用于医学影像快速篡改检测和恢复的无损水印算法

Quick tampering detection and recovering algorithm based on reversible watermark in medical images

针对现有方法中篡改检测效率不高、定位不精确的问题,提出了一种基于无损水印和四叉树分解的医学图像快速篡改检测及恢复的方法。利用对医学图像进行四叉树分解过程中的层次结构特点,提高了篡改检测精确性和定位速度;同时使用分解后块中对角线像素均值作为恢复特征值,保证篡改后图像的修复质量。实验结果表明,与现有方法相比,所提方法在尺寸为512×512的图像中,定位比较次数降至6.7次左右,篡改定位精确性提高了5%左右。

Aiming at the problem of low efficiency of tampering detection and accuracy of location, a medical image tampering detection and recovering method based on reversible watermark and quad-tree decomposition was proposed. The algorithm has higher accuracy and faster tampering location speed by using the hierarchical structure of the quad-tree in the decomposition of the medical images. The method used the diagonal pixel mean in the block as the recovered feature value, which ensures the recovery quality of tampered image. The experimental results show that compared with the existed methods, the proposed algorithm reduces the comparing times for locating tampered region to about 6.7 in the 512×512 images and improves the tampering detection accuracy about 5%.