This paper presents an approach to the registration of individual images to
one another to produce a larger composite mosaic. The approach is based on
the use of the moments of Zernike orthogonal polynomials to compute the re
lative scale, rotation and translation between the images. A preliminary st
age involves the use of an attention-like operation to estimate potential a
pproximate correspondence points between the images based on extrema of loc
al edge element density. Experimental results illustrate that the technique
is effective in a range of environments and over a broad range of image re
gistration parameters. In particular, our method makes few assumptions rega
rding the image content and yet, unlike several alternative approaches, can
perform registration for images with only a limited amount of overlap.