隨著衛星科技的高速演進，衛星影像的種類與特性也是千變萬化，然而這雖然代表了衛星影像具有廣泛的實用性，但也導致了不同資料格式衛星影像間對位的困難。本研究透過自適應性對比度強化影像特徵對位演算法提高影像對位精度，並使其成果更為穩定不失真，期待促使衛星影像間應用更為廣泛。
本研究演算法分為三部分，分別為影像前處理、影像匹配與影像對位。影像前處理主要內容為衛星影像對比度的調整，經由強化影像灰階平面的特徵，並且透過不同程度對比度調整，增加其特徵的種類與變化，同時又經由對比度正規化，平衡不同衛星影像間灰階變化模式，促使其匹配更為穩定，衛星影像前處理可使影像匹配運算精度更高且更穩定。影像匹配演算法透過特徵偵測與特徵匹配尋找影像間相同位置，同時透過隨機抽樣一致演算法排除錯誤資料，獲取高精度匹配成果。影像對位憑藉高精度匹配成果，計算影像間座標轉換矩陣，可轉換影像間像元資料來產製大面域且細節更為細緻的衛星影像，提高衛星影像可用性。
本研究透過不同相似度匹配測試影像驗證演算法精度與穩定性，同時也經由大量資料分析如何達到演算法最佳化，獲得對位最佳成果，提高演算法的可用性。

Remote sensing researches using optical satellite images have been addressed for years. Image matching is one of the key techniques applied to remote sensing applications. For instance, image fusion and orthogonal image generation require the step of image matching in data processing. This study focuses on image matching for optical satellite images. The goal is to improve matching results in terms of algorithm robustness and the number of matched feature pairs. The matching algorithm proposed in this study is based on Speed Up Robust Feature (SURF). The feature points are detected and encoded by SURF algorithm. Furthermore, the matching feature points in input images are matched using their encoded descriptions. Algorithm efficiency and robustness are the advantages of SURF algorithm. However, detailed features in near-homogenous regions of the satellite images may not be successfully detected because of the inefficient spatial resolution of satellite images. In this study, a local image enhancement is performed prior to the feature detection. Image contrast adjustment of different degrees produces an image ranking in the order of image contrast. This image ranking is further processed by SURF algorithm. The image rank will have the achievement about the sum of the contrast image rank’s matching results. From quantitative and qualitative analyses, image matching with the proposed local image enhancement improve the matching results, in terms of matching accuracy and number of matched pairs.

摘要 I
英文延伸摘要 II
誌謝 VIII
目錄 X
圖目錄 XIII
表目錄 XVIII
第一章 緒論 1
1.1研究動機 1
1.2研究目的 2
1.3研究方法與流程 3
1.4研究貢獻 5
1.5論文架構 6
第二章 相關研究 7
2.1衛星影像特性 7
2.2相關影像匹配演算法 8
2.3影像匹配應用 14
第三章 方法 18
3.1演算法流程 18
3.2影像資料前處理 19
3.2.1影像對比度增強演算法 21
3.2.2影像前處理演算法模型 33
3.3影像特徵匹配演算法 35
3.3.1影像特徵偵測 35
3.3.2影像特徵描述 41
3.3.3影像特徵匹配 44
3.3.4影像匹配成果校正 46
3.4影像對位 49
3.4.1影像像座標轉換矩陣 50
3.4.2反向雙線性內插法 53
3.4.3影像融合 56
第四章 實作成果 59
4.1測試資料收集與產製 60
4.2自製影像對位成果 64
4.3影像波段對位成果 73
4.4相同感測器影像對位成果 81
4.5不同感測器同類型波段影像對位成果 89
4.6不同感測器影像對位成果 98
4.7成果討論與演算法最佳化 107
第五章 結論與演算法限制 110
參考文獻 112

Achour, K., and Benkhelif, M., (2001). A new approach to 3D reconstruction without camer calibration. Pattern Recognition Society by Elsevier Science Ltd, 34:2467-2476.
Ahengyou, Z., Rachid, D., Olovier, F., and Quang, T. L., (1994). A robust technique for matching two uncalibrated images through the recovery of the unknown epipolar geometry. Artificial Intelligence, 78:87-119.
Alam, M. S., Bognar, J. G., Hardie, R. C., and Yasuda, B. J., (2000). Infrared image registration and high-resolution reconstruction using multiple translationally shifted aliased video frames. IEEE Trans. Instrumentation and Measurement, 49:915-923.
Stark, J. A., (2000). Adaptive image contrast enhancement using generalizations of histogram equalization. IEEE Transactions on Image Processing, 9:889-896.
Bay, H., Tuytelaars, T., and van, G. L. (2006). SURF: Speeded up robust features. In ECCV, 404-417.
Bay, H., Tuytelaars, T., and van, G. L. (2008). Speeded-Up Robust Features (SURF), Computer Vision and Image Understanding, 110:346-359.
Bin, F., Chunlei, H., Chunhong, P., and Qingqun, K. (2013). Registration of Optical and SAR Satellite Images by Exploring the Spatial Relationship of the Improved SIFT. IEEE Geosci. Remote Sens. Lett., 10:657–661.
Björkman, M., Bergström, N., and Kragic, D., (2014). Detecting, segmenting and tracking unknown objects using multi-label MRF inference. Computer Vision Image Understand. 118:111–127.
Collignon, A., Maes, F., Delaere, D., Vandermeulen, D., Suetens, P., and Marchal, G. (1995). Automated multi-modality image registration based on information theory. In Information processing in medical imaging, 14:263-274.
Elad, M., and Hel-Or, Y., (2001). A fast super-resolution reconstruction algorithm for pure translational motion and common space-invariant blur. IEEE Trans. Image Processing, 10: 1187-1193.
Rafael, C. G., and Richard, E. W., (2008). Digital Image Processing, 3rd Edition, Prentice Hall, 467-477.
Gruen, A., and Akca, D. (2005). Least squares 3D surface and curve matching. ISPRS Journal of Photogrammetry and Remote Sensing, 59:151-174.
Heiko, H., (2008). Stereo processing by semiglobal matching and mutual information. IEEE Transaction on pattern analysis and machine intelligence, 30:328-341
Hongjian, S., and Ward, R., (2002). Canny edge based image expansion. IEEE International Symposium on Circuits and Systems, 1: 785-788.
Irani, M., and Peleg, S., (1991). Improving resolution by image registration. CVGIP: Graphical Models and Image Processing, 53: 231-239.
Irani, M., and Peleg, S., (1993). Motion analysis for image enhancement: Resolution, occlusion, and transparency. Journal of Visual Communication and Image Representation, 4: 324-335.
Ke, Y., and Sukthankar, R. (2004). PCA-SIFT: A more distinctive representation for local image descriptors. In Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), 2:506-513.
Kenji, T., Takafumi, A., Yoshifumi, S., Tastsuo, H., and Koji, K., (2003). High-Accuracy Subpixel Image Registration Based on Phase-Only Correlation. IEICE Trans. Fundamentals, 86:1925-1934.
Kupfer, B., Netanyahu, N. S., and Shimshoni, I. (2015). An Efficient SIFT-Based Mode-Seeking Algorithm for Sub-Pixel Registration of Remotely Sensed Images. IEEE Geosci. Remote Sens. Lett., 12:379–383.
Martin, A. F., and Robert, C. B., (1981). Random Sample Consensus: A Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography. Comm. of the ACM. 24:381–395.
Mikolajczyk, K., and Schmid, C., (2005). A performance evaluation of local descriptors. IEEE Trans. Pattern Anal. Mach. Intell., 27:1615–1630.
Misganu, D., and Andreas, K., (2010). Sub-pixel precision image matching for displacement measurement of mass movements using normalized cross-correlation. ISPRS TC VII symposium, 38:182-186.
Olson, C. F. (2002). Maximum-likelihood image matching. IEEE Transactions on Pattern Analysis and Machine Intelligence, 24:853–857.
Park, S. C., Park, M. K., and Kang, M. G., (2003). Super-resolution image reconstruction: a technical overview. IEEE Signal Processing Magazine, 20:21-36.
Rahul, S., Robert, G. S., and Matthew D. M., (2001). Smarter Presentations: Exploiting homography in camera-projector systems. Proceedings of International Conference on Computer Vision 2001, 1:247-253
Sedaghat, A., Mokhtarzade, M., and Ebadi, H. (2011). Uniform robust scale-invariant feature matching for optical remote sensing images. IEEE Trans. Geosci. Remote Sens., 49:4516-4527.
Sun, Y., Zhao, L., Huang, S., Yan, L., and Dissanayake, G., (2014). L-2-SIFT: SIFT feature extraction and matching for large images in large-scale aerial photogrammetry. ISPRS J. Photogramm. Remote Sens., 91:1-16.
Teke, M., Vural, M. F., Temizel, A., and Yardımcı, Y. (2011). High-resolution multispectral satellite image matching using scale invariant feature transform and speeded up robust features, J. Appl. Remote Sens., 5:1-9.
Tsai, D. M., and Lin, C. T. (2003). Fast normalized cross correlation for defect detection. Pattern Recognition Letters, 24:2625-2631.
Tsai, R. Y., and Huang, T. S., (1984). Multiple-frame image restoration and registration. in Computer Vision and Image Processing. Greenwich, 1:317-339.
Wang, L., Niu, Z., Wu, C., Xie, R., and Huang, H. (2012). A robust multisource image automatic registration system based on the SIFT descriptor. Int. J. Remote Sens., 33:3850-3869.
Wu, C. SiftGPU: A GPU Implementation of Scale Invariant Feature Transform (SIFT). Available online: http://cs.unc.edu/ ccwu/siftgpu.
Xin, L., and Orchard, M. T., (2001). New edge-directed interpolation. IEEE Trans. Image Processing, 10:1521-1527.
Ye, Y., and Shan, J. (2014). A local descriptor based registration method for multispectral remote sensing images with non-linear intensity differences. ISPRS J. Photogramm. Remote Sens., 90:83–95.
Yi, Z., Zhiguo, C., and Yang, X. (2008). Multi-spectral remote image registration based on SIFT. Electron. Lett., 44:107–108.
Yu, G., and Morel, J.-M., (2011). ASIFT: An Algorithm for Fully Affine Invariant Comparison, Image Processing On Line, 1:11-38.
Zhang, L., and Gruen, A. (2006). Multi-image matching for DSM generation from IKONOS imagery. ISPRS Journal of Photogrammetry and Remote Sensing, 60:195-211.
Zhao, L., Huang, S., Yan, L., Wang, J.J., Hu, G., and Dissanayake, G. (2010). Large-scale monocular SLAM by local bundle adjustment and map joining. In: Proc. 11th International Conference on Control Automation Robotics ＆ Vision (ICARCV), 431–436.
Zhili, S., Shuigeng, Z., and Jihong, G. (2014). A Novel Image Registration Algorithm for Remote Sensing Under Affine Transformation. IEEE Trans. Geosci. Remote Sens., 52:4895–4912.
吳文心. (2011). 基於邊緣適應性非均勻內插法之超解析度影像重建技術，國立臺灣科技大學電機工程系碩士學位論文。
林炫佑. (2011). 基於局部不變性特徵之遙測影像匹配與套合研究，國立高雄應用科技大學土木工程與防災科技研究所碩士論文。
陳良健. (1997). 遙測影像之整合於地表變遷偵測之應用(III): 子計畫一: 遙測影像之整合中光學探測器影像之幾何處理，國科會專題研究計畫報告。
陳俊愷. (2011). 影像特徵點萃取與匹配應用於福衛二號影像幾何糾正，國立臺灣師範大學地理學系博士論文。
陳啟城. (2013). 影像特徵選取與描述方法於影像匹配應用之研究，國立高雄應用科技大學土木工程與防災科技研究所碩士論文。