Identifying Image Modifications using DCT and JPEG Quantization Technique

Abstract

The volume of images and the videos are being shared on various social media platform are huge and there are cybercrimes happening in these areas. Hence, identification of the main source of social network, based on the images uploaded or downloaded from social network has become an important activity in the multimedia forensic analysis. When media shared on social network there are possibilities of exploiting different pattern embedding in image content by social network. To make it easier system is proposed with discrete cosine transform (DCT) method and JPEG Quantization to identify the changes in shared images on social networks applications. The quantization used in JPEG compression is used to help in separating the images that have been processed by software. DCT finds the pixel value of the blur images and it is easier in implementation. The combination of DCT and JPEG quantization can give the better accuracy rate which helps in finding the images shared or the source of the images.