RUS  ENG
Full version
JOURNALS // Computer Optics // Archive

Computer Optics, 2013 Volume 37, Issue 3, Pages 360–367 (Mi co749)

IMAGE PROCESSING, PATTERN RECOGNITION

The algorithm for copy-move detection on digital images

N. I. Glumova, A. V. Kuznetsova, V. V. Myasnikovba

a Image Processing Systems Institute of the RAS
b S.P. Korolyov Samara State Aerospace University (National Reseach University)

Abstract: In the paper, we propose two algorithms for undistorted copy-move regions’ (coincident rectangular fragments) detection on digital images. Both algorithms represent pixels’ data as the hash function value, which is built using two different mathematical principles: probability-theoretical and number-theoretical. The comparison of the proposed algorithms, as well as recommendations for their use is also proposed in this paper.

Keywords: digital image, authentication, copy-move region, hash table, absolute frequencies, chinese remainder theorem.

Received: 27.06.2013



© Steklov Math. Inst. of RAS, 2024