Title Deep learning for image forgery detection: a comparative tudy with ELA preprocessing
Translation of Title Gilusis mokymasis vaizdų klastojimo aptikimui: lyginamasis tyrimas su ELA išankstiniu apdorojimu.
Authors Saab, Nathalie
Full Text Download
Pages 56
Keywords [eng] deep-learning ; image forgery detection ; CNN
Abstract [eng] This research aims to design, implement, and evaluate a deep-learning-based forgery detection pipeline through the combination of error level analysis, transfer learning, feature refinement, and explainable techniques while balancing detection accuracy and computational efficiency.
Dissertation Institution Kauno technologijos universitetas.
Type Master thesis
Language English
Publication date 2026