本論文主要研究發展電腦影像技術，應用於手指(或手部)之運動分析與量測，以進行相關之臨床評估與診斷。論文主要分為兩部份: 第一部份提出一種以模型為基礎之手指(節)追蹤系統，應用於運動參數之量測。在此多相機追蹤系統中，使用重要性隨機取樣(importance sampling)方式混合前期機率之非線性粒子濾波器(mixture-prior particle filtering)，將標記點、手指影像特徵與手指模型成功的整合以預測手指運動參數。實驗發現所提出的指節式三維手指模型較一般標記點(marker-based)量測方式可獲得較高可靠性之運動參數，也可有效降低標記點因皮膚滑動所引起的誤差(STA: soft-tissue artifact)。同時，也可以成功地應用在常用的多指的運動量測中。第二部份評估手指皮膚對於手指運動評估之實際影響。首先在手指關節皮膚表面畫出網格圖樣(grid)，並利用多相機追蹤系統來追蹤網格影像以建立3D表面網格模型，藉由此模型評估關節表面的皮膚運動。此外，利用同步的平面動態X光影像系統與所發展的電腦視覺系統，可計算皮膚表面網格點在手指骨側面(sagittal view)的相對3D運動，以估計皮膚運動的滑動參數，趨勢與各網格點的滑動特性。進而可將結果應用於手部運動參數的誤差補償或臨床量化評估。另外，本研究亦實現完整的3D皮膚運動量測，以提供更擬真地觀察個人皮膚的運動特性。後續研究將包含改善更穩定的粒子濾波器與對遮蔽點的網格追蹤以提升系統的穩定性。對於3D皮膚運動研究能作更多受測者實驗，以及探討對其他手指部位的運動研究。

In this thesis, we have developed new approaches based on computer-vision and image-analysis techniques to provide quantitatively assessments for finger (or hand) motion analysis and kinematics measurements. There are two major parts explored in this thesis. The first part is to propose a model-based multiple-view fingers tracking system. We proposed a new mixture-prior particle filtering that merges model and marker information to avoid generating improper particles and achieve successful finger tracking with near real-time performance. The experiments showed that the proposed system achieves more reliable kinematics parameters with lower skin-artefact errors than the conventional methods. The system has also been applied to the kinematics measurements for multiple-fingers motions effectively. The second part is to assess the finger skin movements based on the grid pattern which was drawn on the finger skin around a selected joint by using the proposed multiple-view tracking system. A specific grid pattern was hired to reconstruct the grid model for representing skin motion. By using a fluoroscopy system, we can also assess the skin movements with respect to the bony coordinates defined on the simultaneous fluoroscopy. In the experiments, the skin motion was successfully assessed from which the skin motion model and motion tendency can be obtained. The resulting motion model and parameters will be beneficial in clinical evaluation or kinematics measurement. Then, by adopting the resulting 3-D skin motion associated with an MRI-segmented bone model, we can generate a realistic motion animation for the selected 3-D finger joint. In the future studies, I will concentrate on refining the current particle filtering for more reliable finger joint and skin grid tracking. Some extended researches on assessing joint/skin motion for other fingers/limbs and different subjects will also be investigated.

[1] Marzke MW (1997), “Precision grips, hand morphology, and tools,” Am J Phys Anthropol, 102(1) pp.91–110.
[2] Moran SL, Berger RA (2003), “Biomechanics and hand trauma: what you need,” Hand Clinical, 19(1) pp.17–31.
[3] Edmund YS Chao, An KN, Cooney WP and Linsceid RL (1989), Biomechanics of the hand, JBW prjnters & Binder Pte. Ltd.
[4] Chiu HY, Su FC (1996), “The motion analysis system and the maximal area of fingertip motion: A preliminary report,” J. Hand Surgery – British, 21 pp. 604-608.
[5] Kamper DG, Cruz EG, Siegel MP (2003), “Stereotypical fingertip trajectories during grasp,” J. Neurophysiol., 90(6) pp. 3702-3710.
[6] Yokogawa R, Hara K (2004), “Manipulabilities of the index finger and thumb in three tip-pinch postures,” J Biomech Eng, 126 pp.212.
[7] Cole KJ, Abbs JH (1986), “Coordination of three-joint digit movements for rapid finger–thumb grasp,” J Neurophysiol., 55(6) pp.1407–1423.
[8] Chiu HY, Lin SC, Su FC, Wang ST, Hsu HY (2000), “The use of the motion analysis system for evaluation of loss of movement in the finger,” J. Hand Surgery- British, 25 pp. 195-199.
[9] Su FC, Kuo LC, Chiu HY, Chen M (2003), “Video-computer quantitative evaluation of thumb function using workspace of the thumb,” J. Biomechanics, 36 pp. 937-942.
[10] Kuo LC, Cooney WP, Kaufman KR, Chen QS, Su FC, An KN (2004), “A quantitative method to measure maximal workspace of the trapeziometacarpal joint--normal model development,” J. Orthopedic Research, 22 pp. 600-606.
[11] Dragulescu D, Perdereau V, Drouin M, Ungureanu L, Menyhardt K (2007), “3D active workspace of human hand anatomical model,” BioMed Eng online, 6(15) pp.1-12.
[12] Kuo LC, Chiu HY, Chang CW, Hsu HY, Sun YN (2009), “Functional Workspace for Precision Manipulation in Normal Hands,” J Electromyography and Kinesiology, 19 pp. 829–839.
[13] Lee SW, Zhang X (2007), “Biodynamic modeling system identification and variability of multiple-finger movements,” J Biomechanics, 40(7) pp. 3215-3222.
[14] Liu B, Manner R (2005), “An Efficient and Accurate Method for 3D-Point Reconstruction from Multiple Views,” Int. J. Computer Vision, 65(3) pp. 175–188.
[15] Cerveri P, Pedotti A, Ferrigno G (2005), “Kinematical models to reduce the effect of skin artifacts on marker-based human motion estimation,” J Biomechanics, 38 pp. 2228–2236.
[16] Cerveri P, De Mon E, Marchente M, Lopomo N, Baud-Bovy G, Barros RM (2008), “In vivo validation of a realistic kinematic model for the Trapezio-metacaerepal joint using an optoelectronic system,” Annal of Biomedical Engineering, 36(7) pp. 268-1280.
[17] Li ZM, Tang J (2007), “Coordination of thumb joints during opposition,” J Biomechanics, 40 pp. 502-510.
[18] Buchholz B, Armstrong TJ (1992), “A kinematic model of the human hand to evaluate its prehensile capability,” J. Biomechanics, 25(2) pp. 149-162.
[19] Chua CS, Guan H, and Ho YK (2002), “Model-based 3D hand posture estimation from a single 2D image,” J Image and Vision Computing, 20(3) pp. 191-202.
[20] Lin J, Wu Y, and Huang TS (2002), “Capturing human hand motion in image sequences,” In: Proceeding of IEEE Workshop on Motion and Video Computing, pp. 99-104.
[21] Lee SW and Cohen I (2004), “3D hand reconstruction from a monocular view,” In: Proceeding of the 7th International Conference on Pattern Recognition (ICPR04) Cambridge, United Kingdom, 3 pp. 310-313.
[22] Fuller J, Liu LJ, Murphy MC, Mann RW (1997), “A comparison of lower-extremity skeletal kinematics measured using skin- and pin-mounted markers,” Human Movement Science, 16 pp. 219–242.
[23] Lafortune MA, Lake MJ (1991), “Errors in 3D analysis of human movement,” In: Proc. the First International Symposium on 3D Analysis of Human Movement, pp. 55–56.
[24] Cappozzo A, Catani F, Leardini A, Benedetti M G , Della Croce U (1996), “Position and orientation in space of bones during movement: experimental artifacts,” Clinical Biomechanics, 11 pp. 90–100.
[25] Leardini A, Chiari L, Della Croce U, Cappozzo A (2005), “Human movement analysis using stereophotogrammetry: Part 3. Soft tissue artifact assessment and compensation,” Gait and Posture, 21 pp. 212–225.
[26] Alexander EJ, Andriacchi TP (2001), “Correction for deformation in skin-based marker systems,” J. Biomechanics, 34 pp. 355–361.
[27] Schwartz MH, Rozumalski A (2005), “A new method for estimating joint parameters from motion data,” J Biomechanics, 38 pp. 107-116.
[28] Lucchetti L, Cappozzo A, Cappello A, Croce UD (1998), “Skin movement artifact assessment and compensation in the estimation of knee-joint kinematics,” J. Biomechanics, 31 pp. 977-984.
[29] Gamomilla V, Donati M., Stagni R, cappozzo A (2009), “Non-invansive assessment of superficial soft tissue local displacements during movement: A feasible study,” J. Biomechanics, 42 pp. 931-937.
[30] Garling EH, Kaptein BL, Mertens B, Barendregt W, Veeger HEJ, Nelissen RGHH, Valstar ER (2007), “Soft-tissue assessment during step-up using fluoroscopy and skin-mounted markers,” J. Biomechanics, 40(1) pp. s18-s24.
[31] Cappozzo, A, Catani F, Leardini A, Benedetti MG, Della Croce U (1996), “Position and orientation in space of bones during movement: experimental artifacts,” Clinical Biomechanics, 11 pp. 90–100.
[32] Leardini A, Chiari L, Della Croce U, Cappozzo A (2005), “Human movement analysis using stereophotogrammetry: Part 3. Soft tissue artifact assessment and compensation,” Gait and Posture, 21 pp. 212–225.
[33] Stagni R, Fantozzi S, Cappello A, Leardini A (2005), “Quantification of soft tissue artefact in motion analysis by combining 3D fluoroscopy and stereophotogrammetry: a study on two subjects,” Clinical Biomechanics, 20 pp. 320–329.
[34] Sangeuxa M, Marina F, Charleuxc F, Dürselenb L, M.C. Ho Ba Th (2006), “Quantification of the 3D relative movement of external marker sets vs. bones based on magnetic resonance imaging,” Clinical Biomechanics, 21(9) pp. 984–991.
[35] Beillasa P, Papaioannou P, Tashmanb S, Yang KH (2004), “A new method to investigate in vivo knee behavior using a finite element model of the lower limb,” J. Biomechanics, 37 pp. 1019–1030.
[36] Dunbar WL, Un K, Donzelli PS, Spilker RL (2001), “An evaluation of three-dimensional diarthrodial joint contact using penetration data and the finite element method,” J. Biomechanical Engineering, 123 (4) pp. 333–340.
[37] Richard RL, Lester ME, Miller SF, Bailey JK, Hedman TL, Dewey WS, Greer M, Renz EM, Wolf SE, Blackbourne LH (2009), “Identification of cutaneous functional units related to burn scar contracture development,” J. Burn Care Res., 30(4) pp. 625-31.
[38] Wilhelmi BJ, Blackwell SJ, Phillips LG (1999), “Langer's Lines: To Use or Not t Use,” Plastic & Reconstructive Surgery, 104 (1) pp. 208-214.
[39] Richard H and Andrew Z (2001), Multiple-view computer vision, Cambridge.
[40] Gordon NJ, Salmond DJ, Smith AFM (1993), “Novel approach to nonlinear/non-Gaussian Bayesian state estimation,” Radar and Signal Processing IEE Proceedings F, 140(2) pp. 107-113.
[41] Kitagawa G (1996), “Monte Carlo filter and smoother for non-Gaussian nonlinear state space models,” J. Comput. Graph. Statist., 5(1) pp. 1–25.
[42] Liu JS and Chen R (1998), “Sequential Monte Carlo Methods for Dynamic Systems,” J. American Statistical Association, 93(443) pp. 1032-1044.
[43] Doucet A, Godsill S, and Andrieu C (2000), “On sequential Monte Carlo sampling methods for Bayesian filtering,” Statist. Comput., 10(3) pp. 197-208.
[44] Carpenter J, Clifford P, and Fearnhead P (1999), “Improved particle filter for nonlinear problems,” Proc. Inst. Elect. Eng., Radar, Sonar, Navig., 146(1) pp. 2-7.
[45] Gordon N (1997), “A hybrid bootstrap filter for target tracking in clutter,” IEEE Transactions on Aerospace and Electronic Systems, 33(1) pp. 353-358.
[46] Pitt MK and Shephard N (1999), “Filtering Via Simulation: Auxiliary Particle Filters,” J. American Statistical Association, 94 pp. 590-599.
[47] Chang WY, Chen CS, Jian YD (2008), “Visual Tracking in High-Dimensional State Space by Appearance-Guided Particle Filtering,” IEEE Trans Image Processing. 17(7) pp. 1154-1167.
[48] Kuo LC, Su FC, Chiu HY, and Yu CY (2002), “Feasibility of using a video-based motion analysis system for measuring thumb kinematics,” J. Biomechanics, 35(11) pp. 1499-1506.
[49] Davis RB, Ounpuu S, Tyburski D, Gage JR (1991), “A gait analysis data collection and reduction techniques,” Hum Mov Sci., 10 pp.575–587.
[50] Salvador C, Manuel F, Angel Sanchez-Uran, Ortego M Javier and Rafael A (2010), “Human hand descriptions and gesture recognition for object manipulation,” Computer Methods in Biomechanics and Biomedical Engineering, 8 (iFirst) pp. 1-13.
[51] MacCormick J, Isard M (2002), “Partitioned sampling, articulated objects and interface-quality hand tracking,” In: Proceeding Sixth European Conference on Computer Vision, pp. 3-19.
[52]Comaniciu D, Ramesh V, Meer P (2000), “Real-time tracking of non-rigid objects using mean shift,” In: IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’00). Sourth Carolina, June 13-15, 2: 2142.
[53] Isard M, Blake A (1998), “CONDENSATION – Conditional Density Propagation for Visual Tracking,” Int. J. Computer Vision, 1 (29) pp. 5–28.
[54] Doucet, A, de Freitas, JFG and Gordon NJ (2000), Sequential Monte Carlo Methods in Practice, Springer-Verlag.
[55] Arulampalam, Maskell MS, Gordon S, Clapp NT (2002), “A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking,” IEEE Trans Signal Processing, 50(2) pp. 174-188.
[56] Djuric PM, Kotecha JH, Zhang J, Huang Y, Ghirmai T, Bugallo MF, Miguez J (2003), “Particle Filtering,” IEEE Signal Processing Magazine, 20 (5) pp. 19-38.
[57] William M. S. and Noel T (1983), “Estimation of the intraclass correlation coefficient for the analysis of covariance model,” The American Statistician, 37 (3) pp. 221-224.
[58] Chang CW, Ho SP, Sun YN, Kuo LC, Lin CCK, Ju MS, Su FC, Chou IM (2008), “Multiple-View Multiple Fingers Tracking using 3-d Model and Color Features for Therapeutic Evaluations,” 16th International Conference on Mechanics in Medicine and Biology.
[59] Ryu JH, Miyata N, Kouchi M, Mochimaru M, Lee KH (2006), “Analysis of skin movement with respect to flexional bone motion using MR images of a hand,” J. Biomechanics, 39(5) pp. 844-852.
[60] Comaniciu D, Ramesh V, and Meer P (2003), “Kernel-Based Object Tracking,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 25 (5) pp. 564-577.
[61] Lin WC, Liu Y (2007), “A Lattice-Based MRF Model for Dynamic Near-Regular Texture Tracking,” IEEE Trans. Pattern Analysis and Machine Intelligence, 29(5) pp.777 – 792.
[62] Yang C, Duraiswami R, Davis L (2005), “Efficient mean-shift tracking via a new similarity measure,” Proc. the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05), pp.1063-6919.
[63] Matthews I, Ishikawa T, and Baker S (2004), “The template update problem,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 26 (6) pp.-815.
[64] Schoepflin T, Chalana V, Haynor DR, and Kim Y (2001), “Video Object Tracking With a Sequential Hierarchy of Template Deformations”, IEEE Trans Circuits and System for Video Technology, 11 (11) pp. 1171-1182.
[65] Baltzopoulos V (1995), “A videofboroscopy method for optical distortion correction and measurement of knee-joint kinematics,” Clinical Biomechanics, 10(2) pp. 85-92.
[66] 吳佳興、孫永年(2009), ”以模型為基礎之影像對位系統應用於立體關節運動重構”國立成功大學資訊工程研究所碩士論文.
[67] Kang DJ (1999), “A fast and stable snake algorithm for medical images,” Pattern Recognition Letters, 20 pp. 507-512.
[68] Fritsch FN and Carlson RE (1980), “Monotone Piecewise Cubic Interpolation,” SIAM J. Numerical Analysis, 17 pp. 238-246.
[69] Todeschini R, Ballabio D, Consonni V, Manganaro A, and Mauri A (2009), “Canonical measuren of correlation(CMC) and canonical measure of distance (CMD) between nsets of data Part 1: Theroy and simple chemometric applications,” Analytica chimica acta, 648 pp. 45-51.
[70] Zang T, Yue D, Gu Y, Wang Y, and Yu G (2009), “Adaptive correlation analysis in stream time series with sliding windows,” Computers and mathematics with applications, 57 pp. 937-948.
[71] Sungchan B, Thomas JA (2011), “A finger motion model for reach and grasp,” Int. J. Industrial Ergonomics, 41 pp. 79-89.
[72] Banks A, Hodge WA (1996), “Accurate measurement of three-dimensional knee replacement kinematics using single-plane fluoroscopy,” IEEE Trans. on Biomedical Engineering, 43(6) pp. 638-649.
[73] Chen HC, Jou IM, Wang CK, Su FC, and Sun YN (2010), “Registration-based segmentation with articulated model from multi-postural magnetic resonance images for hand bone motion animation,” Medical Physics, 37(6) pp. 2670-2682.,
[74] Borgefors G (1988), “Hierarchical chamfer matching: A parametric edge matching algorithm,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 10(6) pp. 849-865.