Writer identification by handwritten text analysis

This work has calculated and implemented some methods used by professional person on forensic analysis, for test on dubitative documents. This system obtains different types of characteristics and they are tested with known samples from our database. It has been used writing samples from 30 writers, and we have got a success rate of 94, 66%, applying as classifier Neural Network, and after, the technique of "more voted" algorithm, with 10 Neural Networks.

[1]  Robert M. Haralick,et al.  Document page decomposition by the bounding-box project , 1995, Proceedings of 3rd International Conference on Document Analysis and Recognition.

[2]  Thierry Paquet,et al.  Handwritten Document Analysis for Automatic Writer Recognition , 2005 .

[3]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[4]  Horst Bunke,et al.  A Set of Novel Features for Writer Identification , 2003, AVBPA.

[5]  Horst Bunke,et al.  Writer identification using text line based features , 2001, Proceedings of Sixth International Conference on Document Analysis and Recognition.

[6]  Tieniu Tan,et al.  Writer Identification from Non-uniformly Skewed Handwriting Images , 1998, BMVC.

[7]  Vassilis Anastassopoulos,et al.  Morphological waveform coding for writer identification , 2000, Pattern Recognit..

[8]  A. Harvey,et al.  Skew detection in handwritten scripts , 1997, TENCON '97 Brisbane - Australia. Proceedings of IEEE TENCON '97. IEEE Region 10 Annual Conference. Speech and Image Technologies for Computing and Telecommunications (Cat. No.97CH36162).

[9]  Gyeonghwan Kim,et al.  Chaincode Contour Processing for Handwritten Word Recognition , 1999, IEEE Trans. Pattern Anal. Mach. Intell..