許多早期拍攝之電影、紀錄片、新聞等膠卷影片，在放映機機械操作及無妥善保存的環境下，造成嚴重直條紋的刮痕或是因受潮而產生不規則的污點。為了永久儲存，數位化及其修復技術變得相當重要。礙於傳統修復方式需要大量人力與設備，故以數位方式修復受損之影片資料成為重要的工作。
在本論文中，針對膠卷影片中的刮痕與污點雜訊提出一套自動修復系統。首先利用空間上的資訊偵測出刮痕的位置與寬度，並提出一個多項式近似修復模型。接著以動量估測來偵測污點雜訊的出現，並使用時間上的資訊來達到移除的目的。由實驗結果得知，此方法可以有效且自動的修復膠卷影片之受損區域，節省不必要的人力與時間。

There are many valuable videos such as movies, documentaries, or newsreels which were taken by plastic films in early years, being corrupted by serious vertical line scratches or irregular stain noises caused by mechanical operation in film projection and inappropriate preservation corrupted by damp environments. For permanent storage, the digitization and restoration of the films become more and more important nowadays. Due to the traditional approach requires a lot of manpower and equipments, the automatic restoration techniques have become an important work for recovering the damaged films.
In this thesis, we propose an automatic restoration system for line scratches and stain noises in films. First, we use spatial information to detect the position and width of line scratches and then restore them by the proposed polynomial approximation model. Second, we use motion estimation to detect the blotch noises and remove them by utilizing temporal information. Experimental results show that the proposed method can restore the damaged areas in films successfully and effectively, which can save lots of unnecessary manpower and time.

[1] V. Bruni, and D. Vitulano, “A Generalized Model for Scratch Detection,” IEEE Transactions on Image Procesing, Vol. 13. No. 1, January 2004.

[2] A. Kokaram, “Detection and Removal of Line Scratch in Degraded Motion Picture Sequences,” VIII European Signal Processing Conference (EUSIPCO), Trieste, Italy, pp. 5-8, Sept. 1996.

[3] S. Müller, J. Bühler, S. Weitbruch, C. Thebault, I. Doser, and O. Neisse, “Scratch Detection Supported by Coherency Analysis of Motion Vector Fields,” IEEE International Conference on Image Processing, pp.89-92, Nov. 2009.

[4] M. K. Gullu, O. Urhan, and S. Erturk, “Scratch Detection via Temporal Coherency Analysis and Removal Using Edge Priority Based Interpolation,” in IEEE International Symposium on Circuits and System, 2006.

[5] K. tai Kim, and E. yi Kim, “Automatic Film Line Scratch Removal System Based on Spatial Information,” in IEEE International Symposium on Consumer Electronics, pp. 1-5, 2007.

[6] K. tai Kim, and E. yi Kim, “Film Line Scratch Detection Using Neural Network and Morphological Filter,” in IEEE Conference on Cybernetics and Intelligent Systems, pp.1007-1011, Sept. 2008.

[7] T. Bretschneider, O. Kao, P.J. Bones, “Removal of Vertical Scratches in Digitised Historical Film Sequences Using Wavelet Decomposition,” Image and Vision Computing New Zealand (IVCNZ), Dunedin, New Zealand, pp. 38-43, Nov. 2000.

[8] J. Xu, J. Guan, X. Wang, J. Sun, G. Zhai, and Z. Li, “An OWE-based Algorithm for Lin Scratches Restoration in Old Movies,” in IEEE International Symposium on Circuits and Systems, New Orleans, LA, pp.3431-3434, May 2007.

[9] A. C. Kokaram, P. J. W. Rayner. “A System for the Removal of Impulsive Noise in Image Sequences, ” International Conference on SPIE Visual Communications in Image Processing. pp.322-331, 1992.

[10] A. C. Kokaram, R. D. Morris, W. J. Fitzgerald, and P. J. W. Rayner, “Detection of Missing Data in Image Sequences,” IEEE Trans. Image Processing, vol. 4, no. 4, pp. 1496–1508, Nov. 1995.

[11] M. J. Nadenau and S. K. Mitra, “Blotch and Scratch Detection in Image Sequences Based on Rank Ordered Differences, ” Proceedings of the Fifth International Workshop on Time-Varying Image Processing, pp. 27–35, Sept. 1996.

[12] Sang-Churl Nam, Abe M., and Kawamata M., “Fast and Efficient MRF-based Blotch Detection Algorithm for Degraded Film Sequences” Intelligent Signal Processing and Communication Systems, pp.100-103, 2007.

[13] V.R. Vijaykumar, P.T. Vanathi, P. Xanagasabapathy, and D. Ebenezer, “An Efficient Algorithm to Remove Scratches Stripes and Blotches in Still Images,” International Conference on Intelligent and Advanced Systems, 2007.

[14] W.K. Pratt. “Digital Image Processing,” Second Edition, John Wiley and Sons, 1991.

[15] J. Canny, “A Computational Approach to Edge Detection”, IEEE Trans. Pattern Analysis and Machine Intelligence, pages 679-698, November 1986.

[16] Z. Zhang, H. Cui, H. Lu, R. Chen, and Y. Yan, “A Colorization Method Based on Fuzzy Clustering and Distance Transformation,” International Congress on Image and Signal Processing, 2009.

[17] T. Madhusudhan, and Alwyn Roshan Pais, “Generation of Super-Resolution Video From Low Resolution Video Sequences: A Novel Approach,” International Conference on Computational Intelligence and Multimedia Applications, pp.255-232, 2007.

[18] M. S. Lee, M. Y. Shen, and C.-C. J. Kuo, “A Content-Adaptive Up-Sampling Technique for Image Resolution Enhancement,” Intelligent Information Hiding and Multimedia Signal Processing, pp. 87- 90, 2007.