Novel method for image splicing detection

Author

Mushtaq, Saba ; Mir, Ajaz Hussain

Author_Institution

Electron. & Commun. Eng. Dept., Nat. Inst. Of Technol., Srinagar, India

fYear

2014

fDate

24-27 Sept. 2014

Firstpage

2398

Lastpage

2403

Abstract

Splicing is a very common operation performed for image forgery. It involves merging of two or more different images to form a combined image that is significantly different from the original image. In this paper an image splicing detection method based on texture features of spliced image is proposed. The proposed approach calculates grey level run length matrix (GLRLM) texture features for the forged images and original images from CASIA database and DB2 database. The statistical features thus extracted from GLRLM are used for detection of tampering. Support vector machine is used for classification. The proposed algorithm is very effective in detection of splicing forgery as marked by the results.

Keywords

feature extraction; image classification; image colour analysis; image forensics; image texture; matrix algebra; support vector machines; CASIA database; DB2 database; GLRLM texture features; grey level run length matrix texture features; image classification; image forgery; image merging; image splicing detection; image tampering detection; spliced image texture features; statistical feature extraction; support vector machine; Authentication; Databases; Digital images; Feature extraction; Forgery; Splicing; Support vector machines; GLRLM; SVM; image forensics; image forgery; image texture; splicing;

fLanguage

English

Publisher

ieee

Conference_Titel

Advances in Computing, Communications and Informatics (ICACCI, 2014 International Conference on

Conference_Location

New Delhi

Print_ISBN

978-1-4799-3078-4

Type

conf

DOI

10.1109/ICACCI.2014.6968386

Filename

6968386