摘要
基于特征点的匹配方法是图像配准中必不可少的过程。但是,对于经过仿射变化的参考图像而言,得到的特征点匹配结果中常常有很多的误匹配特征点,造成不理想的特征点匹配结果。原因在于目前对于匹配结果的评价只能采用线性结构进行评价而忽略了非线性结构带来的影响。本文提出了一种基于机器学习的比较特征向量方法,用模式分类问题替代现有的匹配方法。实验结果表明,该算法对误匹配点对的剔除有较好的效果。
Feature point matching is of central importance in feature-based image registration algorithms.However, since most of the existing feature matchings are not so powerful and efficient for images after affine transformation. The underlying reason is that a suitable measure is unavailable currently in literature to adequately evaluate the similarity of two images. In contrast to the traditional evaluation methods, in this paper, the registration evaluation problem is converted into a classification problem through machine learning. Experiments show that the validity and reliability of this proposed method.
出处
《科技广场》
2016年第2期5-8,共4页
Science Mosaic
基金
江西理工大学校级重点课题(编号:NSFJ2014-K18)
关键词
图像配准
特征点误匹配
机器学习
分类器
Image Registration
Feature Mismatches
Machine Learning
Classifier