Device-based Image Matching with Similarity Learning by Convolutional Neural Networks that Exploit the Underlying Camera Sensor Pattern Noise
This addresses a specific challenge in digital image forensics, such as aiding investigations into crimes like child sexual abuse, but it is incremental as it builds on existing methods for sensor pattern noise.
The paper tackles the problem of identifying images captured by the same camera device for digital forensics, achieving 85% accuracy on the Dresden dataset with 1851 images from 31 cameras using a two-part network based on similarity learning and sensor pattern noise.
One of the challenging problems in digital image forensics is the capability to identify images that are captured by the same camera device. This knowledge can help forensic experts in gathering intelligence about suspects by analyzing digital images. In this paper, we propose a two-part network to quantify the likelihood that a given pair of images have the same source camera, and we evaluated it on the benchmark Dresden data set containing 1851 images from 31 different cameras. To the best of our knowledge, we are the first ones addressing the challenge of device-based image matching. Though the proposed approach is not yet forensics ready, our experiments show that this direction is worth pursuing, achieving at this moment 85 percent accuracy. This ongoing work is part of the EU-funded project 4NSEEK concerned with forensics against child sexual abuse.