International Journal of Scientific and Research Publications

IJSRP, Volume 6, Issue 8, August 2016 Edition [ISSN 2250-3153]


Detecting Forgery in Duplicated Region Using Superpixel Segmentation and Feature Point Matching
      Hasbi Shaji, Theresa Jose
Abstract: The region duplication image forgery detection is the one of the major problems in the field of digital image forensics In this paper describes a roboust forgery detection scheme using adaptive oversegmentation and feature point matching It include both block based and keypoint based forgery detection methods. In proposed method, an adaptive oversegmentation method is to segment the host image into non overlapping and irregular blocks called Image Blocks(IB),Then apply Scale Invariant Feature Transform(SIFT) in each block to extract the SIFT feature as block feature(BF) .Then ,the block features are matched with one another to determine labeled feature points(LFP).The indicate the suspected forgery region .Using this approach ,for each image we can determine appropriate block initial size to enhance the accuracy of forgery detection ,and also reduce computational expenses .The proposed forgery region extraction algorithm replace the feature points with small superpixels as feature block and the neighboring feature blocks with local color features that are similar to feature blocks are merged to generate the merged region .The morphological operation are applied to the merged regions to generate the detected forgery region .The proposed copy-move forgery detection scheme can achieve better results compared to existing copy-move forgery detection methods.

Reference this Research Paper (copy & paste below code):

Hasbi Shaji, Theresa Jose (2018); Detecting Forgery in Duplicated Region Using Superpixel Segmentation and Feature Point Matching; Int J Sci Res Publ 6(8) (ISSN: 2250-3153). http://www.ijsrp.org/research-paper-0816.php?rp=P565734
©️ Copyright 2011-2022 IJSRP - All rights reserved. Use of this web site signifies your agreement to the terms and conditions.