Source Cell-phone Identification

• Published in: 2006 IEEE 14th Signal Processing and Communications Applications pp. 1 - 3
• Main Authors: Celiktutan, O., Avcibas, I., Sankur, B., Ayerden, N.P., Capar, C.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 2006
• Subjects: Cameras; Filters; Higher order statistics; Histograms; Image quality; Influenza; Interpolation; Lenses; Support vector machine classification; Support vector machines
• ISBN: 1424402387; 9781424402380
• ISSN: 2165-0608; 2693-3616
• DOI: 10.1109/SIU.2006.1659882

In this paper, we focus on blind source cell-phone identification problem. The main idea is that proprietary interpolation algorithm (involved due to the structure of color filter array [CFA]) leaves footprints in the form of correlations across adjacent bit planes of images. For this purpose, we explore a set of binary similarity measures, image quality measures and higher order wavelet statistics in conjunction with KNN and SVM classifiers to identify the originating cell-phone. We provide identification results among 9 different brand cell-phone cameras