跳到主要內容

臺灣博碩士論文加值系統

(34.204.180.223) 您好!臺灣時間:2021/08/05 16:11
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:楊青矗
研究生(外文):Ging-Chu Yang
論文名稱:偵測影像複製區域技術之研究
論文名稱(外文):Copy-Move Forgery Detection in Digital Image
指導教授:黃仲陵黃仲陵引用關係
指導教授(外文):Chung-Lin Huang
學位類別:碩士
校院名稱:佛光大學
系所名稱:資訊學系
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2009
畢業學年度:97
語文別:中文
論文頁數:41
中文關鍵詞:字典排序區塊比對主成份分析奇異值分解
外文關鍵詞:lexicographic orderblock-matchingprincipal component analysisSingular Value Decomposition
相關次數:
  • 被引用被引用:0
  • 點閱點閱:409
  • 評分評分:
  • 下載下載:57
  • 收藏至我的研究室書目清單書目收藏:0
由於影像處理軟體的強大及容易取得、使用,想要編修數位影像或是想竄改數位影像已經是非常容易的事情。一般偵測影像竄改大致來分有二種,第一種是主動式:先在原數位影像置入浮水印,之後就可以利用浮水印是否有被修改過來偵測是否竄改及竄改何處。第二種是被動式:原數位影像沒有置入浮水印就直接遭受竄改。對於竄改數位影像的方式非常多種,其中一般人最容易的竄改方式為複製貼上(copy-paste)。本研究著重於偵測在同一張影像上複製貼上竄改的技術根據文獻,偵測複製貼上竄改先使用重疊的方式把影像切分成許多區塊,再把每一區塊轉為特徴向量,向量即代表此區塊的特徵值,以特徵值去作演算法的偵測運算,運算複雜度取決於區塊的數量及特徴向量的維度。本研究更改了之前學者的偵測作法,不再用相同位移量累加的偵測方式,改用兩階段區塊比對的方式。使用了更小的解析度影像,且只取特徵向量的第一個特徵值。因為比較小的影像有更少的區塊,也降低了比對區塊的數量,影像解析度變小了,理論上區塊間變異變大了,所以特徴向量維度直接取第一個元素。一般人最容易取得的影像格式為JPEG壓縮的影像,所以本研究以JPEG影像偵測,在實驗結果中顯示,雖然特徴向量維度直接縮減,資料量變少了,但是結果也還算正確,一樣可以正確找出相似的小圖像,達到縮減運算資料量的目的。
Digital images are easy to be tempered and edited due to availability of image editing software. Generally, image tampering protection can be divided into two kinds: (1) active protection: The original digital image is embedded with a watermark which can be used to detect tampering. (2) passive protection: it does not need any digital watermark or signature but relies on the image processing technology to detect the forgery. Passive approaches have not yet been thoroughly researched. The most common ways to temper a digital image is copy-paste forgery which is used to conceal objects or produce a non-existing scene. To detect the copy-paste forgery, we divide the image into blocks as the basic feature for detection, and transfer every block to a feature vector with lower dimension for comparison. The number of blocks and dimension of characteristics are the major factor affecting the computation complexity. In this paper, we modify the previous methods by using less cumulative offsets for block matching. The experimental results show that our method can successfully detect the forgery part even when the forged image is saved in a lossy format such as JPEG. The performance of the proposed method is demonstrated on several forged images.
第一章 研究背景、動機與目的 1
第二章 相關文獻探討 4
2.1 偵測複製貼上竄改介紹及限制 4
2.2 影像特徵區塊提取 7
2.2.1 DCT&量化 9
2.2.2 PCA&量化 10
2.2.3 DWT&SVD 11
2.3 偵測copy-paste演算法 13
第三章 複製區域偵測 18
第四章 實驗結果 26
第五章 結論 34
參考文獻 35
[1]J. Fridrich, D. Soukal, and J. Luk´a_s.,”Detection of copy-move forgery in digital images”, In Proceedings of Digital Forensic Research Workshop, August 2003.

[2] A. Popescu, H. Farid, “Exposing Digital Forgeries by
Detecting Duplicated Image Regions”, Computer
Science, Dartmouth College, Tech. Rep. TR2004-515,2004.

[3] Li G H, Wu Q, Tu D. Sun S J.,” A sorted neighborhood approach for detecting duplicated regions in image forgeries based on DWT and SVD”, In: Proceedings of 2007 IEEE International Conference on Multimedia and Expo, Beijing, China: IEEE, 2007. 1750-1753

[4] Amara Graps, “An Introduction to Wavelets”, IEEE Computational Science and Engineering, 1992, 2(2):50-61.

[5] E. J. Ientilucci, “Using the Singular Value Decompostion”, Chester F. Carlson Center for Imaging Science, Rochester Institute of Technology, 29 May 2003.

[6] B. B. Zhu, M. D. Swanson, and A. H. Tewfik, “When Seeing isn't Believing”, IEEE Signal Processing, March 2004.

[7] A. K. Elmagarmid, P. G. Ipeirotis, “Duplicate Record Detection: A Survey”, IEEE Transactions on Knowledge and Data Engineering, Vol. 19, No. 1, Jan. 2007.

[8] R. C. Gonzalez, R. E. Woods, “Digital Image Processing”, 3rd ed. Reading, MA: Addison-Wesley,1992.

[9] Tian-Tsong Ng, Shih-Fu Chang, Qibin Sun, “Blind Detection of Photomontage Using Higher Order Statistics”, IEEE International Symposium on Circuits and Systems, Vancouver, Canada, May 2004.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top