一种针对手写体图像的取证算法

数字图像取证是信息安全领域的研究热点.在手写体图像方面,现有的取证算法对于图像采用各种重采样技术的篡改,其检测效果并不理想.在本文中,我们根据源区域和篡改区域的特征值不变性,提出一种简单有效的盲取证算法.该算法实现了手写体图像篡改的自动检测和篡改区域的定位,并且比现有的重采样检测方法对相同的手写体图像具有更好的检测率,尤其是在图像中的字有模糊、缺损的情况下检测的优势更明显.实验结果表明,对于经过各种重采样处理的手写体图像,该算法比现有取证算法的检测率高20%,当虚警率小于1%时,本算法的检测率达96.9%以上.

A Forensics Algorithm for Handwriting Images

Digital image forensics is the research focus in the field of information security.In the handwriting images area,when images are tampered by all sorts of resampling techniques,the current forensics algorithms are difficult to detect it.In this paper,a simple and effective detecting algorithm is proposed based on invariant property of the eigenvalues between source and tempered places.The proposed algorithm realizes automatic detection and location of tampered region,and has better detection rate than previous resmpling detection methods on the same handwriting image.Especially,when the words of image are fuzzy and defect,the advantages of the proposed algorithm are getting pretty obvious.Experimental results demonstrate that the accuracy of the proposed approach is higher than the existing forensics algorithm by 20%.The detection rate of the proposed algorithm reaches above 96.9% with a less than 1% false positive rate.