| Autor: |
Priya, B., Sri Devi, R., Praveena, R. |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 5/26/2022, Vol. 2464/2506 Issue 1, p1-5, 5p |
| Abstrakt: |
Blind image Forensics has a lot of research areas and Forgery detection is one of the most active area. Existing algorithms follow the principles of block and key-point methods, or combination of them. In the research areas of image processing such as image classification, forensics, hashing and retrieval, when neural networks is applied, it shows significant results compared to the existing approaches. The drawback associated with the existing system is that changing the colour distribution of a block is not possible even if it is compressed or blurred. Color Moments approach has not been employed to detect image forgery in the traditional approaches. A novel forgery detection approach based on neural network is proposed in the paper. The duplicate regions in an image, which exploits statistical features of an image can be obtained using the proposed approach. Mean and variance are used for this purpose here, by splitting the image into pixel blocks. The contribution of each individual block with respect to pixel intensity of the entire image can be found using mean and to find the variation of each pixel from its neighbour, variance is employed. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
