進階搜尋


下載電子全文  
系統識別號 U0026-0708201315290000
論文名稱(中文) 自動化鑲嵌區域偵測技術應用於航空影像拼接
論文名稱(英文) Automatic Determination of Blending Zone for Aerial Image Mosaicking
校院名稱 成功大學
系所名稱(中) 測量及空間資訊學系碩博士班
系所名稱(英) Department of Geomatics
學年度 101
學期 2
出版年 102
研究生(中文) 周漢思
研究生(英文) Han-Szu Chou
學號 p66004023
學位類別 碩士
語文別 英文
論文頁數 45頁
口試委員 指導教授-林昭宏
口試委員-曾義星
口試委員-張智安
口試委員-徐百輝
中文關鍵字 影像鑲嵌  最佳路徑判定  影像混和 
英文關鍵字 image mosaicking  optimal path determination  color blending 
學科別分類
中文摘要 影像拼接(image mosaicking)在航空正射影像鑲嵌技術中是一個基礎且重要的步驟。隨著數位影像品質的提升、數位影像處理技術的精進與無人空中飛行載具使用頻率的增加,使得高解析度低航高影像的應用逐漸擴大,也使得低航高正射影像鑲嵌的需求日益增加,此代表著需合併更多的影像去覆蓋指定區域,而影像的視差效應(parallax effect)會相當顯著,因此許多研究學者開始致力於提升自動化航空正射影像鑲嵌技術解決此問題。在影像鑲嵌的技術中,其中一個關鍵的步驟在於如何在影像重疊區域內,決定一個最佳的鑲嵌線,使得影像能無縫的拼接。近期大部分的研究,以最佳化自訂的目標函式來尋找單一像素寬的鑲嵌線,此方法雖然可降低由幾何畸變造成的錯位問題,但並沒有考慮到在鑲嵌線周圍的色彩不連續與錯位問題,而使得鑲嵌成果有時候會有明顯的人為痕跡(seam artifact)。
基於同時考慮鑲嵌線的位置與色彩混和區域的平滑問題,我們針對多頻帶色彩混和技術提出一個混和區域決定的演算法。比起只尋找單一像素寬鑲嵌線的方法,此演算法利用階層式架構,使用最短路徑演算法有效率地在重疊區找到擁有k個像素寬並通過高相似像元的混和區域,而基於已決定的混和區域,利用多頻帶色彩混和技術配合加權函式可平滑地接合在重疊區域每個像元的色彩。這個策略使我們提出的方法能無縫地合併鄰近影像區塊,並減輕由色彩不連續造成的人為鑲嵌問題與因錯位而產生的鬼影問題。比起其他相關的航空影像鑲嵌方法,在各種不同航空影像的定性分析(qualitative analysis)與定量分析(quantitative analysis)中,本文所提的方法都有較佳的表現。
英文摘要 Creating an composed image from aerial images is a fundamental but important step in orthomosaic generation. With the development of high-quality digital imaging and digital photogrammetry techniques as well as the increasing use of unmanned aerial vehicles, orthomosaics are created from large sets of low-altitude aerial images. This process means that more images need to be merged to cover a given area and that parallax effects are now more significant. As a result, many researchers have developed techniques to automatically generate orthomosaics from aerial images. One of the processes involved in this technique is determining an optimal seamline in the overlapping regions to seamlessly merge the image patches. Most previous studies solved this optimization problem by searching for a one-pixel wide seamline with a minimum-value defined objective function. This strategy significantly reduces the mismatch problem caused by geometric distortions but does not fully consider color discontinuity and mismatch problems occurring around the seamline, sometimes causing apparent seam artifacts.
By considering both seamline determination and transition smoothing, we propose a blending zone determination scheme with multi-band color blending to reduce the occurrence of unwanted artifacts. Instead of searching for a one-pixel wide seamline, a blending zone, that is, a k-pixel wide seamline, passing through high-similarity pixels in overlapping regions is efficiently determined by utilizing a hierarchal structure with the shortest path search algorithm. Multi-band blending technique is then adopted to smoothly blend the colors of pixels in the overlapping region with a weighting function based on the determined blending zone. This strategy allows for the seamless merging of neighboring image patches and reduces the possibility of obtaining seam artifacts caused by color discontinuity and ghost effects caused by mismatched pixels. Qualitative and quantitative analyses of various aerial images show that the proposed method is superior over other related aerial image mosaicking methods.
論文目次 摘要 I
Abstract II
致謝 IV
Content V
List of Figures VII
List of Tables X
Chapter 1 INTRODUCTION 1
1.1 Seamline Determination 1
1.2 Color Blending 3
Chapter 2 PREPROCESSING 5
2.1 Aerial Orthoimage 5
2.2 Color Balancing 5
Chapter 3 IMAGE MOSAICKING 8
3.1 Registration 8
3.2 Hierarchical Seam Line Determination 11
3.2.1 Cost Function 11
3.2.2 Seamline Determination 12
3.2.3 Optimal Path 16
3.3 Multi-band Color Blending 19
Chapter 4 EXPERIMENTAL RESULTS 24
4.1 Experimental data 24
4.2 Number of Bands 26
4.3 Path Length Condition for Shortest Path 27
4.4 Experimental Results and Analysis 28
4.5 Large-size Aerial Image Mosaicking 39
Chapter 5 CONCLUSIONS AND FUTURE STUDIES 41
5.1 Conclusions 41
5.2 Future Work 41
References 43
參考文獻 Adelson, E.H., Anderson, C.H., Bergen, J.R., Burt, P.J., and Ogden, J.M. (1984), “Pyramid method in image processing”, RCA Engineer, 29(6), pp. 33-41.
Allene, C., Pons, J.P., and Keriven, R. (2008), “Seamless image-based texture atlases using multi-band blending”, 19th International Conference on Pattern Recognition, pp. 1-4.
Botterill, T., Mills, S. and Green, R. (2010), “Real-time aerial image mosaicking”, 25th International Conference of Image and Vision Computing New Zealand, pp. 1-8.
Brown, M. and Lowe, D.G. (2003), “Recognising panoramas”, In Proceedings of the 9th International Conference on Computer Vision, 2, pp. 1218-1225.
Brown, M. and Lowe, D.G. (2007), “Automatic Panoramic Image Stitching using Invariant Features”, International Journal of Computer Vision, 74(1), pp. 59-73.
Burt, P.J. and Adelson, E.H. (1983), “A multiresolution spline with application to image mosaics”, ACM Transactions on Graphics, 2(4), pp. 217-236.
Chon, J., Kim, H. and Lin, C.S. (2010), “Seam-line determination for image mosaicking: A technique minimizing the maximum local mismatch and the global cost”, ISPRS Journal of Photogrammetry and Remote Sensing, 65(1), pp. 86-92.
Fernandez, E., Garfinkle, R. and Arbiol, R. (1998), “Mosaicking of aerial photographic maps via seams defined by bottleneck shortest paths”, Operations Research, 46(3), pp. 293-304.
Kerschner, M. (2001), “Seamline detection in colour orthoimage mosaicking by use of twin snakes”, ISPRS Journal of Photogrammetry and Remote Sensing, 56(1), pp. 53-64.
Lin, C.H., Tsai, P.H., Lai, K.K. and Chen, J.Y. (2013), “Cloud Removal from Multitemporal Satellite Images Using Information Cloning”, IEEE Transactions on Geoscience and Remote Sensing, 51(1), pp. 232-241.
Milgram, D.L. (1977), “Adaptive techniques for photomosaicking”, IEEE Transactions on Computers, C-26(11), pp. 1175-1180.
Mills, A. and Dudek, G. (2009), “Image stitching with dynamic elements”, Image and Vision Computing, 27(10), pp. 1593-1602.
Mills, S. and McLeod, P. (2013), “Global seamline networks for orthomosaic generation via local search”, ISPRS Journal of Photogrammetry and Remote Sensing, 75, pp. 101-111.
Pan, J., Wang, M., Li, D. and Li, J. (2009), “Automatic generation of seamline network using area Voronoi diagrams with overlap”, IEEE Transactions on Geoscience and Remote Sensing, 47(6), pp. 1737-1744.
Pan, J. and Wang, M. (2011), “A seam-line optimized method based on difference image and gradient image”, IEEE 19th International Conference on Geoinformatics, pp. 1-6.
Peleg, S. (1981), “Elimination of seams from photomosaics”, Computer Graphics and Image Processing, 16(1), pp. 90-94.
Soille, P. (2006), “Morphological image compositing”, IEEE Transactions on Pattern Analysis and Machine Intelligence, 28(5), pp. 673-683.
Szeliski, R. (2006), “Image alignment and stitching: A tutorial”, Foundations and Trends in Computer Graphics and Vision, 2(1), pp. 1-104.
Uyttendaele, M., Eden, A., and Szeliski, R. (2001), “Eliminating ghosting and exposure artifacts in image mosaics”, IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. II:509-II:516.
Wan, Y., Wang, D., Xiao, J., Wang, X., Yu, Y. and Xu, J. (2012), “Tracking of vector roads for the determination of seams in aerial image mosaics”, IEEE Geoscience and Remote Sensing Letters, 9(3), pp. 328-332.
Wan, Y., Wang D., Xiao, J., Lai, X., and Xu, J. (2013), “Automatic determination of seamlines for aerial image mosaicking based on vector roads alone”, ISPRS Journal of Photogrammetry and Remote Sensing, 76, pp. 1-10.
Wang, W. and Ng, M.K. (2012), “A variational method for multiple-image blending”, IEEE Transactions on Image Processing, 21(4), pp. 1809-1822.
Yu, L., Holden, E.J., Dentith, M.C. and Zhang, H. (2012), “Towards the automatic selection of optimal seam line locations when merging optical remote-sensing images”, International Journal of Remote Sensing, 33(4), pp. 1000-1014.
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2016-08-22起公開。
  • 同意授權校外瀏覽/列印電子全文服務,於2016-08-22起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw