Detection of copy move forgery using Legendre Moments
| Autor: | Samet Aymaz, Seyma Aymaz, Guzin Ulutas |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
business.industry
Computer science Gaussian Feature vector Feature extraction 020207 software engineering 02 engineering and technology Facial recognition system Image (mathematics) Euclidean distance symbols.namesake Pattern recognition (psychology) 0202 electrical engineering electronic engineering information engineering symbols 020201 artificial intelligence & image processing Computer vision Artificial intelligence business Legendre polynomials Transform coding |
| Zdroj: | SIU |
| DOI: | 10.1109/siu.2016.7495942 |
| Popis: | Today there are different manipulations on images. Copy move forgery is one of them. Copy move forgery means that; any part of an image is taken and added to another part of the same image in a professional way. To detect forgeries, image is divided into overlapped blocks and feature vectors are generated for every block. Feature vectors are tips about copy move forgery. In this article, we used Legendre Moments to detect copy move forgeries. Legendre moments are used in different areas like face recognition, pattern recognition etc. But never used to detect copy move forgery before. Feature vectors are generated using Legendre moments and this vectors are compared using Euclidean distance to find forgery areas. Experimental results show that proposed method can detect single and multi copy-move forgery, and also method can detect forgeries when image is post processed using Jpeg compression and Gaussian bluring. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|