Using Sab-Iomha for an Alpha Channel based Image Forgery Detection

Digital images are a very popular way of transfer-ring media. However, their integrity remains challenging because these images can easily be manipulated with the help of software tools and such manipulations cannot be verified through a naked-eye. Although there exist some techniques to validate digital images, but in practice, it is not a trivial task as the existing approaches to forgery detection are not very effective. Therefore, there is need for a simple and efficient solution for the challenge. On the other hand, digital image steganography is the concealing of a message within an image file. The secret message can be retrieved afterwards by the author to check the image file for its veracity. This research paper proposes Sabiomha, an image forgery technique that make use of image steganography. The proposed technique is also supported by a software tool to demonstrate its usefulness. Sabiomha works by inserting an invisible watermark to certain alpha bits of the image file. The watermark we have used to steganograph an image is composed of a combination of text inputs the author can use to sign the image. Any attempts to tamper the image would distort the sequence of the bits of the image pixel. Hence, the proposed technique can easily validate originality of a digital image by exposing any tampering. The usability of our contribution is demonstrated by using the software tool we developed to automate the proposed technique. The experiment which we performed to further validate our technique suggested that Sabimoha could be flawlessly applied to image files.

@article{Bhatti2018,
  title     = {Using Sab-Iomha for an Alpha Channel based Image Forgery Detection},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {9},
  number    = {8},
  year      = {2018},
  publisher = {The Science and Information Organization},
  author    = {Muhammad Shahid Bhatti and Syed Asad Hussain and Abdul Qayyum and Abdul Karim Shahid and Muhammad Usman Akram and Sajid Ibrahim Hashmi},
  doi       = {10.14569/IJACSA.2018.090849},
  url       = {https://doi.org/10.14569/IJACSA.2018.090849}
}