基于谱图理论的特征匹配原理研究

图像匹配是计算机视觉的重要研究领域,它们广泛应用于工业、农业、物体识别、遥感、生物医学以及军事等方面.基于谱图理论的图像匹配算法引起了人们越来越多的兴趣,该算法直接对图像中的特征进行处理,将高度复杂的经典算法转化为组合（离散）谱问题的简单求解,有效地降低了算法复杂度。文章对基于谱图理论的特征匹配算法进行了较为系统的探索,借助数学理论工具,对不同匹配算法中所用谱分解的原理和本质进行了研究。

Research on Principles for Features Matching Based on Spectral Graph Theory

Images matching is an important research area in computer vision eld,which is widely applied in industrial, agriculture, object recognition, remote sensing, biomedicine, military aair and many other elds. Recently, the algorithms for images matching based on spectral graph theory is the most popular research subject in the world. In the algorithm,features in images are processed directly, and the solution becomes rather simple. The algorithm reduces the complexity effectively. The systematic explore on the algorithm of features matching based on spectral graph theory is addressed in this dissertation. With the help of mathematical theory tools, the research on the principles and essence of the spectral decomposition in the dierent matching algorithms is addressed, too.