A Feature Extracting Method for Tamper Detection in Printed Documents

Abstract

In this paper, we propose our feature extracting method for tamper detection in printed documents. To detect the tamper of the printed document, a feature extracting method is needed. In this paper, we describe our feature extracting method. Our feature extracting method is based on the location of the mean point of each dot. We have estimated our feature extracting method’s probability of the collision, its uniformity of the distribution, its invariability during D/A and A/D transform and its invariability during the ordinal change of paper. We have found that our feature extracting method can extract desirable feature value.