I was just placing ground control points wehere I only have 9-15 pixels to identify the center of a high contrast feature like a manhole cover from an aerial image. It strikes me that I could more accurately place the center if I was looking at a bilinear interpolation of the pixel brightness / color. Even better would be if I had buttons to auto-center based on the marker type (end/corner/symmetrical object).
Topic: Bilinear interpolation or other averaging when zoomed to pixels for gcp placemen (Read 2800 times)
