Verification of handwritten signature images by multiresolution wavelet analysis

We took a multi-resolution approach to the signature verification problem. The top-level representation of signature images was the global geometric features including the size and slant angle. A multiresolution representation of signature images was obtained using the wavelet transformation. We evaluated the feature representation at different scales and developed schemes to use this multi-resolution representation for signature verification. We built a vector quantization classifier and a neural network classifier to demonstrate the advantages of using the multiresolution approach. Reasonably high verification rates were obtained.<
>