Muntadher H. Al-Hadaad, Rasha Thabit, Khamis A. Zidan , Bee Ee Khoo
Abstract
The recent interest of face image manipulation detection has been directed towards providing the ability of detecting various types of manipulations. In the best scenario, the available methods can detect the manipulations and localize the manipulated face region. The ability of recovering the face region after manipulation localization will be very useful in practical applications, however, this has not been highlighted in the previous researches. In this paper, a new face image authentication (FIA) scheme is presented based on image watermarking and Cohen–Daubechies–Feauveau (CDF) wavelets. In the proposed scheme, the CDF is used to generate the recovery bits from the face region in order to be used for recovering the face region when manipulations exist. Several experiments have been conducted to evaluate the performance of the proposed scheme which proved its efficiency in generating high quality watermarked images, detecting various types of manipulations, localizing the manipulated blocks in the face region, and recovering the face region with good visual quality. The comparison with the state-of-the-art detection schemes proved the superiority of the proposed scheme.
Publication Date
3/21/2024
Volume No
Issue No
DOI
DOI: 10.1007/978-981-99-9005-4_70
Pages
553–564
Keywords
DeepFakes reveal, Face image security, Face manipulation detection, Face manipulation localization, Multimedia forensics.
أ د خميس عواد زيدان | Khamis A. Zidan | 9208