System and method for check fraud detection using signature validation
Systems and methods are provided for validating the authenticity of a signature on a document by providing a document from an account, the document including an actual signature and a machine-readable identifier, wherein the machine-readable identifier contains a string of data representing the integral characteristics of all valid account signatures and a person-specific confidence threshold. When the document is presented at a point of presentment, the document is scanned into a document-processing machine and the actual signature is compared against all valid account signatures.
1. A computer readable medium encoded with a computer program for locating one or more signatures on a check, the program comprising:
machine readable instructions for receiving an original or preprocessed check bitmap;
machine readable instructions for dividing the check bitmap into fixed-size tiles;
machine readable instructions for sorting the tiles into predefined classes;
machine readable instructions for identifying handwriting tiles;
machine readable instructions for adjusting the signature positions using a connected components analysis; and
machine readable instructions for generating final signature locations.
2. The computer readable medium encoded with a computer program of claim 1 , wherein the machine readable instructions for sorting the tiles into predefined classes include machine readable instructions for applying neural network classifiers to the tiles.
3. The computer readable medium encoded with a computer program of claim 1 , wherein the predefined classes include printed text, cursive text, handwritten text, graphics, picture and icon.
4. The computer readable medium encoded with a computer program of claim 1 , further comprising machine readable instructions for calculating a confidence score corresponding to the classification of each tile.
5. The computer readable medium encoded with a computer program of claim 1 , further comprising machine readable instructions for grouping the identified handwriting tiles using a cluster analysis technique, wherein clusters of handwriting represent potential signature locations.