抬頭顯示器上目標重疊視訊之位移追蹤補償機制

Offset Compensation Mechanism on Target-Video Overlapping of Head-Up Display

游智宇(Jhih-Yu You)

指導教授 ： 張嘉銘

大同大學/電機資訊學院/資訊工程學系所/碩士(2010年)

摘要

行車安全一直是一個重要的議題，在將抬頭顯示器(Head Up Display)應用於行車資訊系統的應用中，因為所有的車況指示資訊都顯示在眼睛前方，可減少頭部擺動及未注視到前方路況的頻率，進而提升行車的安全性。
但抬頭顯示器顯示之資訊位置與駕駛者之所視位置的資訊是否一致，則是安全性是否有提升的最大考量。因為資訊取得的不一致，可能會造成駕駛者對於車況或路況之位置資訊產生誤判，因此造成抬頭顯示器顯示之訊息與視線所看到的現實狀況不一致。
本系統使用攝影機擷取駕駛之影像，應用OpenCV偵測眼睛，不需戴上其它額外之感測器以感測眼睛之位置，使目標物、抬頭顯示器上顯示之附加資訊標示，能夠保持與眼睛注視方向重疊一致，並可因應不同使用者動態追蹤位移補償機制，進而確保提升行車之穩定與安全性。

關鍵字

Head Up Display ； OpenCV ； 眼睛偵測

並列摘要

The traffic safety is always an important issue. In the application of Head-Up Display (HUD) to display the driving information, since all vehicle conditions are shown in front of the driver's eyes, the frequency of forehead swinging and not gaze at front the state of roads is reduced. Therefore, the driving security is improved.
But the position of information in HUD and target that driver watched to be consistent, is most considered about the promotion of security. Because the display information is inconsistent, it may confuse driver's perception of the condition of vehicle and status of road. That is the information shown on the HUD and the line of sight to realistic condition will be inconsistent.
This system uses camera to pick up driver's picture and applies the OpenCV to detect eyes. Other extra sensors are not needed to sense eye. The system may cause on the targe, the information on the HUD can be maintained with the eye gaze direction in line. It also calibrates the dynamic offset compensation mechanism due to the different user to improve driving stability and the security.

並列關鍵字

Head Up Display ； OpenCV ； Eye Detect

參考文獻

[1] H.C . Lai,M. Savvides,and T.Chen,“Proposed FPGA Hardware Architecture for High Frame Rate (>100 fps) Face Detection Using Feature Cascade Classifiers,” IEEE International Conference on Biometrics: Theory,Applications,and Systems, Sept. 2007.

[3] J. Zhou,et al., “ Haar-like Features Based Eye Detection Algorithm and Its Implementation on TI TMS320DM6446 Platform,” IEEE Internation Workshop on Imaging Systems and Techniques, May 2009.

[4] J.C.Yang,et al., “Evaluation of an Eye Tracking Technology for 3D Display Applications,” 3DTV Conference on Transmission and Display of 3D Video, May 2008.

[5] J. Xiao and H. Feng, “ A Low-cost Extendable Framework for Embedded Smart Car Security System,” ICNSC ’09. International Conference on Networking,Sensing and Control,2009,March 2009.

[6] J. Zhou,et al., “ Haar-like Features Based Eye Detection Algorithm and Its Implementation on TI TMS320DM6446 Platform,” IEEE International Workshop on Imaging Systems and Techniques, May 2009.