Wavelet-Based Multiresolution Features for Detecting Duplications in Images

作者: Noboru Ohnishi , Md. Khayrul Bashar , Hiroaki Kudo , Tetsuya Matsumoto , Yoshinori Takeuchi

DOI:

关键词: Artificial intelligenceImage (mathematics)MathematicsMatching (graph theory)SortingFeature (computer vision)Block (data storage)Lexicographical orderComputer visionRepresentation (mathematics)WaveletPattern recognition

摘要: Duplication of image regions is a common method for manipulating original images using typical software like Adobe Photoshop. In this study, we propose wavelet based feature representation scheme detecting duplicated in images. This technique works by first applying multi-resolution decomposition to small fixed-sized blocks. Normalized coefficients are then stacked into vector an order from lower higher frequencies. kind appears robust block matching. Duplicated detected lexicographically sorting all the blocks and threshold desired frequency offsets blockcoordinates. A semi-automatic that detects accurate number also proposed. Initial experiments with set natural having show impressive results compared linear PCA representation.

参考文章(6)
Hany Farid, None, Exposing digital forgeries in scientific images Proceeding of the 8th workshop on Multimedia and security - MM&Sec '06. pp. 29- 36 ,(2006) , 10.1145/1161366.1161374
S.G. Mallat, A theory for multiresolution signal decomposition: the wavelet representation IEEE Transactions on Pattern Analysis and Machine Intelligence. ,vol. 11, pp. 674- 693 ,(1989) , 10.1109/34.192463
A.C. Popescu, H. Farid, Exposing digital forgeries in color filter array interpolated images IEEE Transactions on Signal Processing. ,vol. 53, pp. 3948- 3959 ,(2005) , 10.1109/TSP.2005.855406
Peter E. Hart, Richard O. Duda, Pattern classification and scene analysis A Wiley-Interscience Publication. ,(1973)