植基於臉部特徵與眨眼檢測之線上學習專注力評估系統

Attention Evaluation of Online Learning System Based on Facial Features and Blink Detection

劉育名(Yu-Ming Liu)

指導教授 ： 徐豐明

國立臺中科技大學/設計學院/多媒體設計系碩士班/碩士(2021年)

摘要

2020年因應COVID-19（新型冠狀病毒肺炎）疫情席捲全球，大部分學生受到此影響無法在校進行學習，此波疫情賦予了遠距教學更多的應用，而這樣的影響卻也推動學術機構發展與投入新的線上課程資源。遠距教學固然方便，與此同時疫情也帶給學生學習就業上的新挑戰，此時本研究是使用人工智慧與影像辨識的技術進行探討，結合上述技術來讓使用者進行評估，經由資料統計並透過影像分析判斷多數使用者專注力的系統就能夠達到實質上的幫助，使用者們經由視訊鏡頭以獲得臉部圖像，透過臉部特徵辨識與眨眼檢測綜合分析之後，讓授課者得到課堂中學生的專注力分析；也能夠達成使用者們本身的自我檢測，經由這些數據來得到評量與自我檢測的目的，經過實測，本研究達成91.4%的人臉辨識率再經由眨眼檢測獲得高於九成的專注度檢測準確率，適合應用於線上學習的檢測評估。

關鍵字

課堂觀察法 ； 線上學習 ； 專注度評估 ； 臉部特徵辨識 ； 眨眼檢測

並列摘要

In 2020, in response to the COVID-19 (new coronavirus pneumonia) epidemic that has swept the world, most students have been affected by this and cannot study in school. This wave of epidemic has given a new definition of distance teaching, and this impact has promoted academics. The institution develops and invests in new online course resources. Although remote teaching is convenient, the epidemic also brings new challenges to students’ learning and employment. At this time, this thesis uses artificial intelligence and image recognition technologies to discuss, combining the above technologies to allow users to evaluate, the system that judges the concentration of most users through data statistics and image analysis can achieve substantial help. Users obtain facial images through video lenses, and through facial feature recognition and blinking. After the comprehensive analysis of the test, the lecturer can obtain the concentration analysis of the students in the classroom; it can also achieve the self-test of the users themselves, and the purpose of evaluation and self-test can be obtained through the data. After the actual test, this research achieves a face recognition rate of 91.4%, and then obtains a concentration detection accuracy rate higher than 90% through blink detection, which is suitable for detection and evaluation of online learning.

並列關鍵字

classroom observation ； online learning ； attention assessment ； facial features recognition ； blink detection

參考文獻

[1] T. Mita, T. Kaneko and O. Hori, Joint haar-like features for face detection, Tenth IEEE International Conference on Computer Vision, 1619-1626, 2005.

[2] P. Viola and M. J. Jones. Robust real-time face detection, International Journal of Computer Vision, vol. 57, no. 2, 137-154, 2004.

[3] K. He, X. Zhang, S. Ren and J. Sun. Deep Residual Learning for Image Recognition,"IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 770-778, 2016.

[4] M. A. Hearst, S. T. Dumais, E. Osuna, J. Platt and B. Scholkopf. Support vector machines, IEEE Intelligent Systems and their Applications, vol: 13, 1998.

[5] G. Levi and T. Hassner. Emotion recognition in the wild via convolutional neural networks and mapped binary patterns, in Proceedings of the ACM on international conference on multimodal interaction, 503-510, 2015.