非等量备份和双认证自修复有限域图像分存
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

国家自然科学基金(61100239);陕西省科技新星计划(2011kjxx17);陕西省自然科学基金(2011JQ8009,2016JM6065);中央高校基本科研业务费支持项目(GK201402036,GK201703057)


Gloise Field Self-Recovery Image Sharing Scheme with Non-Equivalent Backup and Double Authentications
Author:
Affiliation:

Fund Project:

National Natural Science Foundation of China (6110239); Science and Technology New Star Project of Shaanxi Province (2011kjxx17); Natural Science Foundation of Shaanxi Province (2011JQ8009, 2016JM6065); Fundamental Research Fund for the Central Universities of China (GK201402036, GK201703057)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    传统有意义图像分存方案存在认证能力偏低、攻击后不具备修复能力或修复能力整体较弱以及嵌入掩体视觉质量不高等问题.针对以上问题,提出一种结合非等量备份和双认证自修复有限域图像分存方案,包含分存和恢复阶段.在分存阶段,首先对密图做1级离散小波变换,取LL子带按密钥置乱,并对置乱后LL子带每个系数比特按比特位重要程度分组进行非等量备份来构造与密图等大备份图;然后对密图和备份图每个像素及其对应7K-13位认证信息在GF(27)有限域进行(K,N)分存,将产生的7位分存信息和使用密钥产生的1位认证信息使用优化LSB法嵌入到N个掩体2×2分块中;最后对密钥进行(K,N)分存,将子密钥对应的MD5值公开到第3方公信方并将子密钥和嵌入掩体分发给参与者.在恢复阶段,首先对参与者提供的子密钥真实性进行检验,利用检验通过子密钥对密钥进行恢复;其次对分发掩体2×2分块嵌入的分存信息和1位认证信息使用密钥进行第1重认证,利用第1重认证通过分存信息重建GF(27)有限域分存多项式,提取出密图和备份图每个像素及其对应的7K-13位认证信息并对其进行第2重检验和构造初步密图、备份图以及认证图;再次由备份图和认证图重构密图LL子带,然后对其做逆置乱和逆离散小波变换得到密图修复参考图;最后对认证图每一个认证不通过秘密像素,根据其周围像素认证情况选择多项式插值拟合或进行修复参考图像素替代修复.理论和实验结果表明,与现有方法相比,所提方法具备更好的认证能力,并能充分使用双认证和自然图像邻近像素相关性来提升其攻击后修复能力,且分发掩体具备较高视觉质量.

    Abstract:

    In conventional meaningful image sharing schemes, there exist some shortages such as low authentication ability, weak to no repair or recovery ability after being attacked, and low visual quality in embedded carriers. To address these problems, this paper proposes a Gloise field self-recovery image sharing scheme with non-equivalent backup and double authentication. The proposed scheme includes both distributing and recovery stages. The distributing stage include three steps. First, the LL subband of the secret image is obtained by one layer DWT and then the LL subband coefficients is scrambled by a secret key to generate the non-equivalent backup image according to different bit important degree groups where both the backup image and the secret image are in same size. Second, each pixel in both the secret image and the backup image are shared with their 7K-13 authentication bits into 7 bits distributed information on GF(27), and then the distributed 7 bits information with its 1 bit authentication generated by the secret key is embedded into 2×2 blocks of N distributed carriers by the optimized LSB embedding method. Finally, the secret key used to scramble LL subband and generate 1 bit authentication is also shared into N sub-keys by(K,N)-threshold scheme, and then the sub-keys' MD5 values are published into the third reliable party and theNsub-keys and embedded carriers are distributed toNparticipants. The recovery stage has four steps. First, the facticity of participants' sub-keys is verified, and then the sub-keys are used to recovery the secret key. Second, the embedded distributed information with its 1 bit authentication is checked by the first authentication, the checked distributed information is used to reconstruct GF(27) interpolation polynomial, and then the pixels in the secret image and the backup image with their 7K-13 authentication bits are recovered to construct the preliminary secret image, backup image and authentication image. Third, the backup image and authentication image are used to reconstruct the LL subband of the secret image and then to build the repair reference image by the descramble method and inverse DWT. Finally, the unpassed secret image pixels in the authentication image are self-recovered by the polynomial interpolation fitting and pixel replacement in repair reference image. The theory and experiment show the proposed method has a superb authentication ability and can make full use of double authentication ability and the adjoin pixels to improve its self-recovery ability, and the embedded carriers have better visual qualities.

    参考文献
    相似文献
    引证文献
引用本文

欧阳显斌,邵利平,乐志芳.非等量备份和双认证自修复有限域图像分存.软件学报,2017,28(12):3306-3346

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2016-04-11
  • 最后修改日期:2016-09-20
  • 录用日期:
  • 在线发布日期: 2017-03-27
  • 出版日期:
您是第位访问者
版权所有:中国科学院软件研究所 京ICP备05046678号-3
地址:北京市海淀区中关村南四街4号,邮政编码:100190
电话:010-62562563 传真:010-62562533 Email:jos@iscas.ac.cn
技术支持:北京勤云科技发展有限公司

京公网安备 11040202500063号