원문정보
초록
영어
Powerful image editing tools are very common these days. By use of such tools, image tampering is also becoming common. It is becoming very common to remove or add some information on the images mostly for the purpose of some forgery. The focus of this paper is to detect “copy move” type of forgery. In this type of forgery, one portion of image is copied and pasted somewhere else in the same image. The motive of this type of image forgery is to hide some important features from the original image. As may be the case, in an criminal evidencing system or to show large crowd in a scene having sparse population etc. In this paper, we propose a method which is more efficient and reliable in comparison to previous methods. This method is good enough even for some manipulations/attacks like for scaling, rotation, Gaussian noise, smoothing and JPEG compression etc.
목차
1. Introduction
2. Problem definition and Motivation
3. Proposed Method
1. Discrete Wavelet Transform:
2. Sub Blocking:
3. Lexicographical Sorting:
4. Detection of suspected copied regions:
5. Neighbor Shift Matching:
6. Detection of Duplicated Region
7. Detection of Copy-Rotate-Move (CRM) Forgery
4. Experimental Results
5. Conclusion
References