Generalized affine moment invariants for object recogn

This paper introduces a new way of extracting affine in-variant features from image functions. The presented ap-proach is based on combining affine moment invariants (AMI) with multiscale invariants, in particular multiscale autoconvolution (MSA) and spatial multiscale affine invari-ants (SMA). Our approach includes all of these invariants as special cases, but also makes it possible to construct new ones. According to the performed experiments the intro-duced features provide discriminating information for affine invariant object classification, clearly outperforming stan-dard AMI, MSA, and SMA.