Systems and methods for biometric identification using handwriting recognition
A biometric handwriting identification system converts characters and a writing sample into mathematical graphs. The graphs comprise enough information to capture the features of handwriting that are unique to each individual. Optical character recognition (OCR) techniques can then be used to identify these features in the handwriting sample so that drafts from two different samples can be aligned to compare to determine if the features in the writing sample correlate with each other.
1. A method for identifying an originator of a handwriting specimen, comprising:
converting the handwriting specimen into an electronic image;
converting the electronic image into a mathematical graph;
detecting similarities between the mathematical graph and a plurality of stored graphs;
aligning vertices and edges of the mathematical graph with similar stored graphs;
comparing the mathematical graph with the similar stored graphs; and
identifying the originator of the handwriting specimen based on the comparison.
2. The method of claim 1 , wherein converting the handwriting specimen into an electronic image comprises, converting the handwriting specimen into a bi-tonal electronic image.
3. The method of claim 1 , wherein converting an electronic image into a mathematical graph comprises converting the electronic image into an image skeleton.
4. The method of claim 3 , wherein converting an electronic image into a mathematical graph further comprises transforming the image skeleton into one or more edges and one or more vertices.
5. The method of claim 3 , wherein converting an electronic image into a mathematical graph further comprises locating a first primary vertex.
6. The method of claim 5 wherein converting an electronic image into a mathematical graph further comprises following a pixel path from the first primary to the second primary vertex.
7. The method of claim 6 , comprising detecting what type of vertex each located vertex is, wherein the type of vertex is detected by checking a plurality of pixels around a pixel of interest.
8. The method of claim 1 , wherein converting the electronic image into the mathematical graph includes minor feature editing.
9. The method of claim 8 , wherein the minor feature editing comprises closing a small gap.
10. The method of claim 8 , wherein the minor feature editing comprises connecting undershoots.
11. The method of claim 8 , wherein the minor feature editing comprises combining a plurality of close vertices.
12. The method of claim 1 , wherein the comparing step includes the additional steps of:
associating contour information with the mathematical graph and the similar stored graphs; and
comparing the contour information associated with the mathematical graph and the similar stored graphs.
13. The method of claim 1 , further comprising transforming a mathematical graph into at least one feature caddie.
14. The method of claim 13 , wherein the feature caddie comprises a topology.
15. The method of claim 14 , wherein the topology comprises a topology for an individual character.
16. The method of claim 14 , wherein the topology comprises a topology for a group of characters.
17. The method of claim 14 , wherein the topology comprises a topology for a part of a character.
18. The method of claim 13 , wherein the feature caddie comprises a minimum morphing distance.
19. The method of claim 13 , wherein the feature caddie comprises a degree 2 graph vertices.
20. The method of claim 13 , wherein the feature caddie comprises a contour of a letter.
21. The method of claim 13 , wherein the feature caddie comprises a character height.