摘要
G-LSB-M隐写方法是一种±1隐写嵌入方法,通过减少嵌入时的修改次数提高隐写的嵌入效率.该方法嵌入时所用的加减覆盖集只能通过穷举搜索构造,当嵌入消息分段长度n较大时,穷举搜索的计算代价过高导致无法实现.为了解决这一问题,本文提出了基于扩展加减覆盖集的隐写方法,通过从基础加减覆盖集中去除若干非必需的元素,能够以较小的计算代价构造出扩展加减覆盖集,避免了G-LSB-M方法中的穷举搜索困难,使得使用更长的信息分段进行嵌入成为可能.该方法降低了隐写时的平均修改次数,提高了嵌入效率和隐写的隐蔽性.
Generalized least significant bit matching (G-LSB-M) steganography scheme is a ± 1 embedding method which minimizes the number of modifications per pixel to improve the embedding efficiency .The sum and difference covering set (SDCS ) used for embedding can only be obtained through exhaustive search ,which is hard to implement when the number of message bits increases .To solve this problem ,we provide a novel steganography method based on the extended SDCS (ESDCS ) .The ESDCS is constructed by removing redundant elements from the so-called basic SDCS without exhaustive search ,which implies that longer message segments could be used in the embedding process .The proposed method reduces the expect number of modifications and thus improves the embedding efficiency as well as the security of steganography .
出处
《电子学报》
EI
CAS
CSCD
北大核心
2014年第6期1168-1172,共5页
Acta Electronica Sinica
基金
北京市自然科学基金课题(No.4112063)
国家自然科学基金课题(No.61170281
No.61303259)
中国科学院战略性先导科技专项课题(No.XDA06030600)
中科院信工所密码基金(No.Y3Z0012102)
关键词
扩展加减覆盖集
LSB匹配
隐写
嵌入效率
extended sum and difference covering set
LSB matching
steganography
embedding efficiency