首页 | 本学科首页   官方微博 | 高级检索  
     

基于可形变自相关网络的图像篡改检测方法
引用本文:梁鹏,吴玉婷,赵慧民,李春英,何娃,黎绍发. 基于可形变自相关网络的图像篡改检测方法[J]. 计算机工程, 2021, 47(11): 241-246,253. DOI: 10.19678/j.issn.1000-3428.0059658
作者姓名:梁鹏  吴玉婷  赵慧民  李春英  何娃  黎绍发
作者单位:广东技术师范大学 计算机科学学院,广州 510665;广东技术师范大学 电子与信息学院,广州 510665;华南理工大学 计算机科学与工程学院,广州 510641
基金项目:国家自然科学基金面上项目(62072123);国家自然科学基金青年科学基金项目(61807009);广东省普通高校重点领域专项(2020ZDZX3059);广东省自然科学基金面上项目(2018A0303130187);广东省普通高校重点实验室项目(2019KSYS009)。
摘    要:基于深度学习的图像复制-粘贴篡改检测方法在特征提取过程中未考虑特征的空间排列,在小区域篡改样本下检测性能不佳。基于可形变自相关网络提出一种图像篡改检测方法。通过引入可形变卷积和多尺度空间金字塔,自适应地学习篡改目标的空间形变,同时通过构造自相关金字塔式特征层次结构,融合全局特征和局部特征以提升图像篡改检测性能。实验结果表明,该方法在公开的图像篡改检测基准上各项评测指标均优于对比方法,其精确率、召回率、F1值较BusterNet 2019分别提高14.85、15.04、12.81个百分点,在小区域篡改样本下性能优势更为明显。

关 键 词:图像篡改检测  特征提取  可形变卷积  自相关金字塔
收稿时间:2020-10-09
修稿时间:2020-11-23

Method for Image Forgery Detection Based on Deformable Self-Correlation Network
LIANG Peng,WU Yuting,ZHAO Huimin,LI Chunying,HE Wa,LI Shaofa. Method for Image Forgery Detection Based on Deformable Self-Correlation Network[J]. Computer Engineering, 2021, 47(11): 241-246,253. DOI: 10.19678/j.issn.1000-3428.0059658
Authors:LIANG Peng  WU Yuting  ZHAO Huimin  LI Chunying  HE Wa  LI Shaofa
Affiliation:1. College of Computer Science, Guangdong Polytechnic Normal University, Guangzhou 510665, China;2. College of Electronics and Information, Guangdong Polytechnic Normal University, Guangzhou 510665, China;3. School of Computer Science and Engineering, South China University of Technology, Guangzhou 510641, China
Abstract:The deep learning-based copy-move forgery detection methods ignore the spatial layout of the features, leading to a reduction in the detection performance for small-region forgery samples.Additionally,the fixed size of the receptive fields in Convolutional Neural Network (CNN) modules are not suitable for the detection of nonrigid image forgery.To address the problem,a copy-move forgery detection method is proposed based on a deformable self-correlation network.The method introduces deformable convolution and multi-scale spatial pyramid to adaptively learn the spatial deformation of the forgery target.At the same time,a self-correlation pyramidal feature hierarchy is constructed to integrate the global features and local features to improve the performance of image forgery detection. Experimental results show that the proposed method is superior to the compared methods in all indexes of image forgery detection performanc.Compared with BusterNet 2019,the proposed method increases the accuracy by 14.85 percentage point,recall rate by 15.04 percentage point,and F1 score by 12.81 percentage point, especially in the case of small-region forgery samples.
Keywords:image forgery detection  feature extraction  deformable convolution  self-correlation pyramid  
本文献已被 万方数据 等数据库收录!
点击此处可从《计算机工程》浏览原始摘要信息
点击此处可从《计算机工程》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号