Enhancing the Protection of Digital Images against Tampering Using Joint Source Channel Coding

With the rapid development of multimedia products, digital images can be modified easily. This questions the integrity of images. To provide the integrity and authenticity to images, digital watermarking can be effectively used. In addressing the problem of image tampering these algorithms must be enough to accommodate two functionalities such as tampering detection and error concealment. Both these functions can be achieved by using check bits and reference bits as watermark data. Check bits carry information for the tampering detection, whereas reference bits carry information about the whole image which can be used for recovering the lost information. Check bits are MD5 hash output which can detect the tampering that also results in tampering localization. The generation of reference bits can be done by analog joint source channel coding of the image. Analog joint source channel coding directly maps analog image symbols to channel symbols without requiring source coding followed by channel coding. This reduces the computational complexities. This paper presents a novel and effective method for tampering detection and image self recovery.