get math accurate results. Basically, add another channel of all 0 to be the Y component.
