臺灣博碩士論文加值系統

行車的安全是非常重要的事，在長時間專注的駕駛情況，駕駛者容易
疲勞，若是駕駛者有疲勞駕駛的情況對於自己及周遭的人事物都是非常危
險的。
本論文介紹了駕駛疲勞監測的研究現況後，並對駕駛疲勞監測的原理
與整體的實踐方法進行說明。首先，應用Viola-Jones或灰度投影方法來定
位眼睛，最後再根據PERCLOS(Percentage of Eyelid Closure over the Pupil
over Time)判斷駕駛者是否疲勞狀態，若判定是疲勞狀態，則發出警訊提醒
駕駛者。
依據上述的方式，我們利用MATLAB/Simulink 規劃系統並評估方法，
以實現整個系統。考量即時性效率以及計算的複雜度，因此選擇以灰度投
影方式結合在TI DSP 平台DM6437 之中，實現了駕駛疲勞監測系統。並
將開發版的上電自動執行系統的功能完成，以符合實際應用需求。實驗結
果可以證實，本系統已不妨礙駕駛者，並即時的判斷駕駛者是否有疲勞駕
駛的行為。

Driving behavior assessment is a main issue for traffic safety; especially
driver fatigue is a significant risk factor in public transportation.
In this thesis, the recent researches of driving fatigue monitoring are
introduced, the overall principles are explained, and the implementation of
driver fatigue monitoring is practiced. First, we use Viola-Jones or integral
projection to detect eyes. In the last, based on Percentage of Eyelid Closure
over the Pupil over Time (PERCLOS) drowsiness metric was established, the
detection system can determine whether the driver is in a fatigued state or not.
If yes, the buzzer will sound off to alert the driver.
According to the manner described above, we first use the MATLAB /
Simulink development program and assessment methods to design and
implement the whole system. For real-time application, a TI DM6437 platform
is integrated with peripheral devices to realize the driving fatigue monitoring
system. Due to computing complexity and efficiency, only the integral
projection method was deployed in the DSP platform. The experimental results
show that the system in real-time measures is not subject to weather restrictions
and can determine whether the driver is exhibiting fatigue-driving behavior or
not without contacting the human body.

明志科技大學碩士學位論文指導教授推薦書 i
明志科技大學碩士學位論文口試委員審定書 ii
明志科技大學學位論文授權書 iii
誌 謝 iv
中文摘要 v
Abstract vi
目　錄 vii
表目錄 x
圖目錄 xi
第一章 緒論 1
1.1 研究動機 1
1.2 研究方向 2
1.2.1 疲勞原因與反應特徵 2
1.2.2 駕駛疲勞背景研究 2
1.3 論文架構 6
第二章 人眼定位及疲勞監測 7
2.1 光源補償 7
2.1.1 紅外線光源 7
2.1.2 紅外線使用疑慮 9
2.1.3 電路設計與實體 10
2.2 Viola-Jones 12
2.3 色彩空間 16
2.3.1 RGB色彩空間 16
2.3.2 YUV色彩空間 16
2.4 灰階圖像 17
2.5 動態追蹤 18
2.5.1 背景相減法 18
2.6 二值化 20
2.7中值濾波 21
2.8 型態學 22
2.8.1 膨脹(Dilation) 22
2.8.2侵蝕(Erosion) 23
2.8.3閉合(Closing) 23
2.8.4斷開(Opening) 24
2.9 連通標記法 25
2.10積分投影 27
2.11 PERLCOS 29
第三章 駕駛疲勞監測系統實作 31
3.1 駕駛疲勞監測系統架構 31
3.2 實作環境 32
3.2.1 光源模組 33
3.2.2 CCD攝影機 34
3.2.3 MATLAB/Simulink 36
3.3 Viola-Jones駕駛疲勞監測系統實作 37
3.4 以灰度投影的方式駕駛疲勞監測系統實作 40
3.5 實作成果分析 46
第四章 監測系統於DSP平台實現 48
4.1 德州儀器硬體開發平台TMS320DM6437 EVM 48
4.2 德州儀器軟體開發平台Code Composer Studio 50
4.2.1 CCS軟體開發平台與除錯功能 51
4.3 MATLAB/Simulink與CCS、DM6437EVM連接應用 53
4.4 駕駛疲勞監測系統架構 56
4.5 於DSP平台實現駕駛疲勞監測系統 57
4.6 上電自動執行 62
第五章 結論 68
5.1 結論 68
5.2 未來展望 69
參考文獻 70

[1]http://www.thb.gov.tw/TM/Default.aspx , 中華民國交通部.
[2]張光雄, “主動式紅外線影像的疲勞駕駛偵測系統, ”國立中央大學資訊工程研究所,2005年6月.
[3]Hamada, T., T. Ito, K. Adachi, T. Nakano, and S. Yamamoto, "Detecting method for drivers' drowsiness applicable to individual features," in Proc. IEEE Int'l Conf. Intelligent Transportation Systems, Oct.12-15, 2003,vol.2, pp.1405-1410.
[4]Haro, A., M. Flickner, and I. Essa, "Detecting and tracking eyes by using their physiological properties, dynamics, and appearance," in Proc. IEEE Conf. Computer Vision and Pattern Recognition, Hilton Head Island, SCJun.13-15, 2000, vol.1, pp.163-168.
[5]J.R. Parker, Algorithms for Image Processing and Computer Vision, USA: Baker & Taylor Books, 2010.
[6]Gonzalez & Woods, Digital Image Processing, 3rd Edition, Prentice Hall, 2008.
[7]R. Knipling, and P. Rau,“PERCLOS: A Valid Psychophysiological Measure of Alertness as Assessed by Psychomotor Vigilance, ”U.S. Department of Transportation, Federal Highway Administration,1998.
[8]U. Trutschel, B. Sirois, D. Sommer, M. Golz, and D. Edwards, “PERCLOS: An Alertness Measure of the Past, ” Proceedings of the Sixth International Driving Symposium on Human Factors in Driver Assessment, Training and Vehicle Design,2011, pp.172~179.
[9]Texas Instruments Inc., “xDAIS-DM (Digital Media) User Guide”, Literature Number SPRUEC8B, January 2007.
[10]Spectrum Digital Inc., “DM6437 EVM Technical Reference”, Manual No. 509105-0001 Rev. C, Dec 2006.
[11]Texas Instruments Inc., "TMS320DM6437 Digital Media Processor", Literature Number SPRS345D, June 2008.
[12]Texas Instruments Inc., "Code Composer Studio Development Tools v3.3 Getting Started Guide", Literature Number SPRU509H, May 2008.
[13]Texas Instruments Inc., “Code Composer Studio Development Tools v3.3,” Literature No. spru509h, October 2006.
[14]Texas Instruments Inc., "DaVinci Technology Background and
Specifications", Literature Number SPRT401A, January 2007.
[15]盧怡仁、蔡偉和,初版, “數位信號處理平台在嵌入式系統的應用,” 台
北：文魁資訊股份有限公司, 2006。
[16]王志中,” 網際網路電視系統實作於德儀數位訊號處理器,” 明志科技大學論文, 2010。