A Study on Forgeries Detection in Digital Image

In contemporary society, social media platforms have become prevalent venues for individuals to share personal photographs, enhance social connectivity, and exchange information. These images are increasingly susceptible to manipulation due to the ease of access provided by social media. Recent advancement in communication technologies have facilitated the creation of counterfeit images with minimal effort by malicious actors. The field of image forensics emerged to authenticate such images. Modern technologies now employ deep neural network to detect image tampering expeditiously, in contrast to the protracted processes of the past. In this context, convolutional neural network are integrated with deep learning and machine learning methodologies to obscure discrepancies between cloned and original images, rendering them imperceptible to human observers. Extensive research has been conducted to address this issue, resulting in numerous techniques for detecting and localizing copy-move tampering and regional forgeries. This paper present a comprehensive literature review of innovative techniques in copy-move, region duplication, and splicing, retouching, cloning and composite forgeries from 2019 to 2024, highlighting their limitations and suggesting directions for future research. This paper also concludes by addressing future research trends and considerations.