Statistical Feature Descriptors Based Methodology for Photo Forgery Detection

In the digital age, image is a popular medium of communication. The widespread use of image editing tools, the integrity of visual content has become essential. This research addresses the present issue of photo forgery by proposing a robust detection method grounded in statistical feature descriptors-based approaches. Image features plays an important role for identification of real or fake images. The statistical features descriptors provide authentic information about an image. In this paper statistical features descriptorbased methods have been used to detect imminent features of the images, then machine learning techniques are applied to the feature vector for classification of images in to real or fake. Proposed work adapts and evolves to detect new trends in photo-forging techniques by training a model on a dataset of original and manipulated photos. The outcomes of proposed experiments show how well our statistical method performs better to identify several types of forgeries, such as splicing and copy-move. This study advances the continuous endeavor to create reliable solutions for preserving the authenticity of photographs in the digital era.