Server authentication of images

There has been tremendous advancement in multimedia in recent time. There are cameras everywhere around us, and the Internet is overwhelmed with all sorts of images. Mostly there are certain claims about the location and contents associated with images. Whenever an image shows up with associated claims on a media platform, there is always a question about the reliability of these claims associated with the image. There are numerous techniques to address these questions about the reliability of these claims. These include localization, signature usage, hide in an image or watermark images in order to prevent, reduce and detect any form of manipulation or modification. These techniques offer a limited level of reliability these images and associated claims, since the images can still be altered. Even a novice can make use of a sophisticated online tool and remove a watermark; a well scripted code can extract and replace a hidden image as well as alter the metadata of the original image. Thus, there is a need for an image authentication framework that can accurately detect any form of tampering or modification in an image. This research presents a framework to authenticate the image and the claim associated with it. The framework is based on server-client model. The server has the repository of all the authenticated images and associated claims, while the client consists of a web browser plugin. Once an image which is originated from the server is shared over the Internet reaches an Internet surfer, the surfer may want to verify the associated claim of the image. This is where the plugin kicks in. The plugin establishes a connection with the server and communicates the identity of the image already embedded into the image through some steganography technique. The frontend plugin and the server communicate further to authenticate the image and related claim. Primarily the server asks the client to send a random transversal profile of the image. Upon getting the response from the plugin the server takes the correlation of the profile and gets a correlation of the correlation of the same profile of the image from its repository. Once there is a correlation the image is verified, and the associated claims are sent to the plugin which displays the claims to the surfer. The significance of using the correlation is that the correlation depicts the harmonic variations in the image which are greatly preserved even when the resolution of the image has changed. The technique is even effective if the image is cropped. Moreover, the framework is immune to known attacks on the existing authentication techniques