摘要:隨著我國(guó)低空空域?qū)γ裼玫拈_(kāi)放,無(wú)人機(jī) (Unmanned aerial vehicles, UAVs)的應(yīng)用將是一個(gè)巨大的潛在市場(chǎng). 目前,如何對(duì)輕便的無(wú)人機(jī)獲取的圖像進(jìn)行全自動(dòng)處理,是一項(xiàng)急需解決的瓶頸技術(shù). 本文將探索如何將近年來(lái)在視頻、圖像領(lǐng)域獲得巨大成功的三維重建技術(shù)應(yīng)用到無(wú)人機(jī)圖像處理領(lǐng)域, 對(duì)無(wú)人機(jī)圖像進(jìn)行全自動(dòng)的大場(chǎng)景三維重建.本文首先給出了經(jīng)典增量式三維重建方法Bundler在無(wú)人機(jī)圖像處理中存在的問(wèn)題, 然后通過(guò)分析無(wú)人機(jī)圖像的輔助信息的特點(diǎn),提出了一種基于批處理重建(Batch reconstruction)框架下的魯棒無(wú)人機(jī)圖像三維重建方法.多組無(wú)人機(jī)圖像三維重建實(shí)驗(yàn)表明: 本文提出的方法在算法魯棒性、三維重建效率與精度等方面都具有很好的結(jié)果.

Abstract:With the latest deregulation and opening-up policy of Chinese government on low altitude airspace to private sectors, the applications of unmanned aerial vehicles (UAVs) will be a huge potential market. Currently the automatic processing technology of UAV images is far behind the market demand, and has become the bottleneck of various applications. This work is meant to apply hugely successful scene reconstruction techniques in computer vision field to large scene reconstruction from UAV images. To this end, at first, specific problems of direct application of the Bundler, a popular increment reconstruction technique in computer vision are investigated. Then a batch reconstruction method from UAV images is proposed by fully taking into account various pieces of prior information which are usually available in UAV images, such as those from GPS, IMU, DSM, etc. Our method is tested with several sets of UAV images, and the experiments show that our method performs satisfactorily in terms of robustness, accuracy and scalability for UAV images.

[1]Cui Hong-Xia, Lin Zong-Jian, Sun Jie. Research on UAV remote sensing system. Bulletin of Surveying and Mapping, 2005, (5): 11-14(崔紅霞, 林宗堅(jiān), 孫杰. 無(wú)人機(jī)遙感監(jiān)測(cè)系統(tǒng)研究. 測(cè)繪通報(bào), 2005, (5): 11-14) [2]Jin Wei, Ge Hong-Li, Du Hua-Qiang, Xu Xiao-Jun. A review on unmanned aerial vehicle remote sensing and its application. Remote Sensing Information, 2009, (1): 88-92(金偉, 葛宏立, 杜華強(qiáng), 徐小軍. 無(wú)人機(jī)遙感發(fā)展與應(yīng)用概況. 遙感信息, 2009, (1): 88-92) [3]Lowe D G. Distinctive image features from scale-invariant keypoints. International Journal of Computer Vision (IJCV), 2004, 60(2): 91-110 [4]Bay H, Ess A, Tuytelaars T, Van Gool L. Speeded-up robust features (SURF). Computer Vision and Image Understanding, 2008, 110(3): 346-359 [5]Fischler M A, Bolles R C. Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. Communications of the ACM, 1981, 24(6): 381-395 [6]Rousseeuw P J, Leroy A M. Robust Regression and Outlier Detection. New York: John Wiley and Sons, 1987 [7]Nister D. An efficient solution to the five-point relative pose problem. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004, 26(6): 756-770 [8]Hartley R and Hartley, Zisserman A. Multiple View Geometry in Computer Vision (Vol. 2). Cambridge University Press, Cambridge, 2004 [9]Pollefeys M, Van Gool L, Vergauwen M, Verbiest F, Cornelis K, Tops J, Koch R. Visual modeling with a hand-held camera. International Journal of Computer Vision, 2004, 59(3): 207-232 [10]Snavely N, Seitz S M, Szeliski R. Photo tourism: exploring photo collections in 3D. ACM Transactions on Graphics (TOG), 2006, 25(3): 835-846 [11]Strecha C, Pylv?n?inen T, Fua P. Dynamic and scalable large scale image reconstruction. In: Proceedings of the 2010 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). San Francisco, CA: IEEE, 2010. 406-413 [12]Hiep V H, Keriven R, Labatut P, Pons J P. Towards high-resolution large-scale multi-view stereo. In: Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Miami, FL: IEEE, 2009. 1430-1437 [13]Furukawa Y, Ponce J. Accurate, dense, and robust multiview stereopsis. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(8): 1362-1376 [14]Gherardi R, Farenzena M, Fusiello A. Improving the efficiency of hierarchical structure-and-motion. In: Proceedings of the 2010 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). San Francisco, CA: IEEE, 2010. 1594-1600 [15]Shum H Y, Ke Q F, Zhang Z Y. Efficient bundle adjustment with virtual key frames: a hierarchical approach to multi-frame structure from motion. In: Proceedings of the 1999 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR). Fort Collins, CO: IEEE, 1999. 538-543 [16]Martinec D, Pajdla T. Robust rotation and translation estimation in multiview reconstruction. In: Proceedings of the 2007 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Minneapolis, MN: IEEE, 2007. 1-8 [17]Sturm P F, Triggs B. A factorization based algorithm for multi-image projective structure and motion. In: Proceedings of the 4th European Conference on Computer Vision (ECCV). London, UK: Springer-Verlag, 1996. 709-720 [18]Tomasi C, Kanade T. Shape and motion from image streams under orthography: a factorization method. International Journal of Computer Vision (IJCV), 1992, 9(2): 137-154 [19]Kahl F, Hartley R I. Multiple-view geometry under the L∞ norm. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 30(9): 1603-1617 [20]Sim K, Hartley R. Recovering camera motion using L∞ minimization. In: Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR). Washington, DC: IEEE, 2006. 1230-1237 [21]Li Li-Chun, Zhang Heng, Li You, Yu Qi-Feng. A 3D landform reconstruction method based on UAV image sequences. Remote Sensing Technology and Application, 2008, 23(5): 505-510(李立春, 張恒, 李由, 于起峰. 一種基于無(wú)人機(jī)序列成像的地形地貌重建方法. 遙感技術(shù)與應(yīng)用, 2008, 23(5): 505-510) [22]Qiu Zhi-Qiang. Three-Dimensional Structure Reconstruction from Aerial Images Based on Affine Approximation [Ph.D. dissertation], National University of Defense Technology, China, 2004(邱志強(qiáng). 基于仿射近似從航空?qǐng)D像重建目標(biāo)三維結(jié)構(gòu) [博士學(xué)位論文], 國(guó)防科學(xué)技術(shù)大學(xué), 中國(guó), 2004) [23]Shen Yong-Lin, Liu Jun, Wu Li-Xin, Li Fa-Shuai, Wang Zhi. Reconstruction of disaster scene from UAV images and flight-control data. Geography and Geo-Information Science, 2011, 27(6): 13-17(沈永林, 劉軍, 吳立新, 李發(fā)帥, 王植. 基于無(wú)人機(jī)影像和飛控?cái)?shù)據(jù)的災(zāi)場(chǎng)重建方法研究. 地理與地理信息科學(xué), 2011, 27(6): 13-17) [24]Irschara A, Hoppe C, Bischof H, Kluckner S. Efficient structure from motion with weak position and orientation priors. In: Proceedings of the 2011 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPR). Colorado Springs, CO: IEEE, 2011. 21-28 [25]Snavely K N. Scene Reconstruction and Visualization from Internet Photo Collections [Ph.D. dissertation], University of Washington, USA, 2008 [26]Sinha S N, Frahm J M, Pollefeys M, Genc Y. Feature tracking and matching in video using programmable graphics hardware. Machine Vision and Applications, 2011, 22(1): 207-217 [27]Wu C C, Agarwal S, Curless B, Seitz S M. Multicore bundle adjustment. In: Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Providence, RI: IEEE, 2011. 3057-3064

相關(guān)知識(shí)

基于人體姿態(tài)識(shí)別的自重健身輔助教練系統(tǒng)、方法、終端
基于腦電數(shù)據(jù)與面部表情影像的抑郁癥病癥輔助識(shí)別方法技術(shù)
規(guī)劃信息處2018年度重點(diǎn)工作落實(shí)情況
衛(wèi)健委、國(guó)務(wù)院、中華醫(yī)學(xué)會(huì)新發(fā)文未來(lái)五年人工智能輔助醫(yī)學(xué)影像
關(guān)于甲殼素片的基本信息介紹
一種輔助群體健康飲食的系統(tǒng)以及方法
基于時(shí)序巡航圖像的茶樹(shù)生長(zhǎng)監(jiān)測(cè)研究
福建省食品安全信息追溯管理辦法
基于圖注意力機(jī)制的藥物副作用頻率預(yù)測(cè)方法
機(jī)器學(xué)習(xí)在高德用戶(hù)反饋信息處理中的實(shí)踐

網(wǎng)址: 基于輔助信息的無(wú)人機(jī)圖像批處理三維重建方法 http://m.u1s5d6.cn/newsview681594.html