Deluding Image Recognition in SIFT-based CBIR Systems

Thanh-Toan Do (1) , Ewa Kijak (1) , Teddy Furon (2) , Laurent Amsaleg (1)

1 TEXMEX - Multimedia content-based indexing
2 TEMICS - Digital image processing, modeling and communication

Content-Based Image Retrieval Systems used in forensics related contexts require very good image recognition capa- bilities. Therefore they often use the SIFT local-feature de- scription scheme as its robustness against a large spectrum of image distortions has been assessed. In contrast, the security of SIFT is still largely unexplored. We show in this paper that it is possible to conceal images from the SIFT-based recognition process by designing very SIFT-specific attacks. The attacks that are successful in deluding the system re- move keypoints and simultaneously forge new keypoints in the images to be concealed. This paper details several strate- gies enforcing image concealment. A copy-detection oriented experimental study using a database of 100,000 real images together with a state-of-art image search system shows these strategies are effective. This is a very serious threat against systems, endangering forensics investigations.

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Thanh-Toan Do, Ewa Kijak, Teddy Furon, Laurent Amsaleg. Deluding Image Recognition in SIFT-based CBIR Systems. ACM Multimedia in Forensics, Security and Intelligence, ACM, Oct 2010, Firenze, Italy. ⟨inria-00505845⟩