Establishment of Robust Source Camera Identification Technologies for Digital Images

2016 Fiscal Year Final Research Report

• Project/Area Number: 26330152
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Information security
• Research Institution: The University of Aizu (2015-2016); Tokyo University of Agriculture and Technology (2014)
• Principal Investigator: Tomioka Yoichi 会津大学, コンピュータ理工学部, 准教授 (10574072)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: カメラ識別 / クラスタPRNUノイズ / シーンコンテンツ / 拡大率推定

Outline of Final Research Achievements

In this research, we aim to establish source camera identification to identify a camera which took digital images using Photo Response Non-Uniformity (PRNU) Noise. We proposed a theoretical framework for accuracy evaluation of camera identification. To improve the accuracy of camera identification, we proposed a novel camera identification method based on the matching of image feature points on PRNU noise image, PRNU estimation method considering the effects of scene contents, and scaleing factor estimation method for a resized image. Moreover, in scanner identification, we showed the effects of the number and background color of scanned images, and we also showed the effects of aged deterioration.

Free Research Field

監視映像の自動解析とそのFPGA実装，Image forensics