A novel deep learning framework for double JPEG compression detection of small size blocks

Israr Hussain; Shunquan Tan; Bin LiXinghong QinDostdar HussainJiwu Huang

首页> 外文期刊>Journal of visual communication & image representation >A novel deep learning framework for double JPEG compression detection of small size blocks

【24h】

A novel deep learning framework for double JPEG compression detection of small size blocks

机译：A novel deep learning framework for double JPEG compression detection of small size blocks

获取原文

获取原文并翻译 | 示例

掌桥外文数据库（机构版） >>

开具论文收录证明 >>

文献代查 >>

页面导航

摘要
著录项
相关主题

摘要

Double JPEG compression detection plays a vital role in multimedia forensics, to find out whether a JPEGimage is authentic or manipulated. However, it still remains to be a challenging task in the case when thequality factor of the first compression is much higher than that of the second compression, as well as inthe case when the targeted image blocks are quite small. In this work, we present a novel end-to-end deeplearning framework taking raw DCT coefficients as input to distinguish between single and double compressedimages, which performs superior in the above two cases. Our proposed framework can be divided into twostages. In the first stage, we adopt an auxiliary DCT layer with sixty-four 8 × 8 DCT kernels. Using a specificlayer to extract DCT coefficients instead of extracting them directly from JPEG bitstream allows our proposedframework to work even if the double compressed images are stored in spatial domain, e.g. in PGM, TIFFor other bitmap formats. The second stage is a deep neural network with multiple convolutional blocks toextract more effective features. We have conducted extensive experiments on three different image datasets.The experimental results demonstrate the superiority of our framework when compared with other state-ofthe-art double JPEG compression detection methods either hand-crafted or learned using deep networks inthe literature, especially in the two cases mentioned above. Furthermore, our proposed framework can detecttriple and even multiple JPEG compressed images, which is scarce in the literature as far as we know.

著录项

来源
《Journal of visual communication & image representation》 |2021年第10期|103269.1-103269.8|共8页
作者
Israr Hussain; Shunquan Tan; Bin LiXinghong QinDostdar HussainJiwu Huang;
展开▼
作者单位

Guangdong Key Laboratory of Intelligent Information Processing, Shenzhen Key Laboratory of Media Security, Guangdong Laboratory of Artificial Intelligence andDigital Economy (SZ), Shenzhen 518060, China Shenzhen Institute of Artificial Intelligence and Ro;

Department of Computer Sciences, Karakoram International University Gilgit, 15100, Pakistan;

College of Computer Science and Software Engineering, Shenzhen University, Shenzhen 518060, China Guangdong Key Laboratory of Intelligent Information Processing, Shenzhen Key Laboratory of Media Security, Guangdong Laboratory of Artificial Intelligence an;

展开▼
收录信息
原文格式 PDF
正文语种英语
中图分类
关键词
Multimedia forensics; Discrete cosine transform; Deep learning; Double JPEG compression; Convolutional neural network;

A novel deep learning framework for double JPEG compression detection of small size blocks

摘要

著录项

相关主题

期刊订阅