摘要
数字校园是在传统校园的基础上，利用先进的信息化手段和工具，将现实校园中的各项资源数字化，形成的一个数字空间。其中的虚拟校园漫游系统，则是在计算机环境中，利用虚拟现实技术再现校园的景观，来访者足不出产就可以进入到虚拟的校园之中访问漫游，尽情领略校园的文化风情。
虚拟现实技术，是用现代计算机技术创建的一个酷似客观环境又超越客观时空、既能沉浸又能驾驭的和谐人机环境。传统虚拟现实技术是基于计算机图形学的理论来建模，随着虚拟现实技术的发展，对用户的沉浸感和三维图形的视觉真实感要求越来越高，在复杂真实场景的建模表达与绘制速度和真实感上，传统计算机图形学逐渐显出其局限。
基于图像的绘制是近来兴起的一种新的虚拟现实实现方法，它以计算机视觉和图像处理等学科的理论和方法为基础，试图利用真实的图像来代替传统的几何建模，同时又用图像空间的变换操作来代替传统的绘制过程。由于基于图像的绘制技术在构造真实感场景和绘制速度上，与传统计算机图形学相比较有很大的独特优势，近年来逐渐成为图形学发展的一个热点，尤其是在虚拟环境的构造上有着很好的应用前景。
本文围绕着“基于图像绘制的虚拟校园漫游系统”这个中心，对图像拼接、全景图生成、视图插值和视图形态变换等基于图像的绘制技术的主要方法，进行了深入讨论，以比较、分析和设计其中算法为重点逐步展开，旨在改进现有的一些主要算法，并通过实验加以了验证。
通过研究现有全景图素材拼接中，各主要图像匹配对准算法的运算量和缺陷之后，本文设计了一个改进的图像拼接算法，通过理论分析与实践验证，该算法有如下优点：
(1) 能有效地消除彩色图像素材中光照条件与色差的影响，避免基于像素比较法中因色差产生的误匹配；
(2) 有较小的运算量，约为O(n~5)～O(n~4)，而且运算量尺度较小；
(3) 有很好的适应性，在大多数情况下均能较好地避免误匹配。
f＼硕士学位论文
＼W A4ASTER’S ThESIS
在分析视图插值方法的运算中，根据计算机视觉理论建立的虚拟照相机
模型，运用现行的立体视觉匹配方法，本文分别推导出了沿笔直和弯曲道路
行走时所采样的图像，在己知立体匹配点的情况下，匹配点的深度计算和中
间视图的近似插值计算方法，并由此反推出各自的大致采样间隔距离。该方
法有如下特点：
＊）可以大量减少沿道路行走时，全景图方法所需的图像采样数量，并
且能够保证漫游中各视点的平滑连接。
c）只需要前后视点，加上预匹配得到的图中对象的深度，就可以生成
中间视图，中间视图生成快。
经过比较现有主流虚拟漫游系统的实施方法，本文选择利用圆柱全景图，
加上中间采样图像序列来构建虚拟漫游系统，并介绍了本文的基于圆柱全景
图的虚拟校园漫游系统模型、功能模块及有关实现技术。
最后，分析了本文漫游系统模型和采用的相关技术的缺陷，并提出了更进
一步的研究目标。
Because the Internet is able to deliver dynamic, graphic information, a virtual tour of the campus can be made possible to allow people to get a feel for the look of the campus without needing to visit in person.
Image Based Rendering (IBR) is a new method that has been widely used in creating a virtual reality system. It is preferred over the traditional 3D geometry modeling and rendering technology that has its limitations on rendering quality and scene complexity. IBR is based on the theory and method of computer vision and image processing, and it uses photo-realistic renderings and warping. It is appealing since natural scenes, which are very difficult to model using traditional 3D geometry modeling, can be acquired semi-automatically using cameras and other devices. Also IBR has a potential to create high-quality output images if the rendering stage can preserve the quality of the reference images.
This thesis is focused on panoramic images, view interpolation and image morphing in IBR. Many methods and algorithms have been studied and compared. An improved algorithm is introduced for stitching multiple source images to produce a single output image. This new approach overcomes some of the disadvantages of previous methods. This algorithm can effectively eliminate the impact of light and color difference on the matching of two images, and reduce the possibility of mismatch. Also, this algorithm does not have intensive computing involved. It works well for almost all pictures with common characteristics. The algorithm can be easily and efficiently implemented.
This thesis also analyzes view interpolation method, provides calculation method for depth of viewpoint and interpolation of intermediate images. This
method reduces number of original images needed and produces smooth transitions between images.
This thesis proposes using cylindrical panoramic images combined with view interpolation to create a campus navigation system. The structure of the system and flow chart of the working process are provided.
引文
[1] Heung-Yeung Shum and Sing-Bing Kang. "A Review of Image-based Rendering Techniques",IEEE/SPIE Visual Communications and Image Processing (VCIP) 2000,pages 2-13,Perth,June 2000．
[2] Paul E.Debevec,Camillo J.Taylor and Jitendra Malik.Modeling and rendering architecture from photographs: A hybrid geometry-and image-based approach.In Computer Graphics (SIGGRAPH' 96) ,pages 11-20,August 1996．
[3] S.Baker,R.Szeliski,and P.Anandan.A layered approach to stereo reconstruction.In IEEE Computer Society Conference on Computer Vision and Pattern Recognition(CVPR'98) .Santa Barbara,June 1998．
[4] Paul Rademacher and Gary Bishop.Multiple-center-of-projection images.In Computer Graphics Proceedings,Annual Conference Series,pages 199-206． Proc. SIGGRAPH'98(Orlando),July 1998． ACM SIGGRAPH.
[5] Jonathan W.Shade,Steven J.Gortler,Li-wei He and Richard Szeliski.Layer depth images.In Computer Graphics (SIGGRAPH'98) ,pages 231-242,Orlando, July 1998． ACM SIGGRAPH.
[6] Chun-Fa Chang,Gary Bishop and Anselmo Lastra.LDI tree: A hierarchical representation for image-based rendering.Computer Graphics(SIGGRAPH'99) , pages 291-298,August 1999． ACM SIGGRAPH.
[7] Shenchang Eric Chen and Lance Williams.View interpolation for image synthesis.In Computer Graphics (Proc.SIGGRAPH'93) ,pages 279-288． August 1993．
[8] Steven M.Seitz and Charles R.Dyer. View morphing,In Computer Graphics Proceedings,Annual Conference Series,pages 21-30,Proc.SIGGRAPH'96(New Orleans),August 1996．
[9] E.H.Adelson and J.Bergen. The plenoptic function and the elements of early vision.In Computational Models of Visual Processing,pages 3-20． MIT Press, Cambridge,MA,1991．
[10] Leonard McMillan and Gary Bishop.Plenoptic modeling: An image-based rendering system.Computer Graphics (SIGGRAPH'95) ,pages 39-46,August 1995． ACM SIGGRAPH.
[11] Shenchang Eric Chen.QuickTime VR-an image-based approach to virtual environment navigation. Computer Graphics (SIGGRAPH'95) , pages 29-38,
August 1995. ACM SIGGRAPH.
[12]Marc Levoy and Pat Hanrahan. Light field rendering. In Computer Graphics Proceedings, Annual Conference Series, pages 31～42, Proc. SIGGRAPH'96 (New Orleans), August 1996. ACM SIGGRAPH.
[13]Steven J. Gortler, Radek Grzeszczuk, Richard Szeliski and Michael F. Cohen. The lumigraph. In Computer Graphics Proceedings, Annual Conference Series, pages 43～54, Proc. SIGGRAPH'96 (New Orleans), August 1996. ACM SIGGRAPH.
[14]Jin-Xiang Chai, Xin Tong, Shing-Chow Chan and Heung-Yeung Shum. Plenoptic Sampling. In: SIGGRAPH'2000, pages 307～318 New Orleans, Louisiana, USA, 2000.
[15]Heung-Yeung Shum and Li-Wei He. Rendering with concentric mosaics. In Computer Graphics (SIGGRAPH'99), August 1999. Los Angles, CA, USA. pages 299-306
[16]胡金辉，胡占义．同心拼图中深度的计算．计算机学报，2001．6．
[17]Lippman A. Movie maps: an application of the optical videodisc to computer graphics. Computer Graphics(Proc. SIGGRAPH'82), pages 32～43.
[18]张茂军，钟力，孙立峰，李云浩，胡晓峰．虚拟实景空间系统 HVS 的研究与实现．计算机工程，1999．7．
[19]齐越，徐玮，李梦君，李云浩，张茂军．球面虚拟空间的自由漫游．小型微型计算机系统．2001．7．
[20]马向英，杜威，袁晓君，李华．基于图像的室内虚拟漫游系统．中国图像图形学报，2001．1．
[21]漆驰，郑国勤，孙家广．一个基于全景图的虚拟环境漫游系统．计算机工程与应用，2001．15．
[22]杨键，耿卫东，潘云鹤，李明．基于图像的虚拟景观漫游．计算机辅助设计与图形学学报．2001．3．
[23]杨键，耿卫东，潘云鹤．基于图像变形的旅游模拟技术综述．计算机应用研究．1999．6．
[24]Horry Youichi, Anyio Ken-ichi and Arai Kiyoshi. Tour into the picture: Using a spidery mesh interface to make animation from a single image. In: Proceeding of the SIGGRAPH97, Los Angles, CA, USA, 1997, pages 225～232.
[25]张涛，陈一民，薛广涛．基于 PC 和 OpenGL 的全景图像 360 度漫游技术．计算机工程与应用．2001．7．
[26]刘翔，曾芬芳，高世峰，陈洁．基于 IBR 技术的全景图生成．华东船舶工业学院学报．2001．1．
[27]周天，鲁东明，潘云鹤．基于图像的绘制技术研究与发展．计算机科学，2001．5．
[28]马颂德，张正友．计算机视觉——计算理论与算法基础．科学出版社，北京，1998．
[29]成罡，金国藩，刘海松，邬敏贤，何庆生，严瑛白．基于光学沃尔什变换特征提取的图像匹配．光学学报．2001．2
[30]高世海，戴文刚，田仲．应用分形特征的景象匹配仿真．中国图像图形学报．2000．9．
[31]熊惠霖，张天序，桑农，钟胜．基于小波多尺度表示的图像匹配研究．红外与激光工程．1999．6．
[32]Richard Szeliski and Heung-Yeung Shum. Creating full view panoramic image mosaics and texture-mapped models. In: Proceeding of the SIGGRAPH97. Los Angles, LA, USA, 1997, pages 251～258．
[33]汪成为．灵境(虚拟现实)技术的理论、实现及应用．广西科学技术出版社．1996．
[34]钟力，胡晓峰．重叠图像拼接算法．中国图像图形学报．1998．3．
[35]孙华燕，周道炳，李生良．一种序列图像的拼接方法．光学精密工程．2000．1．
[36]阮宗才，粱栋，吴福朝，韦穗．一种自动全景拼图的快速算法．计算机研究与发展．2001．5．
[37]李志刚，纪玉波，薛全．边界重叠图像的一种快速拼接算法．计算机工程．2000．5．
[38]王立峰，潘志庚，石教英．全景图的拼合及优化算法．计算机应用研究．1999．6．
[39]丁迎，洪继光．图像拼接中伪匹配的判别和消解．中国图像图形学报．1999．10．
[40]冯桂，卢健，林宗坚．图像直方图不变特征在影像匹配定位牛的应用．计算机辅助设计与图形学学报．2000．2．
[41]孙远，周刚慧，赵立初，施鹏飞．灰度图像匹配的快速算法．上海交通大学学报．2000．5．
[42]赵景秀，王刚，张静远，赵家刚，杨秀国．基于彩色信息区分度的彩色边缘检测．计算机应用．2001．8．
[43]吴立德．计算机视觉．复旦大学出版社．1993．
[44]章毓晋．图像工程(下册)——图像理解与计算机视觉．清华大学出版社．2000．
[45]Ramin Zabih and John Woodfill. Non-parametric Trans form for computing the Visual Correspondence. Courses of SIGGRAPH, 1994.
[46]Kenneth R. Castleman. Digital image processing. Englewood Cliffs, New Jersey; Prentice Hall, 1996.
[47]徐玮，张茂军，吴玲达．虚拟实景空间的颜色校正方法研究．小型微型计算机系统．2001．2．
[48]何国辉．一种图像变形映射算法的研究．中国图形图像学报．1998．5．
[49]王进，潘志庚，叶澄清．一种基于轮廓多边形的图像变形算法．计算机研究与发展．1999．11．
[50]Jin Li, Harry Shum and Ya-Qin Zhang. On the compression of image based rendering scene. IEEE International Conference on Image Processing (ICIP 2000), Vancouver, BC, September. 2000.