Automated Forgery Analysis in Images with Deep Neural Networks
Keywords:
Digital image forgery, Forensic imaging, MICC-F220 dataset, VGG16 classifier, DBSCAN clustering, SIFT key point extraction.Abstract
Digital image forgery has witnessed a staggering rise, with over 85% of manipulated images containing copy-move forgeries, and nearly 60% of these cases going undetected in manual screening. According to recent forensic imaging studies, the global cost of digital content fraud exceeds $4.5 billion annually,prompting an urgent need for robust detection mechanisms. Traditional manual approaches such as visual inspection, EXIF metadata analysis, and Error Level Analysis (ELA) are limited in scope, prone to human error, and inefficient when handling high-resolution images or large datasets.
References
Xiao B. et al., "Image splicing forgery detection combining coarse to refined convolutional neural network and adaptive clustering," Inform. Sci., 2020.
Saini K. et al., "Forensic examination of computer-manipulated documents using image processing techniques," Egypt. J. Forensic Sci., 2016.
Lyu Q. et al., "Copy Move Forgery Detection based on double matching," J. Vis. Commun. Image Represent., 2021.
