為了擴大遙感影像的覆蓋範圍，影像匹配受到了廣泛關注。 找出兩幅圖像之間的相關性並估計適用於兩幅圖像的模型是圖像匹配的基本目標。 在本論文中首先提出了一種基於幾何直方圖（GH-based）的快速剔除錯誤匹配策略，從初始對應集中構造一個縮減對應集。接下來，本論文提出了一種用於遙感圖像匹配的有效合作式隨機樣本共識（COOSAC）方法。COOSAC 由一個名為初始 RANSAC，和一個小型 RANSAC 組成。在小型 RANSAC 中，提出了一種基於面積約束的採樣策略，一直迭代直到達到指定的置信度，然後初始 RANSAC 利用求得的估計模型來計算初始對應集的內點比率。 COOSAC 重複初始 RANSAC，和一個小型 RANSAC 之間的合作，直到達到指定的置信度，並回傳最終的結果模型。為方便起見，我們的圖像匹配方法稱為 GH-COOSAC 方法。基於多個測試數據集，全面的實驗結果表明，與最先進的圖像匹配方法相比，所提出的 GH-COOSAC 方法具有更低的計算成本和更高的匹配精度優勢。

To extend the areas that remote sensing images can cover, image matching has received extensive attention. Finding the correlation between two images and estimating the model that applies to both images is the basic goal of image matching. In this thesis, a fast geometry histogram-based (GH-based) mismatch removal strategy to construct a reduced correspondence set from the initial correspondence set is first proposed. Next, an effective cooperative random sample consensus (COOSAC) method for remote sensing image matching is proposed. COOSAC consists of initial RANSAC, and tiny RANSAC. In tiny RANSAC, an area constraint-based sampling strategy is proposed to estimate the model solution that iterates until the specified confidence level is reached, and then initial RANSAC utilizes the estimated model solution to calculate the inlier rate of the initial correspondence set. COOSAC repeats the above cooperation between initial RANSAC and tiny RANSAC until the specified confidence level is reached, and report the resultant model solution. For convenience, the proposed image matching method is called the GH-COOSAC method. Based on several testing datasets, thorough experimental results demonstrate that the proposed GH-COOSAC method achieves lower computational cost and higher matching accuracy benefits when compared with the state-of-the-art image matching methods.

[1] O. Chum and J. Matas, “Optimal randomized ransac,” IEEE Trans.
Pattern Anal. Mach. Intell., vol. 30, no. 8, pp. 1472–1482, 2008.
[2] M. A. Fischler and R. C. Bolles, “Random sample consensus: A
paradigm for model fitting with applications to image analysis and
automated cartography,” Commun. ACM, vol. 24, p. 381–395, jun
1981.
[3] D. Capel, “An effective bail-out test for ransac consensus scoring,”
Proc. British Machine Vision Conf, pp. 629–638, 2005.
[4] O. Chum and J. Matas, “Matching with prosac - progressive sam-
ple consensus,” in 2005 IEEE Computer Society Conference on Com-
puter Vision and Pattern Recognition (CVPR’05), vol. 1, pp. 220–226
vol. 1, 2005.
[5] O. Chum and J. Matas, “Optimal randomized ransac,” IEEE Trans.
Pattern Anal. Mach. Intell., vol. 30, no. 8, pp. 1472–1482, 2008.
[6] R. Raguram, O. Chum, M. Pollefeys, J. Matas, and J.-M. Frahm,
“Usac: A universal framework for random sample consensus,” IEEE
Trans. Pattern Anal. Mach. Intell., vol. 35, no. 8, pp. 2022–2038,
2013.
[7] J. Ma, J. Zhao, J. Tian, A. L. Yuille, and Z. Tu, “Robust point match-
ing via vector field consensus,” IEEE Trans. Image Process., vol. 23,
no. 4, pp. 1706–1721, 2014.
[8] M. Ester, H.-P. Kriegel, J. Sander, and X. Xu, “A density-based algo-
rithm for discovering clusters in large spatial databases with noise,”
in Proceedings of the Second International Conference on Knowledge
Discovery and Data Mining, KDD’96, p. 226–231, 1996.
[9] X. Jiang, J. Ma, J. Jiang, and X. Guo, “Robust feature matching using
spatial clustering with heavy outliers,” IEEE Trans. Image Process.,
vol. 29, pp. 736–746, 2020.
[10] J. Ma, J. Zhao, J. Jiang, H. Zhou, and X. Guo, “Locality preserving
matching,” Int. J. Comput. Vision, vol. 127, p. 512–531, may 2019.
[11] G. Wang, X. Sun, Y. Shang, Z. Wang, Z. Shi, and Q. Yu, “Two-view
geometry estimation using ransac with locality preserving constraint,”
IEEE Access, vol. 8, pp. 7267–7279, 2020.
[12] W.-Y. Lin, F. Wang, M.-M. Cheng, S.-K. Yeung, P. H. Torr, M. N. Do,
and J. Lu, “Code: Coherence based decision boundaries for feature
correspondence,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 40,
no. 1, pp. 34–47, 2018.
[13] J. Ma, J. Jiang, H. Zhou, J. Zhao, and X. Guo, “Guided locality
preserving feature matching for remote sensing image registration,”
IEEE Trans. Geosci. Remote Sens., vol. 56, no. 8, pp. 4435–4447,
2018.
[14] C. Yang, M. Zhang, Z. Zhang, L. Wei, R. Chen, and H. Zhou, “Non-
rigid point set registration via global and local constraints,” Multime-
dia Tools and Applications, vol. 77, pp. 1–19, 12 2018.
[15] Y. Liu, Y. Li, L. Dai, T. Lai, C. Yang, L. Wei, and R. Chen, “Motion
consistency-based correspondence growing for remote sensing image
matching,” IEEE Geosci. Remote Sens. Lett., vol. 19, pp. 1–5, 2022.
[16] G. Wang and Y. Chen, “Robust feature matching using guided local
outlier factor,” Pattern Recognit., vol. 117, p. 107986, 2021.
[17] B. N. Datta, Numerical linear algebra and applications. USA:
Brooks/Cole, 1995.