摘要
提出了一种基于压缩感知与小波域奇异值分解的图像哈希认证新方法.该方法首先通过小波变换得到图像的低频子带,然后对低频子带进行基于块的奇异值分解,最后对产生的最大奇异值数组进行基于压缩感知的随机投影,产生哈希认证信息,并偕同原图像传送到接收端.图像接收端通过比较哈希信息完成图像认证,并通过利用传递的哈希信息重建原始图像的奇异值,进一步映射奇异值差值到图像像素域,从而实现对图像篡改的识别与定位.大量实验结果证明了该方法的有效性.
A new hash based on image authentication scheme using compressive sensing with wavelet transform and singular value decomposition was proposed in this paper. The original image is first decomposed by wavelet transform to obtain low frequency subband, which is used to block based singular value decomposition. The hash information is generated by random projecting the maximum singular value data using compressive sensing, which is transmitted to receiver with the original image. In image receiver, the image authentication is achieved by comparing the received hash information and that generated by the received image. The detection and location of tampered image are accomplished by mapping the difference value of between singular value of the received image and that of the original image to image pixel domain. Experimental results has demonstrated the effectiveness of the proposed method.
出处
《中南民族大学学报（自然科学版）》
CAS
2010年第4期89-93,共5页
Journal of South-Central University for Nationalities：Natural Science Edition
基金
湖北省自然科学基金资助项目(2009CDA139)
中南民族大学自然科学基金资助项目(YZY10005)
中南民族大学大学生科研创新基金项目(KYCX100411B)
关键词
图像认证
篡改定位
压缩感知
小波变换
奇异值分解
Image authentication
tamper location
compressive sensing
wavelet transform
singular value decomposition
