Discussions

The experimental results of the proposed tour guidance system presented previously show that it is feasible to use the omni-camera equipped on the vehicle to detect the vertical line features for locating the vehicle by using the vertical-line features. Then, it is also practical to generate the passenger-view image and augment the building name on it for display on user’s mobile–device screen.

However, the proposed system still has some problems. If the features in the

environment are too few for the system to detect, the system might have errors because erroneous localization of the vehicle. In more detail, if the distance between two features is too large, then the system will use the estimated vehicle speed to locate the vehicle for a long distance. The error of the location will possibly accumulate to a very large amount and make the system show wrong AR information on the user’s mobile-device screen. Therefore, the path we choose is important for the system.

Furthermore, if the network of the system breaks down, then the system cannot work because the main process is conducted on the remote server computer. If the system on the vehicle cannot connect to the server, the server cannot get the image and the system will stop. In this case, the system can send the AR image to the mobile device, but the user cannot see anything on the device. So the stability of the network is important for running the proposed system.

Chapter 8

Conclusions and Suggestions for Future Works

8.1 Conclusions

A tour guidance system by augmenting reality techniques for uses in outdoor environments by using an omni-camera imaging device has been proposed. To design such a system, several techniques have been proposed as summarized in the following.

1. A method for learning the environment map has been proposed, which generates the environment map for the system to locate the vehicle position and compute the position of the building.

2. A method for detecting vertical lines in omni-images has been proposed, which is based on the canny edge detector and detects vertical lines, complete or broken, with widths.

3. A method for locating the vehicle along the path has been proposed, which uses the result of vertical-line detection to conduct vertical-line matching by the LCS algorithm to provide the position of the vehicle on a map.

4. A method for generating the passenger-view image has been proposed, which generates, as the base of the AR image, the view of the front passenger on the user’s mobile-device screen, not seen by the user himself/herself but also other passengers in/outside the car with hand-held mobile devices with wireless communication capabilities.

5. A method for tour guidance in the park area has been proposed, by which the user can see augmented passenger-view images with building names on them on the user’s mobile-device screen.

The experimental results shown in the previous chapters have revealed the feasibility of the proposed system.

8.2 Suggestions for Future Works

According to the experience obtained this study, in the following we make suggestions of some interesting issues, which are worth further investigation in the future.

1. Increasing the speed of computations in feature detection and vehicle localization for realtime applications.

2. Developing the capability of detecting features of different shapes adapt the proposed system to more diversified environments.

3. Developing more applications of the proposed augmented reality techniques using the omni-camera system and the vehicle.

4. Adding the capability of detecting various features acquired by the omni-cameras on a fast-moving vehicle.

5. Including more useful information into the environment map for vehicle localization, such as stores, vendors, lakes, etc.

References

[1] B. C. Chen and W. H. Tsai, “A Study on Tour Guidance by Car Driving in Park Areas Using Augmented Reality and Omni-vision Techniques,” in Computer Vision,Graphics, and Image Processing, Aug 2012.

[2] Gandhi and M. M. Trivedi, “Motion analysis for event detection and tracking with a mobile omni-directional camera,” ACM Multimedia Systems Journal, Special Issue on Video Surveillance, vol. 10, no. 2, pp. 131–143, 2004.

[3] P. H. Yuan, K. F. Yang, and W. H. Tsai, “A Study on Monitoring of Nearby Objects around a Video Surveillance Car with a Pair of Two-camera Omni-directional Imaging Devices, ” Proceedings of 2010 Internaitonal Computer Symposium (ICS), National Chiao Tung University, Hsinchu, Taiwan, pp. 325-330, Dec. 2010.

[4] S. W. Jeng and W. H. Tsai, “Using pano-mapping tables for unwarping of omni-images into panoramic and perspective-view images,” Journal of IET Image Processing, Vol. 1, No. 2, pp. 149-155, June 2007.

[5] Y. T. Kuo and W. H. Tsai, "A new 3D imaging system using a portable two-camera omni-imaging device for construction and browsing of human-reachable environments," Proceedings of 2011 International Symposium on Visual Computing, pp. 484-495, Las Vegas, Nevada, USA.

[6] M. Betke and L. Gurvits, “Mobile robot localization using landmarks,” IEEE Transactions on Robotics and Automation, Vol. 13, No.2, pp. 251-263, April 1997.

[7] C. T. Ho and L. H. Chen, “A high-speed algorithm for elliptical object detection,” IEEE Transactions on Image Processing, Vol. 5, No. 3, pp. 547-550,

March 1996.

[8] C. J. Wu, “New Localization and Image Adjustment Techniques Using Omni-Cameras for Autonomous Vehicle Applications,” Ph. D. Dissertation, Institute of Computer Science and Engineering, National Chiao Tung University, Hsinchu, Taiwan, Republic of China, July 2009.

[9] T. Grosch, “PanoAR: Interactive Augmentation of Omni—Directional Images with Consistent Lighting,” Proc. Computer Vision Computer Graphics Collaboration Techniques and Applications (Mirage '05), University of Koblenz-Landau, Germany, pp. 25-34, 2005.

[10] Lee, J.W., You, S., Neumann, U, “Tracking with Omni-Directional Vision for Outdoor AR Systems,” Proceedings of IEEE ACM Int’l Symposium on Mixed and Augmented Reality (ISMAR 2002), Darrnstadtt, Gkrmany, October 2002.

[11] G. Reitmayr and T.W. Drummond, “Going out: Robust model based tracking for outdoor augmented reality,” Proc. IEEE Int'l Symp. Mixed and Augmented Reality (ISMAR), Santa Barbara, California, USA, pp. 109–118, 2006.

[12] M. Tonnis, C. Sandor, G. Klinker, C. Lange, and H. Bubb. “Experimen-tal evaluation of an augmented reality visualization for directing a car driver’s attention.” Proc. of IEEE and ACM International Symposium on Mixed and Augmented Reality, pp. 56–59, Vienna, Austria, Oct. 2005.

[13] B. D. Lucas and T. Kanade, “An iterative image registration technique with an application to stereo vision,” Proceedings of 7th International Joint Conference on Artificial Intelligence, Vancouver, Canada, pp. 674–679, 1981.

[14] T. Mashita, Y. Iwai, and M. Yachida, “Calibration method for misaligned catadioptric camera,” IEICE Transactions on Information & Systems, vol. E89-D, no. 7, pp. 1984-1993, July 2006.

[15] H. Ukida, N. Yamato, Y. Tanimoto, T. Sano, and H. Yamamoto,

“Omni-directional 3D Measurement by Hyperbolic Mirror Cameras and Pattern Projection,” Proceedings of 2008 IEEE Conference on Instrumentation &

Measurement Technology, Victoria, BC, Canada, May 12-15, 2008, pp. 365-370.