An efficient computational approach to balance the trade-off between image forensics and perceptual image quality

Shashidhar TM, K B Ramesh

Abstract


The increasing trends of image processing applications play a very crucial role in the modern-day information propagation with the ease of cost effectiveness. As image transmission or broadcasting is the simplest form communication which determines easy, fastest and effective way of network resource utilization, thereby since past one decade it has gained significant attention among various research communities. As most of the image attributes often contains visual entities corresponding to any individual, hence, exploration and forging of such attributes with malicious intention often leads to social and personal life violation and also causes intellectual property right violation when social media, matrimonial and business applications are concerned. Although an extensive research effort endeavored pertaining to image forensics in the past, but existing techniques lack effectiveness towards maintaining equilibrium in between both image forensics and image quality assessment performances from computational viewpoint. Addressing this limitation associated with the existing system, this proposed study has come up with a novel solution which achieves higher degree of image forensics without compromising the visual perception of an image. The study formulates an intelligent empirical framework which determines cost-effective authentication of an image object from both complexity and quality viewpoint. Finally, the study also presented a numerical simulation outcome to ensure the performance efficiency of the system.

Keywords


Image; Digital Image Processing; Image Forensics; Image Perceptual Quality; Un Supervised Learning



DOI: http://doi.org/10.11591/ijece.v9i5.pp%25p
Total views : 40 times

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.