Title:	AN OFF-LINE PERSIAN HANDWRITTEN FORGERY DETECTION METHOD
Author(s):	Behzad Helli, Mohsen Ebrahimi Moghaddam
ISBN:	978-972-8939-48-9
Editors:	Yingcai Xiao
Year:	2011
Edition:	Single
Keywords:	Handwriting verification, image processing, forgery
Type:	Short Paper
First Page:	297
Last Page:	302
Language:	English
Cover:
Full Contents:	click to dowload
Paper Abstract:	The security of handwritten documents is very important in authentication systems. In this paper, a forgery detection method is proposed for Persian handwritten documents, which uses two types of novel features: Macro and Micro, Macro features show the structure of handwritten while micro ones show more details. Also, the micro features try to extract some online properties from offline data such as pen pressure and velocity. After extracting those features Weighted Euclidean Distance (WED) classifier is used to find forgery ones. It is very important that the weights of this classifier have been adjusted based on true data because forgery data are not identified in adjusting phase. To test the proposed method a Persian handwritten data set was prepared using three kinds of forgeries; Mimic, unskilled, and skilled. The method performance using different reference words showed the best result in correct rejection was 82% while the correct acceptance was 94%. Also, we believe that this method can be extended to other languages by adjusting some parameters.
	Go Back