An image processing device comprises: image display means which displays at least one of transformation target images containing an object of interest; first reference point receiving/determining means which receives information on first reference point candidates according to a user operation, receives according to a user operation a determination signal targeted at the first reference point candidates displayed on the transformation target image based on the information on the first reference point candidates, and determines first reference points based on the information on the first reference point candidates targeted by the received determination signal; second reference point receiving/determining means which determines second reference points by receiving information on the second reference points; and geometric transformation means which outputs a transformed image by conducting geometric transformation to the transformation target image based on the first reference points determined by the first reference point receiving/determining means and the second reference points determined by the second reference point receiving/determining means.