吉林大学学报(信息科学版) ›› 2014, Vol. 32 ›› Issue (2): 131-137.

• 论文 • 上一篇    下一篇

基于Hopfield网络的彩色图像混沌加密算法

卢辉斌1,2, 王丽佳1   

  1. 1. 燕山大学 信息科学与工程学院, 河北 秦皇岛 066004; 2. 河北省特种光纤与光纤传感实验室, 河北 秦皇岛 066004
  • 出版日期:2014-03-25 发布日期:2014-06-12
  • 作者简介:卢辉斌(1964—), 男, 吉林蛟河人, 燕山大学教授, 主要从事网络与信息安全、 保密通信研究, (Tel)86-335-8049286(E-mail)yjsbl@ysu.edu.cn。
  • 基金资助:

    河北省教育厅基金资助项目(2007493)

Color Image Encryption Algorithm of Chaotic Based on the Hopfield Network

LU Huibin1,2, WANG Lijia2   

  1. 1. College of Information Science and Engineering, Yanshan University, Qinhuangdao 066004, China;2. The Key Laboratory for Special Fiber and Fiber Sensor of Hebei, Qinhuangdao 066004, China
  • Online:2014-03-25 Published:2014-06-12

摘要:

一般混沌图像加密, 都是对图像的整体像素置乱处理, 其抵抗明文攻击能力较差。为此, 提出基于离散Hopfield神经网络的彩色图象混沌加密算法。该算法采用自治三维混沌系统对彩色图像单像素比特位进行加密操作, 通过利用三维混沌序列的其中一维置乱图像R\,G\,B分量的像素位置, 用另外两维序列设置置乱每个像素比特位的权值和阈值, 从而改变彩色图像各分量像素的位置和像素值, 达到有效加密的效果。理论分析和实验结果表明, 该单像素加密算法可有效抵抗差分攻击, 使反馈密文提高了像素置乱效果, 并具有良好的加密效果和保密性。

关键词: 三维混沌系统, 反馈神经网络, 图像置乱, 加密

Abstract:

Generally encryption process is the overall processing image pixel so that it is weak to resist plaintext attack. In view of the problem of poor ability to prevent plaintext attack of pixel integral encryption, this article designs a chaotic single-pixel encryption algorithm of color image based on the discrete Hopfield network. This encryption method changes the pixel position of each component of image R,G,B by using one dimension chaotic sequence. Set each pixel bit weights and thresholds with the other two dimensional sequence in order to achieve effective encryption result. Theoretical analysis and experimental results show that the single pixel encryption algorithm can effectively resist differential attack. The feedback ciphertext can improve the pixel scrambling effect as well as good encryption effect and confidentiality of the encryption algorithm.

Key words: three-dimensional chaotic system, feedback neural network, image scrambling, encryption

中图分类号: 

  • TN918