摘要
如何提升不同方法隐写图像检测分析精度是当前图像隐写领域研究的难点。针对不同容量的隐写图像,提出了1种多特征稀疏重构与湿纸码联合的隐写图像检测分析方法。在分析不同特征互补特性与冗余性的基础上,通过稀疏重构,实现了多特征的自适应融合分析。通过采用湿纸码的编解码形式实现隐写图像特征集合的稀疏求解。BOWS2(standard steganographic image data set,标准隐写图像数据集合)实验分析表明,本文方法有效提升了不同隐写速率和隐写实现的分析精度。
How to improve the accuracy of detection and analysis of the hidden image with different algorithms is a difficult problem in the field of image steganography. According to the different capacity of steganography, a new method to detect and analyze the multi feature sparse reconstruction and wet paper code is proposed. Based on the analysis of complementarity and redundancy of different features, multi-feature adaptive fusion analysis is realized by sparse reconstruction. The sparse solution of steganographic image feature set is realized by using the encoding and decoding form of wet paper code. The experimental results of BOWS2(standard steganographic image data set) show that the proposed method effectively improves the analysis accuracy of steganographic implementations at different steganographic rates.
作者
唐西西
韦艳玲
TANG Xixi;WEI Yanling(School of computer and Communication Engineering,Guangxi University of science and technology,Liuzhou 545006,China;School of Electronic Information Engineering,Liuzhou Vocational&Technical College,Liuzhou 545006,China)
出处
《内蒙古农业大学学报（自然科学版）》
CAS
北大核心
2019年第6期77-81,共5页
Journal of Inner Mongolia Agricultural University(Natural Science Edition)
基金
国家自然科学基金(61463008).
关键词
图像隐写
多特征融合
稀疏重构
主成分分析
湿纸码
Image steganography
multi-feature fusion
sparse reconstruction
principal component analysis
wet paper code
