運用PTZ攝影機進行授課教師與手寫資料追蹤的自動化課程錄影系統

Lecturer and Handwritten Data Tracking Using a Pan-Tilt-Zoom Camera for Automatic Lecture Recording

潘銘和(Ming-He Pan)

指導教授 ： 廖珗洲

朝陽科技大學/資訊學院/資訊工程系/碩士(2012年)

摘要

本研究提出使用Pan-Tilt-Zoom(PTZ)攝影機來對以寬黑板的傳統教學進行自動化錄影，此系統稱為ACRS(After Class Reviewing System)。為了能讓學生複習課程，影片當中的手寫資料比教師更為重要。拍攝影片能一起考慮手寫資料與教師是最適合學生複習的。ACRS主要分為兩個階段：廣角階段與放大階段，廣角階段是以時間差異法來偵測教師；放大階段的教師偵測則是以Sobel邊緣偵測與迭代法最佳臨界值來偵測，手寫資料偵測是改進GMM來建立背景模型，以此背景模型透過背景相減法來偵測手寫資料。攝影機是透過兩個預設的參數、教師偵測與手寫資料偵測的結果來控制的。在實驗研究中，經由ACRS所拍攝的影片與攝影師手動記錄的影片做比較。約有83%的影片時間接近手動拍攝並適合學生來複習。

關鍵字

物體追蹤 ； 影片錄影系統 ； 自動化演講錄影

並列摘要

For automatic recording of the traditional lecture with a wide blackboard by using a Pan-Tilt-Zoom (PTZ) camera, a system, called ACRS (After Class Reviewing System), was proposed. To students reviewing the lecture the handwritten data is more important than video of the lecturer. The handwritten data and the lecturer were considered together to record a lecture most suitable for student review. ACRS is consists to two phases: the wide-angle phase and zoom-in phase. In the wide-angle phase, the temporal differencing method was used to detect lecturer. The lecturer detection of zoom-in phase is detected by Sobel edge detection and optimal thresholding method with iteration. We improved the GMM to construct the background model for handwritten data detection. After the construction, we used the background model to detect handwritten data by background subtraction method. The camera is controlled through the two pre-defined parameters, the results of lecturer detection and handwritten data detection. In the experimental study, the video recorded by ACRS was compared with that recorded manually by a cameraman. About 83% of the video time was close to the video recorded manually and suitable for review by students.

並列關鍵字

Automated lecture capture ； video recording system ； object tracking

參考文獻

[17] Chou, H. P., Wang, J. M., Fuh, C. S., Lin, S. C., and Chen, S. W. 2010. Automated lecture recording system. In Proceedings of 2010 International Conference on System Science and Engineering (ICSSE''10). IEEE, Taipei, Taiwan, 167-172.

[8] Lin, Y. T., Tsai, H. Y., Chang, C. H., and Lee, G. C. 2010. Learning-focused structuring for blackboard lecture videos. In Proceedings of the 2010 IEEE Fourth International Conference on Semantic Computing (ICSC''10). IEEE, Berkeley, CA, USA, 149-155.

[16] Wei, J. H., Wang, L. P., Li, M. Y., and Wang, Y. W. 2009. Advantages and deficiencies of the automated multimedia lecture recording system in lecture video production. In Proceedings of the 2009 International Forum on Computer Science-Technology and Applications (IFCSTA''09). IEEE, Chongqing, China, 271-273.

[6] Dickson, P. E., Adrion, W. R., and Hanson, A. 2006. Automatic capture of significant points in a computer based presentation. In Proceedings of the Eighth IEEE International Symposium on Multimedia (ISM''06). IEEE, San Diego, CA, USA, 6 pages.

[9] Friedland, G. and Rojas, R. 2006. Human-centered webcasting of interactive-whiteboard lectures. In Proceedings of the Eighth IEEE International Symposium on Multimedia (ISM''06). IEEE, San Diego, CA, USA, 6 pages.