$\require{mediawiki-texvc}$

연합인증

연합인증 가입 기관의 연구자들은 소속기관의 인증정보(ID와 암호)를 이용해 다른 대학, 연구기관, 서비스 공급자의 다양한 온라인 자원과 연구 데이터를 이용할 수 있습니다.

이는 여행자가 자국에서 발행 받은 여권으로 세계 각국을 자유롭게 여행할 수 있는 것과 같습니다.

연합인증으로 이용이 가능한 서비스는 NTIS, DataON, Edison, Kafe, Webinar 등이 있습니다.

한번의 인증절차만으로 연합인증 가입 서비스에 추가 로그인 없이 이용이 가능합니다.

다만, 연합인증을 위해서는 최초 1회만 인증 절차가 필요합니다. (회원이 아닐 경우 회원 가입이 필요합니다.)

연합인증 절차는 다음과 같습니다.

최초이용시에는
ScienceON에 로그인 → 연합인증 서비스 접속 → 로그인 (본인 확인 또는 회원가입) → 서비스 이용

그 이후에는
ScienceON 로그인 → 연합인증 서비스 접속 → 서비스 이용

연합인증을 활용하시면 KISTI가 제공하는 다양한 서비스를 편리하게 이용하실 수 있습니다.

In-Vehicle AR-HUD System to Provide Driving-Safety Information 원문보기

ETRI journal, v.35 no.6, 2013년, pp.1038 - 1047  

Park, Hye Sun (IT Convergence Technology Research Laboratory, ETRI) ,  Park, Min Woo (Virtual Reality Laboratory, Kyungpook National University) ,  Won, Kwang Hee (Virtual Reality Laboratory, Kyungpook National University) ,  Kim, Kyong-Ho (IT Convergence Technology Research Laboratory, ETRI) ,  Jung, Soon Ki (Virtual Reality Laboratory, Kyungpook National University)

Abstract ▼ AI-Helper 아이콘AI-Helper

Augmented reality (AR) is currently being applied actively to commercial products, and various types of intelligent AR systems combining both the Global Positioning System and computer-vision technologies are being developed and commercialized. This paper suggests an in-vehicle head-up display (HUD)...

주제어

#AR, augmented reality   #HUD, head-up display   #vehicle and pedestrian recognition   #stereo camera   #SVM, support vector machine   #HOG, histograms of oriented gradients   #driving-safety information   #in-vehicle intelligent information system  

AI 본문요약
AI-Helper 아이콘 AI-Helper

* AI 자동 식별 결과로 적합하지 않은 문장이 있을 수 있으니, 이용에 유의하시기 바랍니다.

제안 방법

  • Based on the problems mentioned above, current AR-HUD systems should be further developed, considering driver safety. For the purpose of convenience in this experiment, the proposed system is designed to use a seethrough display installed in front of the passenger seat, fitting the line of vision of the person in the passenger seat. If the proposed system is developed to completion for the commercial market, the display will instead be installed on the driver’s side.
  • For vehicle recognition, we perform an experiment in which a vehicle equipped with the proposed system drives at a speed of 70 km/h to 80 km/h in more than two lanes, and detects other vehicles within 40 m to 45 m of its radius. For the pedestrian recognition, we perform an experiment in which a vehicle equipped with the proposed system drives at a speed of less than 30 km/h along a narrow road in a residential district and detects pedestrians within a range of around 30 m of its radius.
  • The vision-based method cannot generally detect the obstacles in severe weather. However, we perform the experiments to measure the robustness of the proposed method in rainy weather. Table 1 shows the results of the vehicle recognition experiments in sunny and rainy weather.
  • A highway is usually a vehicle-only road. In the case of the highway, we conduct a comparative analysis according to the weather conditions. The proposed method performs the vision-based forward obstacle detection.
  • In the case of the highway, we conduct a comparative analysis according to the weather conditions. The proposed method performs the vision-based forward obstacle detection. It performs the alert and notifies the warning for collision information.
  • The proposed method seeks out the ground from the images input from stereo cameras and detects the foreground region by removing the ground from the image. Within the detected foreground region, the system then detects pedestrians and vehicles based on their feature points, learns and classifies the detected objects through an SVM, and recognizes the vehicles and pedestrians from the input images in real time.
  • The proposed method uses the two steps of hypothesis generation and hypothesis verification. The method cannot increase the processing time.
  • The proposed system can perform real-time obstacle detection in sunny or rainy weather. However, it has not been verified that the proposed system can likewise perform in severe weather or in the nighttime.
  • The proposed system, in which an HUD system is combined in a vehicle with AR technology, matches augmented drivingsafety information with the information on other vehicles and pedestrians in the real world and offers the information to the driver in real time through a transparent display installed in front of the driver, fitting the driver’s view.
  • Thus, this paper introduces an in-vehicle AR-HUD system that detects such driving-safety information and offers the information to the driver, fitting the driver’s viewpoint, as shown in Fig. 1.
  • To maintain a safe distance behind other vehicles and avoid collisions, the proposed system informs and warns the driver of detected obstacles using various mounted devices, under consideration of the previously mentioned factors. Through this system function, the number of traffic accidents can be significantly reduced.
  • The proposed method seeks out the ground from the images input from stereo cameras and detects the foreground region by removing the ground from the image. Within the detected foreground region, the system then detects pedestrians and vehicles based on their feature points, learns and classifies the detected objects through an SVM, and recognizes the vehicles and pedestrians from the input images in real time. In this paper, we proposed an obstacle detection method based on the road geometry.

대상 데이터

  • Figure 7 shows the images used in the learning data for the vehicle recognition module. A total of 217 images are used as positive data, and 410 images are used as negative data. The learning data used in the pedestrian recognition module is from the INRIA person dataset, with 2,416 images used as positive data and 12,180 images used as negative data.
  • The cameras used for this experiment are GS2-FW14S5 models from the Point Grey Research Company, which are 8 mm cameras with a resolution of 1038 × 1036 and can obtain an image at a speed of 30 fps.

이론/모형

  • For this module, the input images and ground obstacle images are needed for detection. A ground obstacle is computed using the billboard sweep stereo matching algorithm by the ground obstacle detection module [35]. Obstacles are deemed dangerous according to height.
본문요약 정보가 도움이 되었나요?

참고문헌 (41)

  1. S.H. Lee, J. Choi, and J. Park, "Interactive e-Learning System Using Pattern Recognition and Augmented Reality," IEEE Trans. Consum. Electron., vol. 55, no. 2, May 2009, pp. 883-890. 

    상세보기 crossref

  2. B. Choi et al., "A Metadata Design for Augmented Broadcasting and Testbed System Implementation," ETRI J., vol. 35, no. 2, Apr. 2013, pp. 292-300. 

    원문보기 상세보기 crossref

  3. N. Parimal et al., "Application of Sensors in Augmented Reality Based Interactive Learning Environments," IEEE Int. Conf. ICST, Dec. 2012, pp. 173-178. 

  4. M.A. Oikawa et al., "AR-Based Video-Mediated Communication: A Social Presence Enhancing Experience," 14th Int. Symp. VR, May. 2012, pp. 125-130. 

  5. Y.M. Aung and A. Al-Jumaily, "AR Based Upper Limb Rehabilitation System," IEEE Int. Conf. BioRob., June 2012, pp. 213-218. 

  6. T. Zysk, J. Luce, and J. Cunningham, "Augmented Reality for Precision Navigation," GPS World Mag., vol. 23, issue 6, June 2012, pp. 47-51. 

  7. H. Chen et al., "Application of Augmented Reality in Engineering Graphics Education," IEEE Int. Symp. ITME 2, art. no. 6132125, Dec. 2011, pp. 362-365. 

  8. Y.C. Cheng et al., "AR-Based Positioning for Mobile Devices," Int. Conf. Parallel Process., no. 6047276, Sept. 2011, pp. 63-70. 

  9. J.J. Mccorvey, "Using Augmented Reality to Market Your Business," Inc., Apr. 2010. http://www.inc.com/magazine/ 20100401/using-augmented-reality-to-market-your-business.html 

  10. Siemens, "From Sedans to Compact Cars - One HUD Fits All," press photo, reference no. 200605004, Siemens AG, Munich, Germany, 2006. http://www.siemens.com/press/en/presspicture/? press/en/pp_sv/2006/sv200605004_(hud)_1376467.htm 

  11. J. Lincoln, D. Lincoln, and M. Carmean, How a Laser HUD Can Make Driving Safer, MicroVision Program Brief, MicroVision, Inc., Mar. 2007. 

  12. R. Parasuraman, R. Molloy, and I.L. Singh, "Performance Consequences of Automation-Induced Complacency," Int. J. Aviation Psychology, vol. 3, no. 1, 1993, pp. 1-23. 

    상세보기 crossref

  13. NHTSA, Measuring Distraction Potential of Operating In-Vehicle Devices, no. DOT-HS-811-231, Dec. 2008. 

  14. S.C. Lee, "Psychological Effects on Aged Driver's Traffic Accidents," Korea J. Psychological Soc. Issues, vol. 12, 2006, pp. 149-167. 

  15. Y. Zhang, Y. Owechko, and J. Zhang, "Learning-Based Driver Workload Estimation," Comput. Intell. Automobile Appl., Ser. Studies Comput. Intell., vol. 132, 2008, pp. 1-24. 

  16. E.R. Michael, J.G. Leo, and J.W. Nicholas, "Effects of Naturalistic Cell Phone Conversations on Driving Performance," Int. J. Safety Research, vol. 35, 2004, pp. 453-464. 

    상세보기 crossref

  17. E. Adella, A. Varhely, and M. Fontana, "The Effects of a Driver Assistance System for Safe Speed and Safe Distance: A Real-Life Field Study," Int. J. Transp. Research Part C 19, 2011, pp. 145-155. 

    상세보기 crossref

  18. E. Johan, J. Emma, and O. Joakim, "Effects of Visual and Cognitive Load in Real and Simulated Motorway Driving," Int. J. Transp. Research Part F, vol. 8, 2005, pp. 97-120. 

    상세보기 crossref

  19. R. Venkatasawmy, The Digitization of Cinematic Visual Effects: Hollywood's Coming of Age, UK: Press Lexington Books, 2013. 

  20. A.S. Cohen and R. Hirsig, Feed Forward Programming of Car Driver's Eye Movement Behavior: A System Theoretical Approach, final technical report, vol. 2, 1980, pp. 1-223. 

  21. V. Charissis and S. Papanastasiou, "Human-Machine Collaboration through Vehicle Head Up Display Interface," Int. J. Cognition, Technol., Work, vol. 12, no. 1, 2010, pp. 41-50. 

    상세보기 crossref

  22. Y.C. Shinko, D. Anup, and M.T. Mohan, "Active Heads-up Display Based Speed Compliance Aid for Driver Assistance: A Novel Interface and Comparative Experimental Studies," IEEE Int. Symp. IV, 2007, pp. 594-599. 

  23. A. Sato et al., "Visual Navigation System on Windshield Head-Up Display," Proc. 13th World Congress ITS, 2006, pp. 167-175. 

  24. H. Watanabe et al., "The Effect of HUD Warning Location on Driver Responses," Proc. 6th World Congress ITS, 1999, pp. 1-10. 

  25. C. Jablonski, "An Augmented Reality Windshield from GM," Mar. 2010. http://www.zdnet.com/blog/emergingtech/anaugmented- reality-windshield-from-gm/2164 

  26. V. Poortere et al., "Efficient Pedestrian Detection: A Test Case for SVM Based Categorization," Proc. ICVW, Sept. 2002, pp. 19-20. 

  27. D.M. Gavrila and V. Philomin, "Real-Time Object Detection for Smart Vehicles," IEEE Int. Conf. CVPR, 1999, pp. 87-93. 

  28. S. Hermann and R. Klette, "Iterative Semi-Global Matching for Robust Driver Assistance Systems," Int. Conf. ACCV, vol. 7726, 2012, pp. 456-478. 

  29. H. Hattori et al., "Stereo-Based Pedestrian Detection Using Multiple Patterns," Int. Conf. BMVC, 2009, pp. 1-10. 

  30. M. Bajracharya et al., "A Fast Stereo-Based System for Detecting and Tracking Pedestrians from a Moving Vehicle," Int. J. Robotics Research, vol. 28, 2009, pp. 1466-1485. 

    상세보기 crossref

  31. A. Shashua, Y. Gdalyahu, and G. Hayun, "Pedestrian Detection for Driving Assistance Systems: Single-Frame Classification and System Level Performance," IEEE Int. Symp. IV, 2004, pp. 1-6. 

  32. S. Munder and D.M. Gavrila, "An Experimental Study on Pedestrian Classification," IEEE Trans. Pattern Anal. Mach. Intell., vol. 28, no. 11, 2006, pp. 1863-1868. 

    상세보기 crossref

  33. Australian Transport Council, The National Road Safety Strategy 2011-2020, 2011, pp. 1-108. 

  34. K.H. Won and S.K. Jung, "Billboard Sweep Stereo for Obstacle Detection in Road Scenes," Int. J. IET, vol. 48, issue 24, Nov. 2012, pp. 1528-1530. 

  35. M.W. Park, K.H. Won, and S.K. Jung, "Vehicle Detection and Tracking Using Billboard Sweep Stereo Matching Algorithm," Korea J. KMMS, vol. 16, no. 6, 2013, pp. 764-781. 

  36. Z.H. Zhou and X. Geng, "Projection Functions for Eye Detection," Int. J. Pattern Recognition, vol. 37, no. 5, 2004, pp. 1049-1056. 

    상세보기 crossref

  37. C. Cortes and V. Vapnik, "Support-Vector Networks," Int. J. Mach. Learning, vol. 20, no. 3, 1995, pp. 273-297. 

  38. T. Joachims, "Making Large-Scale SVM Learning Practical," Advances in Kernel Methods - Support Vector Learning, B. Scholkopf, C. Burges, and A. Smola, Eds., Cambridge, MA: MIT Press, 1999. 

  39. K.A. Paul and H. Ayman, "A Comparative Analysis between Rigorous and Approximate Approaches for LiDAR System Calibration," Korea J. KSGPC, vol. 30, no. 6, 2012, pp. 593-605. 

  40. Wikipedia, Precision and recall. http://en.wikipedia.org/wiki/ Precision_and_recall 

  41. Gamedev.net, Calculate framerate with GetTickCount function. http://www.gamedev.net/topic/621703-calculate-framerate-withgettickcount/ 

저자의 다른 논문 :

LOADING...

관련 콘텐츠

오픈액세스(OA) 유형

BRONZE

출판사/학술단체 등이 한시적으로 특별한 프로모션 또는 일정기간 경과 후 접근을 허용하여, 출판사/학술단체 등의 사이트에서 이용 가능한 논문

저작권 관리 안내
섹션별 컨텐츠 바로가기

AI-Helper ※ AI-Helper는 오픈소스 모델을 사용합니다.

AI-Helper 아이콘
AI-Helper
안녕하세요, AI-Helper입니다. 좌측 "선택된 텍스트"에서 텍스트를 선택하여 요약, 번역, 용어설명을 실행하세요.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.

선택된 텍스트

맨위로