The problem of estimating the motion parameters of an object in the presenc
e of blur is considered using moment functions. Blur invariant moments are
used as normalization constraints to normalize the image of the object to a
canonical form with respect to an assumed affine motion. The crucial point
of this method is that the blur level has no influence on the canonical fo
rm. Then we obtain the motion parameters from the normalizing transformatio
n matrices. Examples of both artificially deformed and real images are empl
oyed to demonstrate the effectiveness of the proposed scheme. (C) 2000 Publ
ished by Elsevier Science B.V. All rights reserved.