Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/8844
標題: 應用類神經模糊網路辨識人體手勢
Application Of Neural Fuzzy Network To Human Gesture Recognition
作者: 曾明凱
Tseng, Ming-Kai
關鍵字: Neural Fuzzy Network;類神經模糊網路;Recognition;辨識
出版社: 電機工程學系所
引用: [1] R. C. Gonzalez and R. E. Woods, Digital Image Processing 2/e, Prentice Hall, 2008 [2] H.D. Cheng, X.H. Jiang, Y. Sun and J. Wang, “Color image segmentation: advances and prospects”, Pattern Recognition, vol. 34, pp. 2259-2281, 2001. [3] Q. Zhou and J.K. Aggarwal, “Tracking and classifying moving objects from video,”Proc. IEEE Int. Workshop Performance Evaluation of Tracking and Surveillance, pp. 52-59, Dec. 2001. [4] T. C. Chen, Vision-based Real-Time 3D Human Posture Significant Points Estimation, Master Thesis, National Chung-Hsing University, Taiwan, R.O.C., 2009. [5] C. F. Juang, C. M. Chang, J. R. Wu, and D. M. Lee, “Computer vision-based human body segmentation and posture estimation,” IEEE Trans. Syst., Man, and Cyber., Part A: Systems and Humans, vol. 39, no. 1, pp. 119-133, Jan. 2009. [6] A. F. Bobick and A. D. Wilson, “A state-based technique to the representation and recognition of gesture,” IEEE Transactions Pattern Analysis and Machine Intelligence, vol. 19, pp. 1325-1337, Dec. 1997. [7] T. Starmer, J. Weaver, and A. Pentland, “Real-time American sign language recognition using desk and wearable computer-based video,” IEEE Transactions Pattern Analysis Machine Intelligence, vol. 20, pp. 1317-1375, Dec. 1998. [8] N. Johnson and D. Hogg, “Learning the distribution of object trajectories for event recognition,” Image Vision Computer, vol. 14, no. 8, pp. 609-615, 1996. [9] H. Fujiyoshi and A. J. Lipton, “Real-time human motion analysis by image skeletonization,” Proceedings of IEEE Workshop on Applications of Computer Vision , pp. 15-21, Oct. 1998. [10] Y. Li, M. Songde, and L. Hanqing, “A multiscale morphological method for human posture recognition,” Proceedings of IEEE International Conference on Automatic Face and Gesture Recognition, pp. 56-61, Apr. 1998. [11] P. Spagnolo, M. Leo, A. Leone, G. Attolico, and A.Distante, “Posture estimation in visual surveillance of archaeological sites,” Proceedings of IEEE International Conference on Advanced Video and Signal Based Surveillance, pp. 277-283, July 2003. [12] C. F. Juang and C. M. Chang, “Human body posture classification by a neural fuzzy network and home care system application,” IEEE Trans. Syst., Man, andCyber., Part A: Systems and Humans, vol. 37, no. 6, pp. 984-994, Nov. 2007. [13] I. Haritaoglu, D. Harwood and L. S. Davis, “ real-time surveillance of people and their activities,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 22, no. 8, pp. 809 - 830, Aug. 2000. [14] E.Ardizzone, A.Chella, R.Pirrone, “Pose Classification Using Support Vector Machines,” Proceedings of IJCNN, vol.6 ,pp.317-322, 2000.
摘要: 
本論文主要是提出一種以特徵向量來辨識特定手勢,另外應用類神經模糊網路來辨識。在手勢辨識上,這篇論文考慮到雙手的兩種手勢,針對單一手臂包含向上、彎曲、水平還有向下,經過雙手組合總共有16種手勢。針對特徵擷取,本論文首先使用移動物體切割演算法從背景中取出人的身體,經由一連串影像處理方式獲得一個完整的人體輪廓。我們使用雙手的兩個區域取出的水平及垂直投影,每一個區域切割成15等份的區塊。然後每一個區塊人體剪影占的百分比可以被找出來,因此產生了總共15種的手臂特徵,將這些分別產生的特徵向量輸入類神經模糊網路來作辨識。實驗結果本論文提出的方法在辨識16種手勢上有很高的準確率。比較eigenvector特徵抽出方法還有template-matching辨識法,證明我們所提出的是較為優秀的辨識方法。

This thesis proposes a new feature vector for static gesture recognition and the application of a neural fuzzy network as a recognizer. For gesture recognition, the paper considers both the gestures of the two human hands. For each hand, four kinds of main gestures, including arm up and bending, in straight, down are considered. In total, the number of gestures is sixteen. For feature extraction, the thesis first uses a moving object segmentation algorithm to segment human body from the extracted background. A sequence of image processing is used to obtain a complete silhouette. Histograms of horizontal and vertical projections of the silhouette are then used to clip two focused body regions for the two hands. Each focus region is equally divided into fifteen non-overlapping grids. The percentage of human body silhouette pixels in each grid is found, generating a total of sixteen features for a hand. Each of the two fifteen dimensional feature vectors is fed as inputs to a neural fuzzy network for recognition. Experimental results show that the proposed method recognizes the sixteen gestures with high accuracy. Comparisons with an eigenvector-based feature extraction method and a template-matching recognition method demonstrate the advantage of the proposed recognition method.
URI: http://hdl.handle.net/11455/8844
其他識別: U0005-1708201014010600
Appears in Collections:電機工程學系所

Show full item record
 

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.