智能科学与技术学报 ›› 2024, Vol. 6 ›› Issue (2): 244-252.doi: 10.11959/j.issn.2096-6652.202414

• 学术论文 • 上一篇    

融入混合注意力的低缩放因子Seam Carving篡改检测算法

赵洁, 常皓婵, 武斌()   

  1. 天津城建大学计算机与信息工程学院,天津 300384
  • 收稿日期:2024-03-15 修回日期:2024-05-22 出版日期:2024-06-15 发布日期:2024-07-31
  • 通讯作者: 武斌 E-mail:wubin@tcu.edu.cn
  • 作者简介:赵洁(1984- ),男,博士,天津城建大学计算机与信息工程学院副教授,主要研究方向为图像处理、计算机视觉、机器学习。
    常皓婵(2000- ),女,天津城建大学计算机与信息工程学院硕士生,主要研究方向为图像取证、机器学习。
    武斌(1966- ),男,天津城建大学计算机与信息工程学院教授,主要研究方向为计算机视觉、模式识别。
  • 基金资助:
    天津市重点研发计划科技支撑重点项目(19YFZCGX00130);天津市企业科技特派员项目(19JCTPJC47200)

Low scaling factor Seam Carving tamper detection algorithm with hybrid attention

Jie ZHAO, Haochan CHANG, Bin WU()   

  1. School of Computer and Information Engineering, Tianjin Chengjian University, Tianjin 300384, China
  • Received:2024-03-15 Revised:2024-05-22 Online:2024-06-15 Published:2024-07-31
  • Contact: Bin WU E-mail:wubin@tcu.edu.cn
  • Supported by:
    Key Research and Development of Tianjin(19YFZCGX00130);The Enterprise Science and Technology Commissioner Project of Tianjin(19JCTPJC47200)

摘要:

针对现有的Seam Carving篡改检测算法对于低缩放因子情况存在检测精度不高、鲁棒性不强的问题,提出一种融入混合注意力机制的Seam Carving篡改检测算法。首先,利用BayarConv2D约束卷积对图像进行预处理,充分学习图像的噪声特征,并通过矩阵乘法与RGB图像进行特征融合;然后,采用ResNet作为骨干网络进行特征学习,引入残差传播和残差反馈机制,凸显Seam Carving的操作痕迹;最后,利用混合注意力机制同时提取相邻位置和通道之间的特征,更好地捕捉全局特征,进而将其输入全连接层进行分类。实验结果表明,在BOSSbase1.01数据集上,当缩放因子为1%和9%时,检测精度分别达到了89.48%和97.94%,优于现有主流方法,同时具有较低的计算复杂度和较好的鲁棒性,能够抵抗JPEG压缩攻击。

关键词: 混合注意力机制, 图像取证, Seam Carving检测, 低缩放因子

Abstract:

The existing seam carving tamper detection algorithms have the problems of low detection accuracy and weak robustness for the case of low scaling factor. A seam carving tamper detection algorithm integrated with hybrid attention mechanism is proposed. Firstly, BayarConv2D constrained convolution is used to preprocess the image, fully learn the noise characteristics of the image, and fuse the features with RGB image through matrix multiplication. Then, ResNet is used as the backbone network for feature learning, and the residual propagation and residual feedback mechanisms are introduced to highlight the operation traces of seam carving. Finally, the hybrid attention mechanism is used to simultaneously extract the features between adjacent locations and channels to better capture the global features, and then input them into the full connection layer to achieve classification. The experimental results show that on the BOSSBase1.01 dataset, when the scaling factor is 1% and 9%, the detection accuracy of the proposed method reaches 89.48% and 97.94% respectively, which is better than existing mainstream methods. At the same time, it has lower computational complexity and better robustness, and can resist JPEG compression attacks.

Key words: mixed attention mechanism, image forensics, Seam Carving detection, low scaling factor

中图分类号: 

No Suggested Reading articles found!