Abstract
In this paper we propose a method for automatic selection of relevant filter responses which are used for extraction of latent images from printed documents containing means of protection. In other words, it constitutes a part of the latent image extraction technique developed earlier, which is based on convolving an analyzing document image with an adaptively constructed Gabor filter bank. Previously selection a set of relevant filter responses was performed manually. Therefore the motivation of this paper is to design an automatic procedure for this stage. The proposed procedure consists in joint classification of both magnitude and phase parts of filter responses using specific feature sets. The results of experimental investigations showing high precision of the method are given.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Costantini, M.: A novel phase unwrapping method based on network programming. IEEE Trans. Geosci. Remote Sens. 36(3), 813–821 (1998)
Dempster, A.P., Laird, N.M., Rubin, D.B.: Maximum likelihood from incomplete data via the EM algorithm. J. R. Stat. Soc. Ser. B (Methodol.) 39(1), 1–38 (1977)
Ghiglia, D.C., Pritt, M.D.: Two-Dimensional Phase Unwrapping: Theory, Algorithms, and Software, vol. 4. Wiley, New York (1998)
James, G., Witten, D., Hastie, T., Tibshirani, R.: An Introduction To Statistical Learning, vol. 112. Springer, New York (2013)
Movellan, J.R.: Tutorial on gabor filters. Open Source Document (2002)
van Renesse, R.L.: Hidden and scrambled images: a review. Proc. SPIE 4677, 333–348 (2002)
Sergeyev, V., Fedoseev, V.: Extraction of latent images from printed media. In: Proceedings of SPIE, vol. 9875, pp. 98750X–98750X-5 (2015). http://dx.doi.org/10.1117/12.2228405
Sergeyev, V., Fedoseev, V.: A method for extracting watermarks from textured printed documents. Comput. Opt. 38(4), 825–832 (2014)
Acknowledgements
This work was supported by RFBR (project No. 15-07-05576) and the Russian Ministry of Education and Science by President’s grant MK-4506.2015.9, state contract 2014/198 (code 2298) and Project 5-100.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing AG
About this paper
Cite this paper
Fedoseev, V., Mishkina, E. (2016). Selection of Relevant Filter Responses for Extraction of Latent Images from Protected Documents. In: Chmielewski, L., Datta, A., Kozera, R., Wojciechowski, K. (eds) Computer Vision and Graphics. ICCVG 2016. Lecture Notes in Computer Science(), vol 9972. Springer, Cham. https://doi.org/10.1007/978-3-319-46418-3_46
Download citation
DOI: https://doi.org/10.1007/978-3-319-46418-3_46
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-46417-6
Online ISBN: 978-3-319-46418-3
eBook Packages: Computer ScienceComputer Science (R0)