SURF Based Copy Move Forgery Detection Using kNN Mapping

Paul, Kelvin Harrison and Akshatha, K R and Karunakar, A K and Seshadri, Sharan (2019) SURF Based Copy Move Forgery Detection Using kNN Mapping. In: Computer Vision Conference, 25/04/2019, Las Vegas, Nevada, United states of America.

[img] PDF
1389.pdf - Published Version
Restricted to Registered users only

Download (2MB) | Request a copy

Abstract

Digital images can be edited with the help of photo editing tools to improve or enhance the image quality. On the other hand, digital images can also be subject to manipulations which can alter the visual information being conveyed by the image. The forged images can also be used to spread false information through various media platforms and in some cases may be surreptitiously used as false evidence in a court of law. Therefore, it is crucial to test the authenticity of such images and ensure that it does not spread falsified information. One of the most common types of forgery being used today is copy-move forgery in which one part of the image is copied and placed over another part of the same image in order to either conceal certain details or multiply certain features seen in the original image. This work introduces a method of detecting copy-move forgery in digital images using speeded-up robust features (SURF) to extract keypoints from the image and then uses knearest neighbor (kNN) training and mapping to yield accurate matches. The SURF algorithm is capable of performing equally or even exceed the more widely accepted SIFT-based counterparts in terms of ensuring distinctive features, reproducibility, and robustness. As a result, this technique ensures a robust detection of copy move forgery while ensuring lower computational costs compared to the SIFT-based techniques used for the same purpose

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Copy move forgery detection�CMFD�SURF�kNN mapping� Image forensics
Subjects: Engineering > MIT Manipal > Electronics and Communication
Engineering > MIT Manipal > MCA
Depositing User: MIT Library
Date Deposited: 03 Jul 2019 08:52
Last Modified: 03 Jul 2019 08:52
URI: http://eprints.manipal.edu/id/eprint/154098

Actions (login required)

View Item View Item