• Divya M
• Anil Kumar M G
• Nagaratna Naik
• Nandana C L
• Sujatha R Upadhyaya

Image tampering is a prevalent issue in the digital age, where images and visual content are manipulated for malicious purposes. In the past, there have been multiple attempts to detect tampered images. Most of these methods focus on copy-move or image splicing, which are recognized as two major types of tampering techniques. While most of such efforts use machine learning as their primary technique for tamper detection, this paper presents a comprehensive approach to address both types of image tampering detection using Vision Learner, a deep learning technique. The system classifies images as authentic or tampered and also localizes the tampered regions with a high degree of accuracy. This research uses UNET, a well-known CNN architecture, for image splicing localization and DCT for copy-move localization. A passive image tampering detection technique is used to classify and localize image tampering based on the inconsistencies within the image. The CASIA dataset, consisting of more than 12600 generic images, has been used to demonstrate the experiments and results.

Image Tampering, CNN, Vision Learner, Localization, Image Splicing, Copy-Move Image, UNet, DCT.

"Vision Learner Based Image Tampering Detection and Localization", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 4, page no.i471-i478, April-2024, Available :http://www.jetir.org/papers/JETIR2404863.pdf

"Vision Learner Based Image Tampering Detection and Localization", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 4, page no. ppi471-i478, April-2024, Available at : http://www.jetir.org/papers/JETIR2404863.pdf