Blind inpainting forgery detection

Trung Thanh Dang (1) , Azeddine Beghdadi (1) , Mohamed-Chaker Larabi (2)

1 L2TI - Laboratoire de Traitement et Transport de l'Information
2 XLIM-SIC - SIC

The increasingly use of digital images in our daily life and the availability of powerful software for processing and editing images, open new challenges regarding illegal or unauthorized image manipulation. Thus, it becomes essential to authenticate digital copies, validate their content, and detect possible forgeries. In this paper, we focus on detection of a specific type of digital forgery, inpainting, where an object is removed using pixels coming from the same image. The problem of inpainting detection is investigated and an efficient and reliable detection method is proposed. The performance of the proposed method is demonstrated on several inpainted images using different inpainting techniques.

Mots clés: Inpainting forgery detection; digital image forensics; digital tamper detection

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

Trung Thanh Dang, Azeddine Beghdadi, Mohamed-Chaker Larabi. Blind inpainting forgery detection. IEEE Global Conference on Signal and Information Processing (GlobalSIP), Dec 2014, Atlanta, United States. ⟨10.1109/GlobalSIP.2014.7032275⟩. ⟨hal-01159130⟩