階梯教室之學生上課動作分析系統

Students' gesture analysis system in a lecture theatre

郭明翰(ming han kuo)

指導教授 ： 方瓊瑤

國立臺灣師範大學/理學院/資訊工程研究所/碩士(2010年)

摘要

本篇論文主要探討階梯教室學生上課動作分析系統。動作分析在教育領域尤其重要，老師可藉由觀察學生的動作知道學生的學習狀況。例如學生舉手，老師可對學生不懂的部份加以說明，提高學生的學習效率。
本研究將攝影機架設在階梯教室前面以擷取學生上課影像。系統先定位教室椅背高度線，並取出影像前景顏色。藉由motion與前景顏色資訊整合判斷出影像前景點。接著利用影像前景點為種子進行region growing，並將擴展出的regions利用region combination演算法定位學生物件。本系統使用學生物件的組成關係變化辨識六種動作，分別為舉右手、舉左手、舉雙手、趴下、站立與正常坐姿。實驗結果顯示提出方法可以解決一些環境的問題，並對實驗遇到其他問題有分析與討論。
最後本研究還有能改進的部份，未來希望能整合其他資訊或者更適合的方法讓本研究更完整。

關鍵字

動作辨識 ； 階梯教室 ； region gorwing ； object segmentation

並列摘要

This paper presents a students’ gestures recognition system in a theater classroom. Gesture recognition is important in many fields, especially in education field. Teacher can know students’ situation by observing their gestures in class. For example, students raise their hands may have question in class, so teacher can explain that make students clear in problem.
In this study the PTZ camera is set in front of the classroom to capture the student sequence. The system first preprocess the input image to locate the main line and extract foreground color. Then system combine motion and foreground color to judge foreground pixels. System use these foreground pixels as seeds to do region growing. Then system apply region combination algorithm to do object segmentation. Six student gesture, including different gesture of raising the right hand, raising the left hand, raising two hands, lining prone, standing up and normal, are classified by the relationship of objects. The experimental results show that the proposed method can solve some experimental situation problem, and the other problem also analyzed and discussed.
Finally, the study have some part can be improved. In the future year, I hope that better technology and information can added to this system in order to make the research more complete.

參考文獻

[Mit07] S. Mitra and T. Acharya, “Gesture Recognition: A Survey,” IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews, Page(s):311 - 324, 2007.

[Nat08] P. Natarajan and R. Nevatia “Online, Real-time Tracking and Recognition of Human Actions,” Proceedings of IEEE Workshop on Motion and video Computing, (WMVC 2008), Page(s):1 - 8, 2008.

[Pen00] P. Hong, M. Turk and T. S. Huang, “Gesture Modeling and Recognition Using Finite State Machines,” Proceedings of Fourth IEEE International Conference on Automatic Face and Gesture Recognition, in France, Page(s):410 - 415, 2000.

[Wan08] K. Wan and H. Sawada, “Dynamic Gesture Recognition Based on the Probabilistic Distribution of arm Trajectory,” Proceedings of International Conference on Mechatronics and Automation, Page(s):426 – 431, 2008.

[Wei08] W. H. Andrew Wang and C. L. Tung, “Dynamic N Using Hand Gesture Recognition Hierarchical Dynamic Bayesian Networks Through Low-Level Image Processing” Proceedings of International Conference on Machine Learning and Cybernetics, Page(s):3247 – 3253, 2008.