Illumination Inconsistency Sleuthing for Exposing Composition Telltales in Digital Images
Resumo
Images are powerful communication tools. Due to this power it is often worrisome when image manipulations come into play allowing forgers to deceive viewers, change opinions or even affect how people perceive reality. Therefore, it is paramount to devise and deploy efficient and effective forgery detection techniques. From all types of image forgeries, composition images (forgeries using parts of two or more images) are of particular interest. Among different techniques for spotting forgeries, image illumination inconsistencies are the most promising. This work builds upon the hypothesis that “image illumination inconsistencies are strong and powerful evidence of image composition” and presents four original and effective approaches to detecting image forgeries. The first method explores eye specular highlight telltales to estimate the light source and viewer positions in an image. The second and third approaches explore the color phenomenon called metamerism, whereby two colors may appear to match under one light source but appear completely different under another one. Finally, the last approach relies on user’s interaction to specify 3-D normals of suspect objects in an image from which the 3-D light source position can be estimated. These approaches represent important advances, which certainly will be a strong tool for forensic analysts against image forgeries.
Referências
Carvalho, T., Farid, H., and Kee, E. (2015b). Exposing Photo Manipulation From User-Guided 3-D Lighting Analysis. In SPIE Symposium on Electronic Imaging, San Francisco, CA, USA.
Carvalho, T., Pedrini, H., and Rocha, A. (To appear in 2015c). Visual computing and machine learning techniques for digital forensics. Revista de Informática Teórica e Aplicada (RITA).
Carvalho, T., Pinto, A., Silva, E., da Costa, F., Pinheiro, G., and Rocha, A. (2012). Escola Regional de Informática de Minas Gerais, chapter Crime Scene Investigation (CSI): da Ficção `a Realidade. UFJF.
Carvalho, T., Riess, C., Angelopoulou, E., Pedrini, H., and Rocha, A. (2013). Exposing Digital Image Forgeries by Illumination Color Classification. IEEE T.IFS, 8(7):1182–1194.
Saboia, P., Carvalho, T., and Rocha, A. (2011). Eye Specular Highlights Telltales for Digital Forensics: A Machine Learning Approach. In IEEE ICIP, pages 1937–1940.
