摘要
提出一种基于特征点的多幅图像自动拼接算法。根据SIFT或SURF算法在图像的尺度空间中提取特征点,对特征点进行亚像素定位,并赋予主方向。根据特征点邻域信息分布计算得到特征向量后,基于k-d树进行最近邻和次最近邻搜索,利用最近邻特征点距离与次近邻特征点距离之比得到初始匹配点对。使用RANSAC(Random Sample Consensus)算法剔除错误匹配特征点对,同时对图像之间的变换参数进行鲁棒估计,使用多频带融合算法消除拼接痕迹。实验验证了该算法能够完成多幅图像的自动无缝拼接。
This paper presents an automatic multiple images stitching algorithm based on feature points. The algorithm first extracts the SIFT or SURF feature points from the scale space of the image, then locates feature points on the sub-pixel coordinates, and gives the main orientation. Initial feature points matching can be calculated by using the k-nearest neighbor search based on k-d tree, and distance ratio of the nearest neighbor feature point and the next nearest neighbor feature point. Then using RANSAC （Random Sample Consensus） algorithm to match the initial feature points set, while transformation parameters between the two images can be estimated robustly. Seamless multi-image automatic stitching can be obtained by using multi-band blending algorithm.
出处
《计算机工程与应用》
CSCD
2012年第34期152-157,共6页
Computer Engineering and Applications
基金
云南省应用基础研究项目(No.2011FZ013)
关键词
图像拼接
特征匹配
鲁棒估计
图像融合
image stitching
feature matching
robust estimation
image blending
