Hybrid deep learning and machine learning approach for passive image forensic

• Published in: IET image processing Vol. 14; no. 10; pp. 1952 - 1959
• Main Authors: Thakur, Abhishek, Jindal, Neeru
• Format: Journal Article
• Language: English
• Published: The Institution of Engineering and Technology 21.08.2020
• Subjects: CMFD image manipulation dataset; deep neural network algorithm; DL algorithm; emerging methods; feature extraction; forged forged categories; forgeries; Gabor filters; image classification; image coding; image forensics; image segmentation; iris recognition; learning (artificial intelligence); machine learning approach; machine learning‐based approach; neural nets; not forged categories; passive image forensic; passive image forgery detection; Special Section: Multidisciplinary advancement of imaging technologies: from medical diagnostics and genomics to cognitive machine vision, and artificial intelligence; machine learning approach; image classification; DL algorithm; CMFD image manipulation dataset; emerging methods; image forensics; deep neural network algorithm; passive image forgery detection; forged forged categories; feature extraction; image segmentation; forgeries; not forged categories; passive image forensic; iris recognition; Gabor filters; learning (artificial intelligence); image coding; neural nets; machine learning-based approach
• ISSN: 1751-9659; 1751-9667
• DOI: 10.1049/iet-ipr.2019.1291

Image forgery detection using traditional algorithms takes much time to find forgeries. The new emerging methods for the detection of image forgery use a deep neural network algorithm. A hybrid deep learning (DL) and machine learning-based approach is used in this study for passive image forgery detection. A DL algorithm classifies images into the forged and not forged categories, whereas colour illumination localises forgery. The simulated results are compared to other algorithms on public datasets. The simulated results achieved 99% accuracy for CASIA1.0, 98% accuracy for CASIA2.0, 98% accuracy for BSDS300, 97% accuracy for DVMM, and 99% accuracy for CMFD image manipulation dataset.