Issue No. 06 - June (2005 vol. 27)
E. Rahtu , Dept. of Electr. & Inf. Eng., Oulu Univ., Finland
This paper presents a new affine invariant image transform called multiscale autoconvolution (MSA). The proposed transform is based on a probabilistic interpretation of the image function. The method is directly applicable to isolated objects and does not require extraction of boundaries or interest points, and the computational load is significantly reduced using the fast Fourier transform. The transform values can be used as descriptors for affine invariant pattern classification and, in this article, we illustrate their performance in various object classification tasks. As shown by a comparison with other affine invariant techniques, the new method appears to be suitable for problems where image distortions can be approximated with affine transformations.
Pattern recognition, Object recognition, Pattern classification, Fourier transforms, Wavelet transforms, Multiresolution analysis, Fast Fourier transforms, Books, Image converters, TV
E. Rahtu, M. Salo and J. Heikkila, "Affine invariant pattern recognition using multiscale autoconvolution," in IEEE Transactions on Pattern Analysis & Machine Intelligence, vol. 27, no. 6, pp. 908-918, .