陈梦琪,,
张贺
北京航空航天大学仪器科学与光电工程学院 ??北京 ??100191
详细信息
作者简介:牛燕雄：男，1967年生，教授，博士生导师，研究方向为光电对抗、图像处理
陈梦琪：女，1993年生，硕士生，研究方向为景象匹配、目标识别
张贺：男，1993年生，硕士生，研究方向为目标检测、目标识别
通讯作者:陈梦琪 chenmengqi6@buaa.edu.cn
中图分类号:TP391.4计量
文章访问数:958
HTML全文浏览量:365
PDF下载量:47
被引次数:0
出版历程
收稿日期:2018-05-09
修回日期:2018-09-26
网络出版日期:2018-11-02
刊出日期:2019-03-01
Fast Scene Matching Method Based on Scale Invariant Feature Transform
Yanxiong NIU,Mengqi CHEN,,
He ZHANG
Institute of Instrument Science and Photoelectric Engineering, Beihang University, Beijing 100191, China
摘要
摘要:传统基于特征的景象匹配方法存在冗余点多、匹配精度低等问题，难以同时满足实时性及鲁棒性要求，对此，论文提出一种基于尺度不变特征变换(SIFT)的快速景象匹配方法。在特征提取阶段，采用高速分段特征检测器(FAST)在多尺度检测角点作为初始特征，经过高斯差分(DOG)算子在尺度空间中进行特征的2次筛选，简化原有遍历式的特征搜索过程；在特征匹配阶段，采用仿射模型模拟变换关系建立几何约束条件，克服SIFT算法由于忽略几何信息而产生的误匹配。实验表明：该方法在匹配精度和实时性方面均优于SIFT算法，且对光照、模糊、尺度等变换具有良好的鲁棒性，能够更好地实现景象匹配。
关键词:景象匹配/
尺度不变特征变换/
高速分段特征检测/
几何约束
Abstract:The traditional feature-based image matching method has many problems such as many redundant points and low matching accuracy, which can hardly meet the real-time and robustness requirements. In this regard, a fast scene matching method based on Scale Invariant Feature Transform (SIFT) is proposed. In the feature detection phase, FAST (Features from Accelerated Segment Test) is used to detect characteristics in multi-scale, after then, combining with Difference Of Gauss (DOG) operators to filter characteristics again. From this, the feature search process is simplified. In feature matching phase, the affine transformation model is used to simulate the transformation relation and establish the geometric constraint, to overcome the mismatching because of ignoring the geometric information. The experimental results show that the proposed method is superior to the SIFT in efficiency and precision, also has good robustness to light, blur and scale transformation, achieves scene matching better.
Key words:Scene matching/
Scale Invariant Feature Transform (SIFT)/
Features from Accelerated Segment Test (FAST)/
Geometric constraint
PDF全文下载地址:
https://jeit.ac.cn/article/exportPdf?id=8dd456ff-5afe-4d8b-b7f6-b99bb40dea57
