道路交通事故的發生主要在於車輛行進中，駕駛人沒察覺到車輛周圍的物體而造成的碰撞意外。沒察覺的主要原因在於車體構造和後照鏡角度的限制所造成的。為了協助駕駛人注意車體周邊狀況，提升駕駛安全，減少人員傷亡和車體損傷，我們在本研究中提出一個廣域全周俯瞰監視與障礙物偵測系統。整個系統包含兩大部份：一是廣域全周俯瞰監視用於輔助駕駛監視車輛周遭的狀況，二是主動偵測車輛附近的障礙物並提醒駕駛人注意。
廣域全周俯瞰監視系統在車輛的四周架設廣角相機以拍攝周遭影像，利用離線程序計算相機內部參數、暗角參數、及扭曲參數。接著利用大型校正板，根據特徵點對應求得四張俯瞰影像之間的相對關係，將俯瞰影像整合成一張俯視車輛周遭的全周俯瞰影像，再以影像中心為圓心，將內圈和外圈兩個區域以不同的函數收縮影像，增加影像中車輛周圍的可視範圍，完成廣域全周俯瞰監視影像。最後將各參數建立一張查找表；在線上處理階段，以四部相機中取得亮度平均值最高的影像做為參考，調整其它影像的亮度，再根據查找表的資訊，內差產生即時的廣域全周俯瞰監視影像。
在障礙物偵測中，我們將四部相機的影像平均分割區塊，保留擁有強烈特徵的區塊，接著合併具有強烈特徵且相鄰的區塊，稱之為障礙物候選區塊。再利用判斷平面立面的程序，確認候選區塊中的物體為障礙物而非路面標線等平面物；最後利用查找表資訊，將障礙物的位置標示於廣域全周俯瞰影像以警示駕駛。
本論文增加了俯瞰監視下的可視範圍，輔助駕駛人在慢速或擁擠的道路上行車時能注意車輛周邊的狀況，預防交通意外的發生。此外車輛周圍增加的視野同時也提供額外車輛周邊資訊，經由本系統進一步分析，進行障礙物偵測，即使盲點區域發生危險，也能輔助駕駛人避免意外發生。

A lot of traffic accidents are caused by driver's incomplete understanding of the whole vehicle surroundings. To reduce the accidents caused by collision with surrounding obstacles, we mount four wide-angle cameras at the front, rear, and both sides of the vehicle to capture consecutive images; then we present a real-time wide-scope top-view monitor and obstacle detection system for driving and parking assistance.
In offline steps of wide-scope top-view monitor system, we first calibrate camera intrinsic parameters, distortion of lens, and vignetting effects of four wide-angle cameras. Then we calibrate the geometric relationships (extrinsic parameters) of four cameras using a big calibration board. Third, we calculate the feathering weights of pixels on overlapped image areas to produce a seamless surrounding top-view image. Fourth, from the image center, we utilize different function for different radius distance to shrink the seamless surrounding top-view to produce a wide-scope top-view image. At last, we build look-up tables for the mapping between the captured images and the surrounding synthesized image to speed up the processing. In online procedure, the proposed system interpolates and generates the surrounding synthesized image by those look-up tables directly.
In obstacle detection system, we have four camera images evenly divided blocks, retained the block with strong feature, and then merge the blocks with strong feature and being adjacent, known as the obstacle candidate blocks. Then utilize the plane-elevation process, confirm the candidate block objects as obstacles rather than pavement markings and other flat objects; and finally using the lookup table information, mark obstacle position on the entire wide-scope top-view image to warn drivers.

[1] Carloni, R., V. Lippiello, M. D'Auria, M. Fumagalli, A. Y. Mersha, S. Stramigioli, and B. Siciliano, "Robot vision: Obstacle-avoidance techniques for unmanned aerial vehicles," IEEE Robotics and Automation Magazine, vol.20, no.4, pp.22-31, 2013.
[2] De Croon, G. C. H. E., E. De Weerdt, C. De Wagter, and B. D. W. Remes, "The appearance variation cue for obstacle avoidance," in Proc. 2010 IEEE Int. Conf. on Robotics and Biomimetics, Tianjin, China, Dec.14-18, 2010, pp.1606-1611.
[3] Devernay, F. and O. Faugeras, "Straight lines have to be straight," Machine Vision and Applications, vol.13, no.1, pp.14-24, 2001.
[4] Fardi, B., T. John, and G. Wanielik, "Non-rigid-motion recognition using a moving mono camera," in Proc. IEEE Intelligent Vehicles Symposium, Xi'an, China, Jun.3-5, 2009, pp.221-226.
[5] Fujitsu, 360-Degree Wrap-around Video Imaging Technology, in http://www.fujitsu.com/us/news/pr/fma_20101019-02.html
[6] Hoiem, D., A. A. Efros, and M. Hebert, "Putting objects in perspective," Int. Journal of Computer Vision, vol.80, no.1, pp.3-15, 2008.
[7] Honda, Multi-view Camera System, in http://world.honda.com/news/2008/4080918Multi-View-Camera-System/
[8] Kang, S. B. and R. S. Weiss, "Can we calibrate a camera using an image of a flat, textureless lambertian surface?," in Proc. 6th European Conf. on Computer Vision, Dublin, Ireland, Jun.26-Jul.1, 2000, pp.640-653.
[9] Kato, T., Y. Ninomiya, and I. Masaki, "An obstacle detection method by fusion of radar and motion stereo," IEEE Trans. on Intelligent Transportation Systems, vol.3, no.3, pp.182-187, 2002.
[10] Klappstein, J., F. Stein, and U. Franke, "Monocular motion detection using spatial constraints in a unified manner," in Proc. 2006 IEEE Intelligent Vehicles Symposium, Meguro-Ku, Tokyo, Jun.13-15, 2006, pp.261-267.
[11] Kulchandani, J. S. and K. J. Dangarwala, "Moving object detection: Review of recent research trends," in Proc. Int. Conf. on Pervasive Computing, Pune, India, Jan.8-10, 2015, pp.1-5.
[12] Kundu, A., C. V. Jawahar, and K. M. Krishna, "Realtime moving object detection from a freely moving monocular camera," in Proc. IEEE Int. Conf. on Robotics and Biomimetics, Tianjin, China, Dec.14-18, 2010, pp.1635-1640.
[13] Lalonde, J., R. Laganière, and L. Martel, "Single-view obstacle detection for smart back-up camera systems," in Proc. IEEE Conf. on Computer Vision and Pattern Recognition Workshops, Providence, RI, Jun.16-21, 2012, pp.1-8.
[14] Lucas, B. D. and T. Kanade, "An iterative image registration technique with an application to stereo vision," in Proc. 7th Int. Joint Conf. on Artificial Intelligence, Vancouver, BC, Aug.24-28, 1981, pp.674-679.
[15] Marquardt, D., "An algorithm for least-squares estimation of nonlinear Parameters," SIAM Journal on Applied Mathematics, vol.11, no.2, pp.431-441, 1963.
[16] Naito, T., T. Ito, and Y. Kaneda, "The obstacle detection method using optical flow estimation at the edge image," in Proc. IEEE Intelligent Vehicles Symposium, Istanbul, Turkey, Jun.13-15, 2007, pp.817-822.
[17] Nissan, Around View Monitor, in http://www.nissan-global.com/EN/TECHNOLOGY/OVERVIEW/avm.html
[18] Oniga, F. and S. Nedevschi, "Processing dense stereo data using elevation maps: Road surface, traffic isle, and obstacle detection," IEEE Trans. on Vehicular Technology, vol.59, no.3, pp.1172-1182, 2010.
[19] Saxena, A., S. H. Chung, and A. Y. Ng, "3-D depth reconstruction from a single still image," Int. Journal of Computer Vision, vol.76, no.1, pp.53-69, 2008.
[20] Sung, K., J. Lee, J. An, and E. Chang, "Development of image synthesis algorithm with multi-camera," in Proc. IEEE 75th Vehicular Technology Conf., Yokohama, Japan, May.6-Jun.9, 2012, pp.1-5.
[21] Yamaguchi, K., T. Kato, and Y. Ninomiya, "Vehicle ego-motion estimation and moving object detection using a monocular camera," in Proc. 18th Int. Conf. on Pattern Recognition, Hong Kong, Aug.20-24, 2006, pp.610-613.
[22] Zhang, Z., "A flexible new technique for camera calibration," IEEE Trans. on Pattern Analysis and Machine Intelligence, vol.22, no.11, pp.1330-1334, 2000.
[23] Zheng, Y., S. Lin, C. Kambhamettu, J. Yu, and S. B. Kang, "Single-image vignetting correction," IEEE Trans. on Pattern Analysis and Machine Intelligence, vol.31, pp.2243-2255, 2009.
[24] 周達華, 廣域全周俯瞰監視與影像式倒車導引, 碩士論文, 資訊工程學系, 國立中央大學, 中壢, 2010.
[25] 楊善雯, 亮度一致的全周俯瞰監視與障礙物偵測, 碩士論文, 資訊工程學系, 國立中央大學, 中壢, 2013.