簡易檢索 / 詳目顯示
| 研究生： |
洪鈺敏 Yu-Min Hung |
|---|---|
| 論文名稱： |
使用卷積神經網路進行基於Bayer CFA模組之新穎相機型號辨認 Novel Bayer CFA Module-Based Camera Model Identification Using Convolutional Neural Networks |
| 指導教授： |
鍾國亮
Kuo-Liang Chung |
| 口試委員: |
鍾國亮
Kuo-Liang Chung 陳建中 Jiann-Jone Chen 范國清 Kuo-Chin Fan 貝蘇章 Soo-Chang Pei 廖弘源 Hong-Yuan Liao |
| 學位類別： |
碩士 Master |
| 系所名稱： |
電資學院 - 資訊工程系 Department of Computer Science and Information Engineering |
| 論文出版年： | 2019 |
| 畢業學年度： | 107 |
| 語文別： | 英文 |
| 論文頁數： | 35 |
| 中文關鍵詞： | Bayer 彩色濾波陣列模組識別 、相機型號辨別 、卷積神經網路 、RGB 全彩影像 |
| 外文關鍵詞： | Bayer color filter array (CFA) module recognition, Camera model identification, Convolution neural networks (CNNs), RGB full-color image |
| 相關次數： | 點閱：279 下載：1 |
| 分享至: |
| 查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
對於一張相機所拍攝的 RGB 全彩影像要辨認其中所繼承之相 機型號，是具有挑戰性的，且於鑑識取證中也具有重要的應用。本篇論文，提出了一個利用卷積神經網路 (CNN) 進行基於 Bayer 彩色濾波陣列 (CFA) 排列模組之新穎且強健的相機型號辨認的方 法。在所提出的方法中，對於一張 RGB 全彩影像，首先遞送到 CNN-based Bayer CFA 排列模組的識別器，之後應用辨認出來的 Bayer CFA 模組排列去驅動 CNN-based 相機型號的辨認方法來識 別 RGB 全彩影像的相機型號，根據我們的實驗，這樣的方式可以產生更高的相機型號辨認的準確性。根據 Dresden 資料集，綜合實驗數據證明了，相對於現有技術的方法，所提出的相機型 號辨別方法優於現有的最新方法。
Identifying the camera model inherited in the captured red-green- blue (RGB) full-color image IRGB is challenging and has important application in forensics. In this paper, we propose a novel and robust Bayer (color filter array) CFA module-based camera model identification method using convolutional neural networks (CNNs). For IRGB, in the proposed method, the CNN-based Bayer CFA module recognizer is delivered first then a Bayer CFA module driven- and CNN-based camera model identification method is applied to identify the camera model of IRGB, leading to higher camera model identification accuracy. Based on the Dresden dataset, the comprehensive experimental data have justified the camera model identification accuracy merit of the proposed method relative to the state-of-the-art methods.
[1] B. E. Bayer, “Color imaging array,” U.S. Patent, 3 971 065, Jul. 20, 1976.
[2] L. Bondi, L. Baroffio, D. Guera, P. Bestagini, E. J. Delp, S. Tubaro, “First steps toward camera model identification with con- volutional neural networks,” IEEE Signal Processing Letters, vol. 24, no. 3, pp. 259-263, Mar. 2017.
[3] C. Chen and M. C. Stamm, “Camera model identification frame- work using an ensemble of demosaicing features,” Information Forensics and Security (WIFS), 2015 IEEE International Work- shop, pp. 1-6, Nov. 2015.
[4] K. S. Choi, E. Y. Lam, and K. K. Y. Wong, “Automatic source camera identification using the intrinsic lens radial distortion,” Optics Express, vol. 14, pp. 11551-11565, Nov. 2006.
[5] A. Dirik, H. Sencar, and N. Memon, “Digital single lens reflex camera identification from traces of sensor dust,” IEEE Trans- actions on Information Forensics and Security, vol. 3, no. 3, pp. 539-552, Sept. 2008.
[6] A. E. Dirik, S. Bayram, H. T. Sencar, and N. D. Memon, “New features to identify computer generated images,” Proceedings of the 2007 IEEE International Conference on Image Processing, pp. IV-433-IV-436, 2007.
[7] T. Gloe and R. Böhme, “The dresden image database for bench- marking digital image forensics,” Journal of Digital Forensic Practice, vol. 3, no. 24, pp. 150-159, 2010.
[8] S. Ioffe and C. Szegedy, “Batch normalization: Accelerating deep network training by reducing internal covariate shift,” Interna- tional Conference on Machine Learning, 2015.
[9] J. J. Jeon, J. S. Hyun, and I. K. Eom, “Estimation of Bayer CFA pattern configuration based on singular value decomposition,” EURASIP Journal on Image and Video Processing, vol. 24, no. 3, pp. 259-263, Jul. 2017.
[10] D. Kiku, Y. Monno, M. Tanaka, and M. Okutomi, “Residual inter- polation for color image demosaicking,” IEEE International Con- ference on Image Processing, pp. 2304-2308, Sep. 2013.
[11] M. Kirchner, “Efficient estimation of CFA pattern configuration
in digital camera images,” Proc. SPIE Conf. Media Forensics and
Security, San Jose, CA, 2010.
[12] M. Kirchner and T. Gloe, “Forensic camera model identification,”
Handbook of Digital Forensics of Multimedia Data and Devices. Chichester, U.K.: Wiley, pp. 329–374, 2015.
[13] Kodak True Color Image Collection. [Online]. Available: http:// r0k.us/graphics/kodak/
[14] J. Lukas, J. Fridrich, and M. Goljan, “Digital camera identifica- tion from sensor pattern noise,” IEEE Transactions on Informa- tion Forensics and Security, vol. 1, no. 2, pp. 205-214, Jun. 2006.
[15] F. Marra, G. Poggi, C. Sansone, and L. Verdoliva, “Evaluation of residual-based local features for camera model identification,” New Trends in Image Analysis and Processing - ICIAP Workshop: BioFor, Genoa, Italy, September 7-8, pp. 11-18, 2015.
[16] R. Ramanath, W. E. Snyder, Y. Yoo, and M. S. Drew, “Color image processing pipeline,” IEEE Signal Processing Magazine, vol. 22, no. 1, pp. 34-43, Jan. 2005.
[17] H. J. Shin, J. J. Jeon, and I. K. Eom, “Color filter array pattern identification using variance of color difference image,” Journal of Electronic Imaging, vol. 26, no. 4, pp. 043015, 2017.
[18] M. C. Stamm, M. Wu, and K. J. R. Liu, “Information Forensics: An Overview of the First Decade,” IEEE Access, vol. 1, pp. 167- 200, 2013.
[19] A. Swaminathan, M. Wu, and K. J. R. Liu, “Digital image foren- sics via intrinsic fingerprints,” IEEE Transactions on Information Forensics and Security, vol. 3, no. 1, pp. 101-117, Mar. 2008.
[20] A. Swaminathan, M. Wu, and K. J. R. Liu, “Nonintrusive com- ponent forensics of visual sensors using output images,” IEEE Transactions on Information Forensics and Security, vol. 2, no. 1, pp. 91-106, 2007.
[21] The available code. Accessed: Dec. 27, 2018. [Online]. Available: ftp://140.118.175.164/CNNsBased_code/
[22] T. H. Thai, R. Cogranne, and F. Retraint, “Camera model identifi- cation based on the heteroscedastic noise model,” IEEE Transac- tions on Image Processing, vol. 23, no. 1, pp. 250-263, Jan. 2014.
[23] L. Zhang, X. Wu, A. Buades, and X. Li, “Color demosaicking by local directional interpolation and nonlocal adaptive threshold- ing,” Journal of Electronic Imaging, vol. 20, no. 2, pp. 023016-1- 023016-16, Apr. 2011.
