Zernike Moment-Based Approach for Detecting Duplicated Image Regions by a Modified Method to Reduce Geometrical and Numerical Errors
In the paper, the approach is focused on the Zernike Moment-based model of ROI image (Region of Interest) and its parameters for an efficient image processing in the forensic issue. By considering the factors affecting the identification of an duplicated image, the change of ROI’s size is determined through the proposed algorithm. The proposed technique has shown a good improvement in reducing significantly Geometrical Errors (G.E) and Numerical Errors (N.E) performed better than that of the Zernike-based traditional technique. The duplicated detection program has been written by C++ and supporting OpenCV and Boost libraries that help to verify the images authentication.
KeywordsZernike moments Geometrical errors Numerical errors Geometric moments Region of interest (ROI) FLANN library-Fast library for approximation nearest neighbors
Unable to display preview. Download preview PDF.
- 1.Popescu, A.C., Farid, H.: Exposing Digital Forgeries in Color Filter Array Interpolated Images. IEEE Trans. Signal Processing 53(10), October 2005Google Scholar
- 4.Chong, C.-W.: A formulation of a new class of continuous orthogonal moment invariants, and the analysis of their computational aspects. Ph.D thesis, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia, May 2003Google Scholar
- 7.Papakostas, G.A., Boutalisn, Y.S., Papaodysseus, C.N., Fragoulis, D.K.: Numerical error analysis in Zernike moments computation. Image and Vision Computing 24 (2006)Google Scholar
- 11.Arai, K., Barakbah, A.R.: Hierarchical K-means: an algorithm for centroids initialization for K-means. Reports of the Faculty of Science and Engineering, Saga University, pp. 840–8502 Saga Prefecture, Saga, Japan, 36, (1), (2007)Google Scholar
- 12.Liao, S.: Image analysis by moments. Ph.D thesis, University of Manitoba, Winnipeg, Manitoba, Canada (1993)Google Scholar
- 14.Thacker, N.A., Aherne, F.J., Rockett, P.I.: The Bhattacharyya Metric as an Absolute Similarity Measure for Frequency Coded Data. TIPR 1997, pp. 9–11, Prague, June, 1997Google Scholar