臺灣博碩士論文加值系統

本研究主要提出基於粒子濾波器之使用者指尖位置定位(Positioning)與追蹤
(Tracking)演算法，粒子濾波器(Particle Filtering)為一個有效且被廣泛使用之追蹤架構，本研究將透過評估與調整每個粒子之指尖可能性權重，達成指尖追蹤與定位目標，為進一步使系統能夠有效克服複雜背景及手部姿態變化，本研究整合膚色
與輪廓資訊。首先將擷取影像中膚色區域，於偵測出輪廓線條後，利用輪廓與膚
色特徵估算手指指尖輪廓與區域之可能性機率。於實驗中，本研究拍攝數段影片，
並量化分析指尖定位準確度，可明顯發現結合凸包輪廓特徵的粒子濾波器(Particle
Filtering)可有效提高辨識效率與速度，於手部姿態變化場景中。

The paper proposes an algorithm for positioning and tracking fingertip by fusing
multiple cues including skin color and hand contour under particle filtering which is an
effective and widely used framework for object tracking. First of all, we extract the
skin-color regions from the observed image. Then, the skin-color ratio of the region
centered at the particle is calculated. In addition, eight edge templates of the fingertip
and convex hull are used for evaluating the likelihood probability of the particle as well.
The combination of the skin color and contour cues makes the system robust to
illumination variation and complex background. Particularly, the embedding of Convex
Hull algorithm improves the effectiveness in case of variable hand pose. In the
experiment, twelve videos are used to validate the performance of the proposed method.
The proposed method outperforms our previous work in both accuracy and efficiency.

摘要 ................................................................ I
英文摘要 ........................................................... II
誌謝 .............................................................. III
目錄 ............................................................... IV
表目錄 ............................................................. VI
圖目錄 ............................................................ VII
第一章 緒論 ......................................................... 1
1.1 研究動機 .................................................... 1
1.2 目標 ........................................................ 2
1.3 困難與挑戰 .................................................. 3
1.3.1 複雜背景 .............................................. 4
1.3.2 不同光照條件 ........................................... 4
1.3.3 變化的手勢姿態 ......................................... 5
1.3.4 運動模糊化 ............................................. 5
1.4 相關文獻 .................................................... 6
1.5 系統架構 .................................................... 8
第二章 粒子濾波器推論架構 ........................................... 9
2.1 狀態定義與說明 .............................................. 9
2.2 粒子濾波器 (Particle Filtering) ............................ 11
2.2.1 預測(Prediction) ..................................... 13
2.2.2 測量(Measurement) .................................... 13
2.2.3 估算(Estimation) ..................................... 13
2.2.4 重新取樣(Resampling) ................................. 14
第三章 膚色切割 .................................................... 17
3.1 膚色模型(Skin Model) ....................................... 17
3.2 膚色區域偵測 ............................................... 19
第四章 可能性評估 .................................................. 21
4.1 指尖膚色分佈特性 ........................................... 21
4.2 輪廓可能性評估 ............................................. 23
4.3 指尖樣板比對 ............................................... 26
第五章 實驗 ........................................................ 30
5.1 實驗環境 ................................................... 30
5.2 實驗參數 ................................................... 31
5.3 影片資料庫 ................................................. 31
5.4 實驗分數 ................................................... 34
5.5 實驗結果 ................................................... 35
第六章 結論 ........................................................ 47
參考文獻 ........................................................... 48

[1] Zhe Lin, Larry S. Davis, David Doermann, and Daniel DeMenthon, “Hierarchical
Part-Template Matching for Human Detection and Segmentation”,ICCV, pp. 1-8,
2007
[2] Jong-Min Kim , Woong-Ki Lee, “Hand Shape Recognition using Fingertips”, Fifth
International Conference on Fuzzy Systems and Knowledge Discovery pp.
44-48,2008
[3] Oka K. Sato Y. Koike H. “Real-time fingertip tracking and gesture recognition”,
Computer Graphics and Applications, IEEE Volume 22, Issue 6, Nov.-Dec. pp64 –
71 2002
[4] Sze, M.M., Dy, A.L., Sarmenta, L.F., Inexpensive Camera based Gestural Interface
for PC''s, 4th Philippine Computing Science Congress, University of the Philippines,
Los Banos Laguna, 2004 F.
[5] Sung Kwan Kang, Mi Young Nam , Phill Kyu Rhee, “Color Based Hand and Finger
Detection Technology for User Interaction”, International Conference on
Convergence and Hybrid Information Technology pp. 229-236,2008
[6] Zhi-Wei Chen, Yu-Cheng Lin and Cheng-Chin Chiang, “The Design of a
Vision-based Fingertip Writing Interface”, The 18th International Conference on
Pattern Recognition pp. 104-107,2006.
[7] X. Wen, Y. Niu, “A Method for Hand Gesture Recognition Based on Morphology
and Fingertip- Angle”, The 2nd International Conference on Computer and
Automation Engineering (ICCAE), vol. 1, pp. 688-691, 2010
[8] Lianwen Jin, Duanduan Yang; Li-Xin Zhen, Jian-Cheng Huang, “A Novel Vision
based Finger-writing Character Recognition System”, 18th International
Conference on Pattern Recognition pp. 1104-1107, 2006.
[9] Sheng-Ming Liang,” Fingertip Positioning and Tracking Using Particle Filtering”Consumer Electronics (ISCE), 2013 IEEE 17th International Symposium on pp.
205-206,2011
[10] Ko-Jen Hsiao, Tse-Wei Chen, and Shao-Yi Chien, “Fast fingertip positioning by
combining particle filtering with particle random diffusion,” IEEE International
Conference on Multimedia and Expo (ICME2008), pp. 977-980, Hannover,
Germany, Jun. 2008
[11] C. Garcia, and G.Tziritas. “ Face detection using quantized skin color regions
merging and wavelet packet analysis.” IEEE Transactions on Multimedia vol. 1 ,
No. 3, 1999 , pp. 264-277.
[12] Y. P. Lew, A. R. Ramli, S. Y. Koay, R. Ali and V. Prakash, ”A Hand Segmentation
Scheme Using Clustering Technique in Homogeneous Background”, Research and
Development, pp.305-308 2002. SCOReD 2002. Student Conference on 16-17 July
[13] M. Soriano, S. Huovinen, B. Martinkauppi, and M. Laaksonen, “Using the skin
locus to cope with changing illumination conditions in color-based face tracking,”
in Proceedings of the IEEE Nordic Signal Processing Symposium, pp. 383-386,
2000.
[14] R. C. Gonzalez, R. E. Woods, Digital Image Processing, Prentice Hall, 2008, 3rd.
[15] Chien-Cheng Lee.”Fingertip-Writing Alphanumeric Character Recognition for
Vision-Based Human Computer Interaction”. Broadband, Wireless Computing,
Communication and Applications (BWCCA), 2010 pp.533 – 537
[16] Lae-Kyoung Lee.” Robust Fingertip Extraction with Improved Skin Color
Segmentation for Finger Gesture Recognition in Human-Robot Interaction”
Evolutionary Computation (CEC), 2012. Page(s):1 – 7
[17] Xiao-Heng Jiang .”A robust method of fingertip detection in complex background”
Machine Learning and Cybernetics (ICMLC), 2012. Page(s):1468 – 1473
[18] R. L. Graham. “An efficient algorithm for determining the convex hull of a finite
planar set”, Information Processing Letters, 7:175–180, 1972.
[19] Xie, Qunqun.” A Fast and Robust Fingertips Tracking Algorithm for Vision-Based
Multi-touch Interaction ” Control and Automation (ICCA), pp.1346 –
1351 ,2013 10th.
[20] C. Manresa, J. Varona, R. Mas, F.J. Perales, “Real –Time Hand Tracking and
Gesture Recognition for Human-Computer Interaction”, Electronic Letters on
Computer Vision and Image Analysis 0(0):1-7, 2000
[21] C.Garcia and G.tzirita, “Face detection using quantized skin color region merging
and wavelet packet analysis “, IEEE trans . Multimedia , pp.264-277 ,1999.
[22] Canny, J.，”A Computational Approach To Edge Detection”, IEEE Trans. Pattern
Analysis and Machine Intelligence, 8:pp.679-714, 1986.