小基高比条件下高精度影像匹配技术研究

发布时间：2019-05-03 20:51

【摘要】：小基高比摄影测量可有效解决传统摄影测量在复杂区域的地形测绘难题，但却要求极其精确的视差量测精度。高精度影像匹配是实现小基高比条件下高精度定位的关键。本文在深入分析小基高比摄影测量理论的基础上，探讨了基高比、影像匹配精度和交会精度之间的联系，深入研究了小基高比条件下影像预匹配、高精度影像匹配等方面的内容，取得了一定的进展。论文完成的主要工作和创新点有： 1.论述了本文的研究意义和研究背景，介绍了国内外小基高比摄影测量尤其是高精度影像匹配方面的研究现状。 2.从数学的角度研究了基高比、噪声等因素对影像匹配精度和交会精度的影响。研究表明，在小基高比条件下可以通过提升影像匹配的精度弥补交会精度的损失。 3.在深入分析小基高比摄影测量理论的基础上，设计相关实验，利用实际影像探讨了基高比、影像匹配精度和交会精度之间的联系，验证了小基高比条件下通过提升影像匹配精度获取可靠高程信息的可能性。 4.分析了局部匹配算法和全局匹配算法的优势与不足，结合小基高比影像的特点，，引入了对纹理重复区域、阴影区域和断裂区域都能有效处理的半全局匹配算法，为小基高条件下高精度影像匹配算法的研究提供稳定可靠的初值。在此基础上，验证了半全局匹配算法在不同区域匹配的可靠性和稳定性。 5.基于相位相关基础理论，分别采用局部上采样技术和奇异值分解技术对相位相关互功率谱矩阵进行处理，探讨了两种频率域的高精度影像匹配算法。实验表明，基于相位相关的高精度影像匹配算法，在一定条件下能够获得较高的匹配精度。
[Abstract]:Small base-to-height ratio photogrammetry can effectively solve the difficult problem of traditional photogrammetry in terrain mapping in complex areas, but it requires extremely accurate parallax measurement accuracy. High-precision image matching is the key to achieve high-precision positioning under the condition of small base-to-height ratio. On the basis of deep analysis of the theory of small base height ratio photogrammetry, this paper discusses the relationship among base height ratio, image matching precision and rendezvous precision, and deeply studies the image pre-matching under the condition of small base height ratio. High-precision image matching and other aspects of the content, has made some progress. The main work and innovations completed in this paper are as follows: 1. This paper discusses the research significance and background of this paper, and introduces the research status of small base-height ratio photogrammetry, especially high-precision image matching at home and abroad. 2. The effects of fundamental height ratio, noise and other factors on image matching accuracy and rendezvous precision are studied from the point of view of mathematics. The research shows that the loss of rendezvous precision can be compensated by improving the accuracy of image matching under the condition of small base-height ratio. 3. On the basis of in-depth analysis of the theory of small base-height ratio photogrammetry, relevant experiments are designed and the relationship among base-height ratio, image matching accuracy and rendezvous precision is discussed by using actual images. The possibility of obtaining reliable elevation information by improving the image matching accuracy under the condition of small base-to-height ratio is verified. 4. The advantages and disadvantages of local matching algorithm and global matching algorithm are analyzed. A semi-global matching algorithm, which can deal with texture repeat region, shadow region and fracture region effectively, is introduced according to the characteristics of small base-to-height ratio image. It provides a stable and reliable initial value for the research of high precision image matching algorithm under small base and high condition. On this basis, the reliability and stability of the semi-global matching algorithm in different regions are verified. 5. Based on the phase correlation theory, the phase correlation mutual power spectrum matrix is processed by local upsampling technique and singular value decomposition (SVD) technique, and two high precision image matching algorithms in frequency domain are discussed. Experimental results show that the high-precision image matching algorithm based on phase correlation can obtain high matching accuracy under certain conditions.
【学位授予单位】：解放军信息工程大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：P236

【参考文献】