Open AccessAn Improved Forgery Detection Method for Images
Author(s) -
Tiss Tom,
Durgesh Nandini,
S. Prince Mary,
B. Ankayarkanni
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/590/1/012032
Subject(s) - artificial intelligence , chrominance , computer science , pixel , computer vision , pattern recognition (psychology) , hash function , feature (linguistics) , feature extraction , feature detection (computer vision) , image (mathematics) , luminance , image processing , linguistics , philosophy , computer security
A forgery image detection system is introduced to detect the forgery in images using global, local and pixel based features. Global features will consider the whole images which include luminance and chrominance features. These global features can be extracted by using Zernike moments. The local feature consists of descriptors of multiple interest points. Image authentication is done by using hash method. The set of images will be trained in the database earlier itself. Then the test image and reference image will be compared based on pixels. By analyzing the hash distance the system can identify the test image is forged or not. As an improvement, pixel based feature extraction is used. Pixel based feature extraction is carried out by using supervised classification algorithm.