U.S. flag

An official website of the United States government, Department of Justice.

Improvements on source camera-model identification based on CFA interpolation

NCJ Number
306091
Date Published
March 2023
Length
10 pages
Annotation

In this paper, the authors discuss their use of updated technologies and methodologies to better detect the interpolation artifacts in smooth image parts.

Abstract

The authors of this paper report on their attempt to improve their previous approach to the source camera-model identification problem. The authors’ prior work is discussed in section one, which explains that they had studied the source of camera-model identification problem by identifying and selectively combining a set of image features based on image quality metrics and higher-order statistics of images. The authors’ premise for their current work is that most proprietary algorithms in smooth image parts will deploy simpler forms of interpolation, and as a result can be captured more effectively. In section two, the authors briefly describe the image formation process in digital cameras. In section three, the authors review the results of section two and provide details of the improved approach. They finally present their experimental results in section four, and their conclusions in section five.

Date Published: March 1, 2023