即時性連續手勢追蹤與辨識

Real Time Tracking and Recognition for Continuous Gestures

許倫維(Lun-Wei Hsu)

指導教授 ： 張厥煒

國立臺北科技大學/機電學院/資訊工程系所/碩士(2006年)

https://doi.org/10.6841/NTUT.2006.00215

摘要

利用手勢辨識可協助無人看護或行動不便人士等對周遭設備(如電腦和電器)的操作。本論文針對連續手勢提出一套辨識方法。利用顏色的屬性追蹤手勢及做出手勢切割。先以Central Moments的方法求得特徵向量，再利用隱藏式馬可夫模型進行複雜背景下一至多種手勢的辨識。本論文所提方法，可使電腦具有更便利、省人工和人性化之辨識方法。

關鍵字

手勢辨識 ； Mean Shift ； Particle Filter ； 隱藏式馬可夫模

並列摘要

People with disabilities can operate Machines and Computers with the help of hand gesture recognition. This thesis proposed a method for hand gesture recognition. Detecting and segmentation human hand are finished by the color attribute. We use Central Moments as the shape feature, and recognize gesture under complex environment through hidden markov model. The method we proposed make users interact with machines more convenient, humanistic and easily.

並列關鍵字

Gesture Recognition ； Mean Shift ； Particle Filter ； Hidden Markov Model

參考文獻

[1]. K. Abe, H. Saito, and S. Ozawa, “Virtual 3-D interface system via hand motion recognition from two cameras,” IEEE Trans. on Systems, Man, and Cybernetics-Part A, Vol.32, No.4, pp.536-540, July 2002.

[2]. M. S. Arulampalam, S. Maskell, N. Gordon, and T. Clapp, “A tutorial on particle filters for online nolinear/non-gaussian Bayesian tracking,” IEEE Trans. on Signal Processing, pp. 174-188, Vol. 50, No. 2, 2002.

[3]. V. Athitsos and S. Sclaroff, “An appearance-based framework for 3D hand shape classification and camera viewpoint estimation,” Proc. IEEE Int. Conf. Automatic Face and Gesture Recognition, pp. 45-50, 2002.

[6]. D. Comaniciu, V. Ramesh, and P. Meer, “Real-time tracking of non-rigid objects using mean shift,” Proc. IEEE Int. Conf. Computer Vision and Pattern Recognition, pp. 142-149, 2000.

[7]. B. Han, D. Comaniciu, Y. Zhu, and L. Davis, “Incremental density approximation and kernel-based Bayesian filtering for object tracking,” Proc. IEEE Int. Conf. Computer Vision and Pattern Recognition, pp. 638-644, Vol. 1, 2004.