簡易檢索 / 詳目顯示

回結果列表
研究生： 林柏瑋
Po-Wei Lin
論文名稱： 使用階層式支撐向量機分類器過濾不當影片
Objectionable Video Filtering Using Hierarchical SVM Classifier
指導教授： 吳怡樂
Yi-Leh Wu
口試委員: 林彥君
Yen-Chun Lin
唐政元
Cheng-Yuan Tang
陳延禎
Yen-Jen Chen
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： 2008
畢業學年度： 96
語文別： 英文
論文頁數： 46
中文關鍵詞： 不當影片分類 、基於影片內容的影片分析 、支撐向量機 、快速傅立葉轉換
外文關鍵詞： objectionable video classification, content-based video analysis, support vector machine, fast Fourier transform
相關次數： 點閱：345  下載：2 
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
  •  上一筆

    • 隨著p2p軟體的盛行，人們接觸到令人反感的資訊的機會變多了。 這些令人反感的資訊是不適合一般大眾及幼童並會對心靈產生傷害的，所以如何阻絕或過濾這些不當的資訊已經成為一個重要的議題。色情影片 是令人反感的資訊的其中之一，其內容充斥著色情及淫穢的影像片段，有許多論文已經在過濾不當圖片做了研究，但有關於過濾不當影像的文獻在目前還是相當的少。

    在這篇論文中，我們提出了一個具高準確度的不當影片分類系統。我們藉由從影片中擷取圖片及兩層式的支撐向量機架構來分類不當影片。在第一層中，我們採用了傳統的影像分類器去對影片圖片作分類。在第二層中，我們提出了幾個方法來分析圖片過濾器的結果並產生特徵，利用這些特徵並透過第二層的支撐向量機去分類不當影片，藉此達到過濾的效果。 我們將會證明藉由我們的方法，即使使用不完美的圖片過濾器，我們所提出的二階層式分類器仍然能呈現令人滿意的結果。 最後，我們的實驗結果將會證明我們的方法在現實世界中也是令人滿意和實用的。

    As the P2P software prevails on the internet, people contact with the objectionable information more often than before. Because the objectionable information is not suitable for the minors, how to block or filter the objectionable information has became a critical issue. One of the major objectionable information is pornographic videos. Many studies have been researched on filtering objectionable images, but few studies have been investigated on filtering objectionable videos.
    In this paper, we propose a high accuracy objectionable video classifying system. We extract frames from videos to classify objectionable videos with a two-tier SVM classifier. In the first tier, we adopt the traditional image classifier to classify video frames. In the second tier, we propose methods to analyze the classification results from the image classifier in the first tier and generate features to classify videos with a second tier SVM classifier. We show that even if the image classifier in the first tier is far from perfect the proposed two-tier classifier can still produce satisfactory result in classifying videos. Finally, our experiment results suggest that the proposed methods are promising and applicable in real world situations.

    1. Introduction.............................................................1 2. Image Filter.............................................................3 3. Support Vector Machines (SVMs)...........................................8 4. The Classification Process...............................................10 5. Feature Generation Method................................................12 5.1. Histogram...........................................................12 5.2 Continuous Property..................................................13 5.3 Transforming Time Series Data into Symbolic Sequences Data...........14 5.4 Fast Fourier Transform...............................................14 6. Experiments..............................................................15 6.1 Sample Data Description..............................................15 6.2 Experiment - Feature Verification....................................15 6.2.1 Histogram Verification.........................................16 6.2.2 Continuous Property Verification...............................19 6.2.3 Transforming Time Series Data into Symbolic Sequences Data Verification................................................................20 6.2.4 Fast Fourier Transform Verification............................22 6.3 Experiment – Accuracy Verification..................................25 7. Conclusion and Future Work...............................................28 Appendix....................................................................29 References..................................................................45

    [1] Yi-Leh Wu, Edward Y. Chang, Kwang-Ting Cheng, Cheng-Wei Chang, Chen-Cha Hsu, Wei-Cheng Lai, and Ching-Tung Wu. MORF: A Distributed Multimodal Information Filtering System. In Proceedings of the third IEEE Pacific-Rim Conference on Multimedia (PCM 2002), Pages 279-286, Taiwan, December 2002.
    [2] Chih-Chung Chang and Chih-Jen Lin. LIBSVM: A library for support vector machines. 2001. Software is available at http://www.csie.ntu.edu.tw/~cjlin/libsvm/
    [3] Hogyun Lee, Seungmin Lee, and Taekyong Nam. Implementation of High Performance Objectionable Video Classification System. In Proceedings of the 8th International Conference on Advanced Communication Technology (ICACT 2006), vol. 2, pages 959-962, Korea, February 2006.
    [4] Qian Wang, Wei-Ming Hu, Tie-Niu Tan. Detecting Objectionable Videos. ACTA AUTOMATICA SINICA, vol.31, no.2, pages 280-286, Beijing, March 2005.
    [5] W. T. Cochran, J.W. Cooley, D. L. Favin, H. D. Helms, R.Kaenel,W.W. Lang, G. C. Maling, D. E. Nelson, C. M. Rader, and P. D.Welsh. What is the fast Fourier transform? IEEE Trans. on Audio Electroacoustics, vol. 15, no. 2, pages 45-55, June 1967.
    [6] Kai Ou-Yang, Wenyan Jia, Pin Zhou, and Xin Meng. A new approach to transforming time series into symbolic sequences. In Proceedings of the 1st Joint BMES/EMBS Conference, vol. 2, pages 974, Atlanta, GA, USA, October 1999.
    [7] Fu-Lai Chung, Tak-Chung Fu, Vincent Ng, and Robert W. P. Luk. An evolutionary approach to pattern-based time series segmentation. IEEE Trans. on Evolutionary Computation, vol.8, no. 5, pages 471-489, October 2004.
    [8] Seungwan Han, Chiyoon Jeong, Taekyong Nam. Multi-layer objectionable video classification system using local-global information. In Proceedings of the 9th WSEAS International Conference on Computers, Athens, Greece, July 2005.
    [9] S. Vakkalanka, C. Krishna Mohan, R. Kumaraswamy, and B. Yegnanarayana. Combining multiple evidence for video classification. In Proceedings of the International Conference on Intelligent Sensing and Information Processing, Pages 187-192, January 2005.
    [10] N. Watcharapinchai, S. Aramvith, S. Siddhichai, and S. Marukatat. A discriminant approach to sports video classification. In Proceedings of International Symposium on Communications and Information Technologies ( ISCIT '07), pages 557-561, October 2007.
    [11] Yu-Fei Ma, Hong-Jiang Zhang. Motion pattern based video classification using support vector machines. In Proceedings of the 2002 IEEE International Symposium on Circuits and Systems (ISCAS 2002), vol. 2, pages II-69 - II-72, May 2002.
    [12] Time Series Analysis, http://www.statsoft.com/textbook/sttimser.html
    [13] Open Computer Vision Library (OpenCV), http://opencvlibrary.sourceforge.net/

    QR CODE