Publications

@article{bennabhaktula_camera_2022,

title = {Camera model identification based on forensic traces extracted from homogeneous patches},

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

url = {https://www.sciencedirect.com/science/article/pii/S0957417422010430},

year  = {2022},

date = {2022-01-01},

journal = {Expert Systems with Applications},

volume = {206},

pages = {117769},

abstract = {This work addresses the challenge of identifying the source camera model in digital image forensics, crucial for investigations by Law Enforcement Agencies. The proposed solution extracts small homogeneous regions from the integral image and uses a hierarchical classification approach with convolutional neural networks. This method outperforms traditional classifiers and achieves a 99.01% accuracy on the Dresden dataset’s ‘natural’ subset, marking the best result reported to date.},

note = {Publisher: Pergamon},

keywords = {Camera Model Identification, Digital Image Forensics, Sensor Pattern Noise},

pubstate = {published},

tppubtype = {article}

}