Downloads: 110 | Views: 326
Research Paper | Computer Science & Engineering | India | Volume 3 Issue 5, May 2014 | Rating: 6.3 / 10
A Joint Forensic System to Detect Image Forgery using Copy Move Forgery Detection and Double JPEG Compression Approaches
Dhara Anandpara | Rohit Srivastava [2]
Abstract: With advent of many powerful editing tools in the digital image processing; image forgery is the big concern today in Digital Forensics Industry. Image forgery can be apply either in single image by coping some region of image and pasting it to another place in the same image or in composite image by combining two or more images together. The focus of my research work is to develop a forensic system to detect both type of forgery within a single place. Many Copy-move Forgery Detection (CMFD) algorithms have been developed to detect forgery within single image but are not robust to geometric transformation. Double JPEG compression is used extensively for localization of regions for composite images forgery such as Image Slicing; In-painting etc. A proposed system is a fusion based system which will allow to detect the image tampering using both techniques i. e. CMFD and DJPG. This gives insights of using both image detection algorithms within same image and in single framework so that detection is evident at single place. A system will compute a likelihood map to indicate the forged area that is accrued due to Copy. To reduce computational cost of system features are extracted from taking the mean value of DCT (discrete cosine transform) coefficients. The proposed scheme is not only robust to copy-move forgery; but also to blurring or nosing adding and with low computational complexities.
Keywords: Digital Image Forensics, image forgery detection, manipulation detection, copy-paste forgery detection, tampering detection
Edition: Volume 3 Issue 5, May 2014,
Pages: 1404 - 1408