沈军,廖鑫,秦拯,刘绪崇.基于卷积神经网络的低嵌入率空域隐写分析方法.软件学报,0,(0):0 |
基于卷积神经网络的低嵌入率空域隐写分析方法 |
Spatial Steganalysis of Low Embedding Rate Based on Convolutional Neural Network |
投稿时间:2019-06-18 修订日期:2019-09-23 |
DOI:10.13328/j.cnki.jos.005980 |
中文关键词: 隐写分析 卷积神经网络 低嵌入率 迁移学习 |
英文关键词:Steganalysis Convolution neural network Low embedding rate Transfer learning |
基金项目:国家自然科学基金(61972142,61402162,61772191);湖南省自然科学基金(2017JJ3040);模式识别国家重点实验室开放课题(201900017);湖南省科技计划重点项目(2015TP1004,2016JC2012);网络犯罪侦查湖南省普通高校重点实验室开放课题(2017WLFZZC001) |
|
摘要点击次数: 1096 |
全文下载次数: 1792 |
中文摘要: |
近年来,基于深度学习的空域隐写分析研究在高嵌入率下已经取得了较好的成果,但是对低嵌入率的检测效果还不太理想.因此本文设计了一个新的卷积神经网络结构,使用SRM滤波器进行预处理来获取隐写噪声残差,采用三个卷积层并对卷积核大小进行合理设计,通过适当选择批量归一化操作和激活函数来提升网络的性能.实验结果表明,与现有方法相比,本文提出的网络结构对WOW、S-UNIWARD和HILL三种常见的空域内容自适应隐写算法取得了更好的检测效果,且在低嵌入率0.2bpp、0.1bpp和0.05bpp下的检测效果有非常明显的提升.本文还提出了逐步迁移(step by step)的迁移学习方法,进一步提升低嵌入率条件下的隐写分析效果. |
英文摘要: |
In recent years, the research of spatial steganalysis based on deep learning has achieved good results under high embedding rate, but the detection performance under low embedding rate is still not ideal. Therefore, a new convolutional neural network structure is proposed, which uses the SRM filter for preprocessing to obtain implicit noise residuals, adopts three convolution layers and designs the size of convolution kernel reasonably, and selects appropriate batch normalization operations and activation functions to improve the network performance. The experimental results show that compared with the existing methods, the proposed network can achieve better detection performance for WOW, S-UNIWARD and HILL, three common adaptive steganographic algorithms in spatial domain, and significant improvement in detection performance at low embedding rates of 0.2bpp, 0.1bpp and 0.05bpp. A step-by-step transfer learning method is also designed to further improve the steganalysis effect under low embedding rate conditions. |
HTML 下载PDF全文 查看/发表评论 下载PDF阅读器 |