【摘要】 针对现有脆弱水印算法大多无法区分水印被篡改还是内容被篡改的问题,提出了一种基于奇异值分解的自嵌入水印算法.算法基于原始图像2×2大小分块,图像块的高6位进行奇异值分解,每块的最大奇异值经过量化生成恢复水印嵌入到偏移子块的次低位,每块的奇异值范数二值编码生成认证水印嵌入到本身的最低位.实验结果表明,该算法不仅篡改定位准确,而且能区分是水印被篡改还是图像内容被篡改,可有效地恢复被篡改的区域.更多还原

【Abstract】 The existing fragile watermarking algorithms almost can′t recognize the modification made to the watermarked image is is on the image content or on the embedded watermark.A self-embedding watermark algorithm is proposed based on SVD.The algorithm based on the 2×2 pixel image block.The high 6 bit of image block is singular value decomposition,the maximum SVs from each block to be quantized construct recovery watermark and is embedded into the Hypo-LSB of the corresponding excursion block selected by a key.Binary of singular value norm of each block to be construct authentication watermark is embedded into its LSB.Experimental results demonstrate that the proposed algorithm can not only locate tampered blocks of tamper accurately,but also discriminating tampering watermark from that of content.And it can restore the tampered region is effective.更多还原