基于HSB与LSB的半脆弱水印算法

提出了一种能够精确定位并能恢复原图像的半脆弱数字水印算法。本算法利用LSB(最低比特位)所含水印信息大容量的特点,由于在图像篡改和水印攻击情况下HSB(最高比特位)具有相对稳定性,故利用HSB作为水印信息,利用一种新的置乱算法产生水印嵌入位置,结合混沌序列,修改该位置图像像素的LSB,完成水印的嵌入调制。实验结果表明,该算法不仅可对图像内容的恶意篡改进行精确地检测与定位,而且能够大致恢复出被篡改的原图像信息,并对原图质量的影响非常小。

A Semi-fragile Watermark Scheme based on HSB and LSB

We propose a Semi-fragile watermark scheme which not only exactly locates the tampered place of the protected image but also recovers the lost information of the tampered place.Because the HSB(highest significant bit) is relatively stable under the malicious tamper and attack,by taking advantage of the high volume of LSB containing watermark information,this scheme uses HSB as the watermark information which is embedded into LSB(lowest significant bit) in the different place that is produced by a novel scrambling algorithm.The experiment shows that this scheme can detect and indicate the exact position of tampered part of image,moreover roughly recover the original image information on the tampered part,meanwhile cause the less significant effect on the original image after embedding watermark.