An Effective Image Detection Algorithm for USM Sharpening Based on Pixel-Pair Histogram
- 1.7k Downloads
USM sharpening is a popular method for enhancement of image quality, detection of image sharpening has attracted much attention in recent years. A novel image sharpening detection algorithm is proposed in this paper. In the scheme, different from some image forensic schemes, which used Cb or Cr channel of YCbCr color model to extract image features for forensics, in this paper, color images are firstly transformed into the YCbCr model, then the luminance channel of YCbCr color model is selected to extract pixel-pair histogram features based on four directional differential matrixes, these features within some threshold scope constitute the final image features. LIBSVM is used to implement classification for real and sharpened image. Widely used UCID database is employed to conduct test with various sharpening strength and range. Experimental results show that the proposed algorithm has superior performance; extensive comparisons with some existing algorithms show that it outperforms state-of-art methods investigated, even if the sharpening intensity is very weak (σ = 0.3).
KeywordsUSM sharpening Pixel-pair histogram Image forensic
The work was supported by the Program of Natural Science Fund of Tianjin, China (Grant No. 16JCYBJC15700).
- 14.Cao, G., Zhao, Y., Ni, R.: Detection of image sharpening based on histogram aberration and ringing artifacts. In: Proceedings of IEEE International Conference on Multimedia and Expo (ICME), pp. 1026–1029 (2009)Google Scholar
- 18.Gu, Y., Wang, S., Lin, X., Sun, T.: USM sharpening detection based on sparse coding. In: Proceedings of 2016 International Conference on Digital Image Computing: Techniques and Applications (DICTA), pp. 1–5 (2016) Google Scholar
- 22.Schaefer, G., Stich, M.: UCID—an uncompressed colour image database. Storage Retr. Methods Appl. Multimed. 5307, 472–480 (2003)Google Scholar