可见光和红外反相图像的SURF特征双向匹配方法.doc

文档介绍：
可见光和红外反相图像的 SURF 特征双向匹
配方法#
纪利娥,杨风暴,王志社,陈磊**
5
10
15
(中北大学信息与通信工程学院,山西太原,030051)
摘要:特征匹配的准确率影响图像配准的精度,是基于特征配准方法的重点和难点之一。为
了解决单向最近邻/次近邻法所导致特征点一对多的误匹配问题,提出了一种红外和可见光
图像的特征双向匹配方法。首先,对红外图像进行反相和直方图均衡化处理,增强两类图像
的相似性,提取数量更多重复率高的共有特征;其次,对提取的 SURF(speed-up robust feature)
特征进行双向最近邻/次近邻粗匹配,确保特征匹配的一致性,降低误匹配率,并利用
RANSAC(random sample consensus)算法对特征点进行二次匹配,实现特征点精确匹配。实
验结果表明,该算法在正确匹配率和配准精度方面都优于传统 SURF 的单向最近邻/次近邻
匹配方法,具有有效性。
关键词:图像配准;可见光图像;红外图像反相;SURF 特征;双向匹配
中图分类号:
Bi-directional matching algorithm based on SURF features
for visible and negative image of infrared image
20
JI Lie, YANG Fengbao, WANG Zhishe, CHEN lei
(Information munication Engineering College,North University of
China,Taiyuan,030051)
Abstract: Feature matching accuracy affects the precision of image registration,which is one of
25
30
35
40
difficult key points for image registration base on order to solve feature points
one-to-many mismatching problem caused by the ratio of the closest neighbor and second closest
neighbor from one direction,this paper put forward a bidirectional matching method of features for
image registration of visible and infrared ,for enhancing the similarity of two
images,image reverse and histogram equalization are adopted to process infrared image,so that
more consistent features of high repetition rate are ,SURF fea