基于多分类预测误差扩展的信息隐藏方法  

Information hiding method based on multi-class prediction error expansion

作　　者：顼聪[1] 刘朝霞[1] 陶永鹏[1] XU CONG;LIU Zhaoxia;TAO Yongpeng(School of Software,Dalian University of Foreign Languages,Dalian 116044,China)

机构地区：[1]大连外国语大学软件学院,辽宁大连116044

出　　处：《合肥工业大学学报（自然科学版）》2021年第11期1470-1475,共6页Journal of Hefei University of Technology：Natural Science

基　　金：国家自然科学基金资助项目(61806038);辽宁省教育厅科学研究重点资助项目(2019JYT01);辽宁省教育厅科学研究经费资助项目(LJKZ1033);大连外国语大学科研创新团队资助项目(2017CXTD01)。

摘　　要：传统的加密图像数据隐藏方法往往计算复杂,并且有嵌入率低和直接解密后的图像视觉质量差等缺点。为了解决这些不足,文章提出一种多分类的预测误差扩展方法,将图像分成多个子块,根据块内像素的位置采取中值边缘检测器(median edge detector,MED)、菱形检测器和均值计算检测器的组合预测误差方法,创建出更多差值直方图的峰值点,从而扩大了嵌入量。在信息嵌入过程中利用平滑函数将图像区域进一步划分成平滑区域和复杂区域,其中复杂区域不嵌入数据,有利于提高直接解密后的图像视觉质量。该算法易于实现,实验结果表明该算法具有较好的嵌入量和率失真性能。Traditional encrypted image data hiding methods are often complicated to calculate,and have the disadvantages of low embedding rate and poor visual quality of directly decrypted images.In order to solve these problems,a multi-class prediction error expansion method is proposed,which divides the image into multiple sub-blocks,and adopts the prediction error methods of median edge detector(MED),diamond detector and mean detector according to the different positions of pixels in the block to create more peak points of the difference histogram.In the process of information embedding,the smoothing function is used to divide the image area into smooth area and complex area.The complex area does not embed data,which is beneficial to the visual quality of the image after direct decryption.The algorithm is simple to implement,and the experimental results show that the algorithm has better embedding capacity and rate-distortion performance.

关 键 词：加密图像 可逆信息隐藏 预测误差 图像恢复 

分 类 号：TP391.413[自动化与计算机技术—计算机应用技术]