Technical Note—Least-Squares Image Registration

Two-dimensional image registration estimates how one set of k 2-dimensional points is rotated, scaled, and translated into a second set of 2-dimensional points. We give an exact algorithm to solve the “least-squares” formulation of the problem. On average, the algorithm requires k² pivots; it can be viewed as a refinement of the O(k³) approximation method proposed by K. Zikan and T. M. Silberberg. The approach can be extended to the registration of images of different cardinalities.

INFORMS Journal on Computing, ISSN 1091-9856, was published as ORSA Journal on Computing from 1989 to 1995 under ISSN 0899-1499.