臺灣博碩士論文加值系統

近年來，因為駕駛員疲勞導致了許多安全事故的發生，往往對交通安全造成很大的傷害，如何能提前警示駕駛員避免發生意外便成為了重要的議題。

本論文提出了採用光體積變化描記圖之瞌睡偵測系統，是一種非接觸式的量測方式，當webcam拍攝人臉時，利用血紅素吸收光線的變化量來轉換成心跳信號，然後使用心率變異度(Heart rate variability, HRV)在頻域上的分析，藉由我們的演算法來進行計算與判斷，能提早發現駕駛員是否有瞌睡情況，而當我們系統偵測到瞌睡狀態發生的情況下，會提前發出警示鈴聲，提醒駕駛員，有利於避免交通意外的發生。

本系統在LINUX環境下開發人機介面(Graphical User Interface, GUI)，能及時觀察到PPG的波形和低高頻功率的比值(LF/HF-ratio)。

In recent years, fatigue driving led to the occurrence of many accidents, which often caused great harms. How to warn the driver in advance to avoid accidents has become an important issue.

In this thesis, a drowsiness detection system is proposed by using PPG Imaging. It is a non-contact measurement method. When the webcam captures the human face, hemoglobin absorption of light can be changed into heartbeat signal. And then use the heart rate variability (Heart rate variability, HRV) in the frequency domain analysis. By the proposed algorithm to calculate and judge, early drowsiness detection can be achieved. When the proposed system detects the occurrence of drowsiness, it will send out the warning ring to warn the driver. It is a good help to avoid traffic accidents.

Graphical user interface (GUI) is developed under Linux Operation System (OS). It can immediately display PPG waveform and low frequency / high frequency ratio(LF/HF-ratio).

誌謝 i
中文摘要 ii
Abstract iii
目錄 iv
圖目錄 vi
表目錄 viii
第1章 緒論 1
1-1 研究背景與動機 1
1-2 論文架構 2
第2章 文獻探討 3
2-1 光體積變化描記圖(PPG) 3
2-2 光體積變化描記圖影像 4
2-3 PERCLOS介紹 6
2-4 心率變異度介紹 7
2-4-1 心跳區間 7
2-4-2 時域分析 8
2-4-3 頻域分析 10
2-5 瞌睡偵測介紹 11
第3章 瞌睡偵測系統演算法架構與模擬 13
3-1 系統演算法架構 13
3-2 心跳區間偵測(RRI detection) 14
3-3 瞌睡判斷 15
3-3-1 清醒跟瞌睡狀態 15
3-3-2 判斷條件 16
3-4 瞌睡偵測演算法 18
3-5 瞌睡準則(Drowsiness Rules) 20
3-5-1 瞌睡準則1(Rule 1) 21
3-5-2 瞌睡準則2(Rule 2) 22
3-5-3 瞌睡準則3(Rule 3) 23
3-5-4 瞌睡準則4(Rule 4) 24
3-5-5 瞌睡準則5(Rule 5) 24
3-5-6 瞌睡準則6(Rule 6) 26
3-6 實驗結果 27
第4章 系統設計與實作 40
4-1作業系統介紹 40
4-2軟體介紹 41
4-3攝影機介紹 42
4-4 GUI介面 43
第5章 結果與討論 44
參考文獻 45

[1]呂宗憲、林筱莉、陳智傑, ECG與PPG信號之相關性研究, 逢甲大學自動控制工程學系畢業專題, 2006
[2]邱裕仁、楊哲彰, 以血氧連續脈動訊號擷取心率訊號, 元智大學老人福祉科技研究中心, 2009
[3]M.-Z. Poh, D. J. McDuff, and R. W. Picard, "Advancements in noncontact, multiparameter physiological measurements using a webcam," IEEE Transactions on Biomedical Engineering, vol. 58, pp. 7-11, 2011.
[4]D. Dimage and R. Perclos, "A valid psychophysiological measure of alertness as assessed by psychomotor vigilance," Report No. FH-WA2MCRT2982006, Federal Highway Administration, Office of Motor Carriers1998.
[5]L. Tijerina, W. Wierwille, M. Goodman, S. Johnston, D. Stoltzfus, and M. Gleckler, A Preliminary Assessment of Algorithms for Drowsy and Inattentive Driver Detection on the Road, National Highway Traffic Safety Administration, 1999.
[6]W. W. Wierwille, S. Wreggit, C. Kirn, L. Ellsworth, and R. Fairbanks, "Research on vehicle-based driver status/performance monitoring; development, validation, and refinement of algorithms for detection of driver drowsiness," National Highway Traffic Safety Administration Final Report: DOT HS 808 247, 1994.
[7]http://medi-core.com/download/HRV_clinical_manual_ver3.0.pdf
[8]翁根本,何慈育,歐善福,林竹川,謝凱生,“心率變動性分析,”台灣醫界, vol. 52, no. 6, 2009.
[9]唐善美, Relationship Between Heart Rate Variability and Anxiety Among Patients Receiving Cardiac Catheterization, 國立成功大學醫學院護理學系碩士學位論文, 2004.
[10]動態心電圖對心率變異性的分析
http://www.ceheart.com/Article/Print.asp?ArticleID=203&Page=1
[11]H. A. Rahima, A. Dalimia, and H. Jaafarb, "Detecting Drowsy Driver Using Pulse Sensor," 2015.
[12]高堉雅, Implementation of Intelligent Drowsiness Detection Warning System, 國立中興大學電機工程學系碩士學位論文, 2014.
[13]林勤恆, Contact-Free Measurement of Cardiac Pulse using IR Video Imaging, 國立中興大學電機工程學系碩士學位論文, 2015.
[14]李瑋晟, Design and Embedded System Real-time Implementation of Low-complexity Drowsy Driver Detection System, 國立中興大學電機工程學系碩士學位論文, 2015.
[15] 坂卷佳壽美, 看圖瞭解數位信號處理, 建興文化事業有限公司.
[16]維基百科 Linux https://zh.wikipedia.org/wiki/Linux
[17]【介紹】What is Qt? http://program-lover.blogspot.tw/2008/08/qt_21.html