Combining PRNU and noiseprint for robust and efficient device source identification
This work addresses performance impairments in digital multimedia forensics for low-quality or limited data scenarios, though it appears incremental by integrating existing methods.
The paper tackles the problem of device source identification in challenging conditions like compressed or cropped images by combining PRNU with noiseprint, resulting in significant performance improvements as shown in numerical experiments on widely used datasets.
PRNU-based image processing is a key asset in digital multimedia forensics. It allows for reliable device identification and effective detection and localization of image forgeries, in very general conditions. However, performance impairs significantly in challenging conditions involving low quality and quantity of data. These include working on compressed and cropped images, or estimating the camera PRNU pattern based on only a few images. To boost the performance of PRNU-based analyses in such conditions we propose to leverage the image noiseprint, a recently proposed camera-model fingerprint that has proved effective for several forensic tasks. Numerical experiments on datasets widely used for source identification prove that the proposed method ensures a significant performance improvement in a wide range of challenging situations.