- 會議論文
閱讀疲勞即時檢測系統
Instant Detection System for Reading Fatigue
摘要
在這個科技越來越進步的年代，學習的方法越來越多元，也不再局限於教室內，能夠藉由網路獲得的資訊相當豐富，相對的也更容易出現閱讀疲勞的情況。一旦出現疲勞不僅會導致專注力無法集中，閱讀所獲得的訊息也會大幅減少，使得學習的效率大大的降低。因此本研究想藉由人臉檢測技術與YOLO物件偵測技術來偵測長時間學習所帶來的疲勞狀態。本研究將實作出一套閱讀疲勞檢測系統，透過攝像鏡頭取得使用者臉部的即時影像，再將影像輸入這套由68特徵原理人臉關鍵點檢測與YOLO物件偵測技術組成的系統，人臉關鍵點檢測使用者眼部開閉狀態來判斷疲勞狀態，YOLO物件偵測則是透過提前蒐集的圖像做訓練，對使用者眼部的狀態來判斷疲勞狀態，如果使用者出現疲勞狀態系統會發出提示音效，提示須適度休息。
並列摘要
In this era of more and more advanced technology, learning methods are becoming more and more diverse, and they are no longer limited to classrooms. The information that can be obtained through the Internet is quite rich, and it is relatively more prone to reading fatigue. Once fatigue occurs, not only will it lead to inability to concentrate, but the information obtained by reading will also be greatly reduced, which will greatly reduce the efficiency of learning. In this study, a reading fatigue detection system will be implemented. The real-time image of the user's face is obtained through the camera lens, and the user's face state is used to determine the fatigue state. Moderate rest is required.