Volume 2 - Issue 2, March - April 2026

Due to the swift progress of digital imaging editing software, the issue of image forgery has emerged as a significant challenge in preserving the authenticity and trustworthiness of visual materials. Identifying altered or edited images is crucial for fields such as digital forensics, journalism, and cybersecurity. This initiative presents SIFD-NET (Strong Image Forgery Detection Network), a framework based on deep learning intended to reliably detect and pinpoint forged areas in digital images. SIFD-NET employs a Convolutional Neural Network structure paired with attention mechanisms to recognize subtle discrepancies in texture, lighting, and border artifacts resulting from manipulation methods such as splicing, copy-move, and removal forgeries. The model is trained using publicly accessible datasets that include both genuine and forged images, ensuring high levels of generalization and resilience. The network incorporates feature extraction layers to grasp advanced semantic patterns and utilizes residual learning to improve the detection of small pixellevel variations. In summary, SIFD-NET offers a powerful, automated, and scalable approach to image forgery detection, which greatly aids in the authentication and verification of integrity in digital media within practical applications.

Bairagoni Vaishnavi, Maloth Tharun, Pothuganti Nikhil, Chandupatla Siddhartha,"A Robust System for Detection of Forgery Images" International Journal of Advanced Multidisciplinary Research and Educational Development, V2(2): Page(1209-1212) Mar-Apr 2026. ISSN: 3107-6513. www.ijamred.com. Published by Scientific and Academic Research Publishing.