A review of image tampering detection methods based on deep learning

With the increasing popularity of digital image editing tools, image tampering has become much easier. A large number of tampered false images are now circulating on the Internet and social media, threatening the authenticity and credibility of critical domains such as law, journalism, and scientific research. Image tampering detection aims to identify and locate altered areas within tampered images, thereby safeguarding their credibility. This paper provides a comprehensive review of deep learning-based methods for image tampering detection. First, it introduces the current research status in this field. Next, it classifies deep learning approaches developed over the past five years. The paper also highlights the main datasets and evaluation metrics used, along with a performance comparison of various methods. Finally, it discusses the limitations of current tampering detection methods and offers insights into future development directions.