Multiclassification Tampering Detection Algorithm Based on Spatial‐Frequency Fusion and Swin‐T

Deep learning methods for image forgery detection often struggle with compression attack robustness. This paper proposes a novel multi‐class forgery detection framework combining spatial‐frequency fusion with Swin‐Transformer, outperforming existing methods in compression attack scenarios. Our appro...

Full description

Saved in:
Bibliographic Details
Published inIET image processing Vol. 19; no. 1
Main Authors Li, Li, Zhang, Kejia, Lu, Jianfeng, Zhang, ShanQing, Chu, Ning
Format Journal Article
LanguageEnglish
Published 01.01.2025
Online AccessGet full text
ISSN1751-9659
1751-9667
1751-9667
DOI10.1049/ipr2.70007

Cover

More Information
Summary:Deep learning methods for image forgery detection often struggle with compression attack robustness. This paper proposes a novel multi‐class forgery detection framework combining spatial‐frequency fusion with Swin‐Transformer, outperforming existing methods in compression attack scenarios. Our approach integrates a frequency domain perception module with quantization tables, a spatial domain perception module through multi‐strategy convolutions, and a dual‐attention mechanism combining spatial and channel attention for feature fusion. Experimental results demonstrate superior performance with an F 1 score of 87% under JPEG compression ( q  = 75), significantly surpassing current state‐of‐the‐art methods by an average of 15% in compression resistance while maintaining high detection accuracy.
ISSN:1751-9659
1751-9667
1751-9667
DOI:10.1049/ipr2.70007