基于边缘特征点对对齐度的图像配准方法

针对基于特征的图像配准方法存在特征提取的多样性和相似度计算的复杂性等问题,在定义边缘特征点对的角度直方图和对齐度的基础上,提出了一种基于边缘特征点对对齐度的图像配准方法。该方法首先利用小波多尺度积准确地提取边缘图像和特征点,然后根据特征点的角度直方图得到的旋转角度,并通过计算所有特征点对在边缘图像中的对齐度来精确地确定匹配点对。大量的实验结果表明,该方法具有较强的适用性、精确性和有效性。

A New Method for Image Registration Based on Alignment Metric of Edge Feature Points

Variety of detected features and complexity of similarity calculation are two main problems in image registration. In this paper, angle histogram and alignment metric of edge feature points are defined, and a new approach for image registration is presented based on alignment metric of edge feature points. The first step is to detected edge image and feature points by using wavelet multi-scale product. In this step, wavelet multi-scale product can ensure the accuracy of edge image and feature points. The second step is to calculate alignment metric of feature points based on rotating angle obtained from angle histogram. Then mapping points can be accurately determined by an appropriate threshold. The calculation complexity of alignment metric is greatly simplified since the process is based on edge images. Several experiments conducted illustrate that this method have a good performance of robust, accuracy and efficiency, and also it can be applied to many situations.