To allow images to be tiled together into a seamless overlay we need to know which parts of the image are part of the map and which are the other items which surround it e.g. borders, descriptions of the map etc.
This is the type of transformation which is applied to warp the image onto the map using the points you have selected:
This type of transformation is suitable for most maps and applies different amounts of warp based on the points you add.
Treats the image like a sheet of elastic. Each point you place stretches that location to the place specified and then the effect tapers off.
Requires many points to pull complex images into the correct shape but can give the best fit when a complex transformation is required.
This type of transformation is only suitable for accurately drawn maps, e.g. Ordnance Survey and performs the same amount of warp to the whole image.
Combines rotation, scaling and skew. Only 3 points are used so if you add more they will be averaged out.
Requires the least number of points and gives a good fit when the image is reasonably accurate.