臺灣博碩士論文加值系統

近年來電腦圖像技術力不斷在進步，若遭有心人士利用在竄改身分、偽造醜聞照片之類不法應用，後果將不堪設想，這使得辨識影像真偽成為現代科技的一門重大課題。在本論文中，藉由一種由去馬賽克演算法所造成的週期性特徵，提出一個辨識真實影像PIM和電腦圖像PRCG的理論，這種週期性現象在傅立葉空間特徵中測量為一種特徵，並結合一簡要的分類器去分辨真實影像和電腦影像。基於此理論的演算法組成如下:色彩空間轉換，高通濾波特徵萃取，和分類。實驗結果顯示，提出的理論不但可以偵測真實影像，還能判斷所使用的相機廠牌。與現存的方式比較，提出的理論擁有較高的辨認率，結果也展現出此理論的性能優於現存的方法。

In recent years, computer graphics technology is constantly in progress, if some people employ the technology in crime, such as falsi cation of identity, or forgeries of scandal photos, the consequences would be disastrous. Thus it makes Photorealistic computer graphics (PRCG) detection to become a major issue of modern technology. In this thesis, a scheme to distinguish photographic images (PIM) from PRCG images by using a periodic phenomenon of variance of pixel values resulting from demosaicing process is proposed. The periodic phenomenon is measured as a feature in Fourier domain feature and combined with a simple classi er to distinguish PIM images from PRCG images. Based on the phenomenon, the proposed scheme is composed of color space transformation, high-pass ltering feature extraction, and classi cation. Experimental results show that proposed scheme can not only detect PIM images but also determine what brand of a camera is used. Compared with existing methods, the precision and recall rates of proposed scheme are higher. The result shows that the performance of the proposed scheme is better than existing methods.

[1] [Online]. Available: http://bbs.chinaemu.org/read-htm-tid-88418.html
[2] E. H. Adelson and J. R. Bergen, The plenoptic function and the elements of early vision, Computation Models of Visual Processing, M. Landy and J. A. Movshon, Eds. Cambridge, MA: MIT Press, 1991, pp. 3-20.
[3] B. E. Bayer, Color imaging array, U.S. Patent 3 971 065, 1976.
[4] W. Chen, Y. Shi, and G. Xuan, Identifying computer graphics using hsv color model and statistical moments of characteristic functions, in Multimedia and Expo, 2007 IEEE International Conference on, july 2007, pp. 1123 -1126.
[5] J. A. Ferwerda, Three varieties of realism in computer graphics, Proc. SPIE Human Vision and Electronic Imaging, San Jose, CA, 2003, pp. 290V297.
[6] M. K. C. G. S. Lin and S. T. Chiu, A feature-based scheme for detecting and classifying video-shot transitions based on spatio-temporal analysis and fuzzy classi cation, International Journal of Pattern Recognition and Arti cial Intelligence, vol. 23, no. 6, pp. 11791200, 2009.
[7] M. F. C. D. P. G. G. W. Meyer, H. E. Rushmeier and K. E. Torrance, An experimental evaluation of computer graphics imagery, ACM Trans. Graph., vol. 5, no. 1, pp. 30-50, Jan. 1986.
[8] A. Gallagher, Detection of linear and cubic interpolation in jpeg compressed images, in Computer and Robot Vision, 2005. Proceedings. The 2nd Canadian
Conference on, may 2005, pp. 65 - 72.
[9] A. Gallagher and T. Chen, Image authentication by detecting traces of demosaicing, in Computer Vision and Pattern Recognition Workshops, 2008. CVPRW'08. IEEE Computer Society Conference on, june 2008, pp. 1 -8.
[10] M. Grossberg and S. Nayar, What is the space of camera response functions? in Computer Vision and Pattern Recognition, 2003. Proceedings. 2003 IEEE Computer Society Conference on, vol. 2, june 2003, pp. II - 602-9 vol.2.
[11] G. Healey and R. Kondepudy, Radiometric ccd camera calibration and noise estimation, Pattern Analysis and Machine Intelligence, IEEE Transactions on, vol. 16, no. 3, pp. 267 -276, mar 1994.
[12] T. Ianeva, A. de Vries, and H. Rohrig, Detecting cartoons: a case study in automatic video-genre classi cation, vol. 1, pp. I - 449-52 vol.1, july 2003.
[13] J. F. J. Lukas and M. Goljan, Digital camera identi cation from sensor noise, IEEE Trans. Inform. Sec. Forensics, vol. 1, no. 2, pp. 205V214, June 2006.
[14] M. Levoy and P. Hanrahan, Light eld rendering, ACM SIGGRAPH, New Orleans, LA, 1996, pp. 31V42.
[15] S. Lyu and H. Farid, How realistic is photorealistic? Signal Processing, IEEE Transactions on, vol. 53, no. 2, pp. 845 - 850, feb. 2005.
[16] T. T. Ng, Statistical and geometric methods for passive-blind image forensics, Ph. D. Research Work Columbia University,2007.
[17] T. T. Ng and S. F. Chang, Identifying and pre ltering images: Distinguishing between natural photography and photorealistic computer graphics, IEEE SIGNAL
PROCESSING MAGAZINE pp.49- 58,MARCH 2009.
[18] J. Nicodemus, F.E. Richmond, H. J.J., I. Ginsberg, and T. Limperis, Geometric considerations and nomenclature for reflectance, Monograph 160, National Bureau of Standards (US).
[19] A. Pentland, Image and vision computing, On describing complex surface shapes (1985). 3(4):153-162.
[20] P. Sutthiwan, X. Cai, Y. Shi, and H. Zhang, Computer graphics classi cation based on markov process model and boosting feature selection technique, in Image Processing (ICIP), 2009 16th IEEE International Conference on, nov. 2009, pp. 2913 -2916.
[21] Y. H. L. X. T.T. Ng, S.F. Chang and M. Tsui, Physics-motivated features for distinguishing photographic images and computer graphics, ACM Multimedia, Singapore, 2005, pp. 239V248.
[22] Y. Wang and P. Moulin, On discrimination between photorealistic and photographic images, IEEE Int. Conf. Acoustics, Speech, and Signal Processing(ICASSP), Toulouse, France, 2006.