将载体图像按照不同的扫描方式转换为不同的一雏序列,通过计算一维序列前后两个元素的差值,再统计出该差值序列中频率出现最多的两个值,计算不同扫描方式中这两个值出现的总次数。选择总次数最大的扫描方式生成的一维序列中进行信息隐藏,在频率最大的两个数据中隐藏信息。实验结果表明,该算法隐藏容量大,载体图像的峰值信噪比高。
A high-capacity reversible data hiding algorithm based on neighboring pixels is proposed in this paper. Cover image is first converted into one dimensional (1D) sequence by different scan patterns. The difference values between each two neighbor elements in the 1D sequence are calculated and the two difference values with biggest frequencies are distinguished. For each scan pattern, the occurrence number of the two difference values is achieved. The scan pattern with the biggest occurrence number is used for data hiding, where the two difference values with biggest frequencies are exploited to embed secret messages. Experimental results show that the proposed algorithm has large embedding capacity and high PSNR value.