This chapter provides an overview of digital image forgery detection techniques, categorized into active and passive approaches. Active methods, such as digital watermarking and digital signatures, require pre-embedded information, while passive methods analyze inconsistencies without prior data. Traditional passive techniques include pixel, format, camera, geometry, and environment-based methods. The chapter also highlights the rise of deep learning approaches using CNNs, Autoencoders, GANs, RNNs, Transformers, and hybrid models. It concludes by emphasizing the need for advanced, adaptable systems to address increasingly sophisticated forgeries.