跳到主要內容

臺灣博碩士論文加值系統

(44.201.97.0) 您好!臺灣時間:2024/04/13 10:53
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:蔡宗育
研究生(外文):Tsung-Yu Tsai
論文名稱:利用類神經網路實現人形擷取與融合系統
論文名稱(外文):A Figure Extraction and Synthesis System by Neural Networks
指導教授:張傳育
指導教授(外文):Chuan-Yu Chang
學位類別:碩士
校院名稱:國立雲林科技大學
系所名稱:資訊工程研究所
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:2008
畢業學年度:96
語文別:中文
論文頁數:55
中文關鍵詞:影像融合類神經網路人形擷取
外文關鍵詞:Figure ExtractionSynthesisNeural Networks
相關次數:
  • 被引用被引用:0
  • 點閱點閱:206
  • 評分評分:
  • 下載下載:44
  • 收藏至我的研究室書目清單書目收藏:4
隨著多媒體娛樂事業的發展,影像的擷取與融合技術變得十分熱門,從複雜環境的影片中將人形區域完整的擷取出來,是一項相當困難的技術。因此,本篇論文提出一個新的人形擷取與融合技術,可完整地將正確的人形區域,從複雜環境的連續影像中擷取出來。
本篇論文利用臉部偵測技術的結果與一些影像處理技術,建立出一個人形樣板 (figure template) 。依據此模型的相對位置從人形影像 (figure image) 中取出人形、非人形及未知區域的影像特徵,透過自組織映射圖 (Self-Organizing Map, SOM) 與學習向量量化 (Learning Vector Quantization, LVQ) 類神經網路做訓練分類,將人形影像中未知區域正確分類成人形或非人形區域,擷取出的人形區域融合到使用者所選擇的另一張背景影像中,最後,將融合新背景的影像製作成影片,方便使用者保存並瀏覽觀看。實驗結果顯示,本文所提出的方法能夠有效的從背景複雜的環境中,正確擷取出人形區域並做融合,也能製作出一段句個人風格的影片。
Extracting complete figures from videos with complicated environments is difficult. A new figure extraction and synthesis system with capability of extracting figures from consecutive frames in a messy environment is proposed in this thesis. A figure template is constructed based on the face detection results and some image processing techniques.
Figural and non-figural features are extracted from the figure images. By means of these features, a self-organizing map (SOM) and a learning vector quantization (LVQ) neural network are applied to classify the uncertain regions into figural and non-figural objects. The extracted figure can be further synthesized into an optional cinestrip. Experimental results showed the proposed method successfully extract the figure object from a complex background environment.
摘要 ------------------------------------------------------------------------- i
ABSTRACT ------------------------------------------------------------------------- ii
誌謝 ------------------------------------------------------------------------- iii
目錄 ------------------------------------------------------------------------- iv
圖例索引 ------------------------------------------------------------------------- vi
表格索引 ------------------------------------------------------------------------- viii
第一章 緒論------------------------------------------------------------------- 1
1.1 研究動機------------------------------------------------------------- 1
1.2 相關文獻------------------------------------------------------------- 2
1.3 研究方法與成果---------------------------------------------------- 3
1.4 章節大綱------------------------------------------------------------- 5
第二章 前處理程序及相關影像處理技術------------------------------- 6
2.1 群組內部變異數極小化------------------------------------------- 8
2.2 點處理運算---------------------------------------------------------- 9
2.3 形態學影像處理---------------------------------------------------- 9
2.3.1 凸形封包------------------------------------------------------------- 10
2.3.2 斷開------------------------------------------------------------------- 11
2.4 人形樣板建立------------------------------------------------------- 12
第三章 利用類神經網路實現人形擷取與融合之方法---------------- 14
3.1 特徵擷取------------------------------------------------------------- 15
3.2 SOM類神經網路--------------------------------------------------- 19
3.2.1 計算訓練範例與各輸出層單元的距離------------------------- 20
3.2.2 找出優勝單元與調整權重值------------------------------------- 20
3.2.3 使用SOM類神經網路-------------------------------------------- 21
3.3 LVQ類神經網路---------------------------------------------------- 22
3.3.1 LVQ網路架構------------------------------------------------------- 24
3.3.2 使用LVQ類神經網路--------------------------------------------- 25
3.4 影像融合與影片製作---------------------------------------------- 27
第四章 實驗結果與討論 -------------------------------------------------- 30
4.1 實驗環境------------------------------------------------------------- 30
4.2 實驗模擬結果------------------------------------------------------- 32
4.3 效能評估------------------------------------------------------------- 37
4.4 正確率探討---------------------------------------------------------- 38
4.4.1 人形樣板與正確率探討------------------------------------------- 39
4.4.2 類神經網路使用與正確率探討---------------------------------- 40
4.5 其他實驗模擬------------------------------------------------------- 42
4.6 其他實驗模擬數據比較------------------------------------------- 45
4.7 影片製作展示------------------------------------------------------- 46
第五章 結論 ------------------------------------------------------------------ 49
參考文獻 ------------------------------------------------------------------------- 50
附錄 A Figure Extraction and Synthesis System by Learning Vector Quantization Neural Networks---------------------------- 52
[1]S.K. Weng, C.M. Kuo, and S.K. Tu, “Video Object Tracking using Adaptive Kalman Filter”, Journal of Visual Communication & Image Representation. Vol 17, pp. 1190-1208, 2006.
[2]S.W. Tu, “A Neuro-Fuzzy Approach to Detection of Human Face and Body for MPEG Video Compression”, A Thesis Submitted to Graduate School of Electrical Engineering, National Sun-Yat Sen University, 2001.
[3]S.Y. Chien, S.Y. Ma, L.G. Chen, “Efficient Moving Object Segmentation Algorithm using Background Registration Technique”, IEEE Trans. on Circuits and Systems for Video Technology, Vol. 12, pp. 577-586, 2002.
[4]C. Kim, J.N. Hwang, “Fast and Automatic Video Object Segmentation and Tracking for Content-Based Application”, IEEE Trans. on Circuits and Systems for Video Technology, Vol. 12, No. 2, pp. 122-129, 2002.
[5]Y.C. Luo and J.Y. Chang, “Foreground Image Extraction in the HSV Color Space”, Images & Recognition. Vol 13. NO.3, pp. 4-13, 2007.
[6]C.Y. Chang, Y.C Tu, and H.H. Chang, “Adaptive Color Space Switching Based Approach for Face Tracking”, Lecture Notes in Computer Science, Vol. 4223, pp.244-252, 2006.
[7]T. Kohonen, “Self-Organized Formation of Topologically Correct Feature Maps”, Biological Cybernetics, vol. 43, pp. 59-69, 1982.
[8]T. Kohonen, “Learning Vector Quantization for Pattern Recognition”, Technical Report TKK-F-A601, Helsinki University of Technology, Finland, 1986.
[9]T. Kohonen, “Improved Versions of Learning Vector Quantization”, Proceedings of the International Joint Conference on Neural Networks, San Diego, CA, vol. 1, , pp.545-50, 1990.
[10]C.Z. Chang, “Tracking Multiple Moving Objects Using Level Set Method ”, Master Thesis, Graduate School of Electrical Engineering, Yuan-ze University, 2001.
[11]R.C. Gonzalez and R.E. Woods, Digital Image Processing, Prentice Hall, New Jersey, 2002.
[12]葉怡成,「類神經網路模式應用與實作 第七版」,儒林出版社,2000。
[13]R. M. Gray, “Vector Quantization,” IEEE Acoustics, Speech, and Signal Processing Magazine, vol. 1, pp. 4-29, 1984.
[14]Manuel G. Penedo, Maria J. Carreria, Antonio Mosquera, and Diego Cabello, “Computer-Aided Diagnosis: A Neural-Network-Based Approach to Lung Nodule Detection”, IEEE Trans. on Medical Imaging, 17, 872-880, 1998.
[15]C.H. Chang, “Learning Vector Quantization Neural Networks for LED Wafer Defect Inspection”, A Thesis Submitted to Graduate School of Computer Science and Information Engineering, National Yunlin University of Science & Technology, 2007.
[16]N. Doulamis, A. Doulamis, and S. Kollias, “Improving the Performance of MPEG Compatible Encoding at Low Bit Rates Using Adaptive Neural Networks”, Real-Time Imaging, vol 6, NO 5, 2000.
[17]I. Kompatsiaris and M.G. Strintzis, “Spatiotemporal Segmentation and Tracking of Object for Visualization of Videoconference Image Sequence”, IEEE Trans. on Circuits and Systems for Video Technology, Vol. 10, No. 8, 2000.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top