Publications

@article{bennabhaktula_device-based_2020,

title = {Device-based image matching with similarity learning by convolutional neural networks that exploit the underlying camera sensor pattern noise},

author = {Guru Swaroop Bennabhaktula and Enrique Alegre and Dimka Karastoyanova and George Azzopardi},

url = {https://arxiv.org/abs/2004.11443},

year  = {2020},

date = {2020-01-01},

journal = {arXiv preprint arXiv:2004.11443},

abstract = {This paper addresses the challenge of identifying whether two images originate from the same camera, aiding forensic investigations. A two-part network is proposed to quantify the likelihood of a shared source, evaluated on the Dresden dataset (1851 images from 31 cameras). While not yet forensics-ready, the approach achieves 85% accuracy, showing promising results. This research is part of the EU-funded 4NSEEK project focused on combating child sexual abuse.},

keywords = {4NSEEK project, Camera Identification, deep learning, DigitalImage Forensics},

pubstate = {published},

tppubtype = {article}

}

@article{timmerman_video_2020,

title = {Video camera identification from sensor pattern noise with a constrained convnet},

author = {Derrick Timmerman and Guru Swaroop Bennabhaktula and Enrique Alegre and George Azzopardi},

url = {https://arxiv.org/abs/2012.06277},

year  = {2020},

date = {2020-01-01},

urldate = {2020-01-01},

journal = {arXiv preprint arXiv:2012.06277},

abstract = {This work presents a method for identifying the source camera of a video by analyzing camera-specific noise patterns extracted from video frames. The proposed approach utilizes an extended version of a constrained convolutional layer to process color inputs and extract noise pattern features. The system classifies individual video frames, and a majority vote is used to determine the source camera. The approach was evaluated on the VISION dataset, which contains 1539 videos from 28 different cameras. The results show promising accuracy, with up to 93.1% achieved, even with video compression techniques like WhatsApp and YouTube. This research is part of the EU-funded 4NSEEK project, aimed at forensics in cases of child sexual abuse.},

keywords = {4NSEEK project, Forensic Analysis, VISION},

pubstate = {published},

tppubtype = {article}

}