Format-Controlled Multi-Scale JPEG Compression Response Analysis for Image-Level Forgery Screening
arXiv:2607.06615v1 Announce Type: cross Abstract: Image forgery detection is a critical task in digital forensics, yet many deep-learning localization approaches are typically GPU-accelerated and computationally heavier than handcrafted screening methods. We propose a lightweight, interpretable feature engineering pipeline for image-level forgery screening using only CPU computation and gradient boosted trees. Our method introduces \emph{multi-scale Error Level Analysis} (ELA) computed at seven JPEG quality levels, combined with novel \emph{cross-quality ELA ratio} features that capture double
The proliferation of synthetic media necessitates more efficient and accessible methods for forgery detection, moving beyond computationally heavy, GPU-accelerated approaches.
This development offers a lightweight, CPU-based solution for image forgery screening, making advanced digital forensics more accessible and scalable for various applications.
Digital forensics can now leverage significantly less resource-intensive methods for initial image forgery detection, democratizing access to critical screening capabilities.
- · Digital Forensics Teams
- · Law Enforcement
- · News Agencies
- · Social Media Platforms
- · Heavy GPU-dependent Forensics Software
Widespread adoption of lightweight image forgery detection tools.
Reduced cost and increased speed of initial image authentication processes for numerous industries.
Enhanced trust in digital images, or conversely, a further escalation in the arms race between forgers and detectors as detection becomes more routine.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG