Digital Image Forgery Detection Using Basic Manipulations In Facebook

Abstract In this modern age in which we are living digital images play a vital role in many application areas like social networking websites for example Facebook. But at the same time the image retouching techniques has also increased which forms a serious threat to the security of digital images in Facebook. To cope with this problem the field of digital forensics and investigation has emerged and provided some trust in digital images. In this paper we present a new algorithm to detect digital image forgery based on cellular automata and data embedding in spatial domain. The original JPEG image which the user uploads initially on hisher profile will be partitioned into some regions. We use region-based segmentation to specifying the desired regions of interest from the input image. First we extract the visual attributes of the original image and achieve the statistical information for the selected region and save it in the database. Then we apply linear cellular automata rules to create a robust cipher key from these values. We embed the cipher key into the spatial domain to authenticate and validate the original image. The proposed algorithm is applied on 100 numbers of grayscale images size 800 x 600. The results have demonstrated the robustness and stable time complexity of the proposed method.