[1]刘向增,徐雪灵,刘如意,等.面向图像匹配的局部特征提取研究进展[J].计算机技术与发展,2022,32(02):1-13.[doi:10. 3969 / j. issn. 1673-629X. 2022. 02. 001]
　LIU Xiang-zeng,XU Xue-ling,LIU Ru-yi,et al.Research Progress of Local Feature Extraction for Image Matching[J].,2022,32(02):1-13.[doi:10. 3969 / j. issn. 1673-629X. 2022. 02. 001]
点击复制

面向图像匹配的局部特征提取研究进展()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
32
期数:
2022年02期
页码:
1-13
栏目:
综述
出版日期:
2022-02-10

文章信息/Info

Title:
Research Progress of Local Feature Extraction for Image Matching
文章编号:
1673-629X(2022)02-0001-13
作者:
刘向增; 徐雪灵; 刘如意; 宋建锋; 苗启广
西安电子科技大学 计算机科学与技术学院,陕西 西安 710071
Author(s):
LIU Xiang-zeng; XU Xue-ling; LIU Ru-yi; SONG Jian-feng; MIAO Qi-guang
School of Computer Science and Technology,Xidian University,Xi’an 710071,China
关键词:
图像匹配; 局部特征; 特征描述; 不变特征; 深度学习
Keywords:
image matching; local feature; feature descriptor; invariant feature; deep learning
分类号:
TP311. 5
DOI:
10. 3969 / j. issn. 1673-629X. 2022. 02. 001
摘要:
图像匹配作为计算机视觉领域的重要研究方向,广泛应用于图像配准、图像融合、变化检测、视觉导航、3D 重建、视觉同时定位与地图构建( SLAM) 等领域,精确稳健的局部特征提取是实现其高效处理的前提与关键。 以图像匹配研究为导向,从传统特征设计到现代特征学习对局部特征提取方法进行了分类总结,首先,为增强对现代局部特征提取方法的理解,重点介绍了基于传统特征设计的相关方法,接着回顾了基于经典机器学习的方法,搭建起传统方法到深度学习方法的桥梁,最后详细讨论了基于深度学习的现代特征提取方法。 针对跨传感器、多视角、不同时段环境下的图像匹配需求,全面分析了各阶段主流方法的优缺点,提出了目前存在的问题与挑战,并给出了相应的研究建议,为相关研究人员全面深入理解图像局部特征提取方法并利用深度学习方法对其进行改进提供基础性参考。
Abstract:
As an important research direction in the field of computer vision,image matching is widely used in image registration,imagefusion,change detection,visual navigation,3D reconstruction, simultaneous visual positioning and map construction ( SLAM) and otherfields. Among them,accurate and robust local feature extraction is the prerequisite and a key step to achieve efficient feature matching.Guided by image matching research,the local feature extraction methods are classified and summarized from traditional feature design tomodern feature learning. Firstly,in order to enhance the understanding of modern local feature extraction methods,we focus on the relatedmethods based on traditional feature design,then review the methods based on classic machine learning,build a bridge from traditionalmethods to deep learning methods,and finally discuss in detail based on modern feature extraction methods for deep learning. In view ofthe requirements of image matching under cross-sensor,multi-view,and different time periods,the advantages and disadvantages of mainstream methods at each stage are comprehensively analyzed,and the current problems and challenges of local feature extraction in imagematching are put forward,and corresponding research recommendations are proposed. It provides a basic reference for related researchersto fully understand the topic of local feature extraction and improve the results with deep learning.

相似文献/References:

[1]冉柯柯 王继成.基于比值法图像拼接的等比例改进算法[J].计算机技术与发展,2010,(02):5.
　RAN Ke-ke,WANG Ji-cheng.An Improved Mosaic Algorithm Based on Ratio Matching Using Geometric Proportion[J].,2010,(02):5.
[2]肖若秀 蔡光程 贾建波.利用旋转模板匹配方法对SIFT算法的改进[J].计算机技术与发展,2009,(05):127.
　XIAO Ruo-xiu,CAI Guang-cheng,JIA Jian-bo.Using a Rotated Template to Improve SIFT＇s Processing[J].,2009,(02):127.
[3]张宇 黄亚博 焦建彬.一种适用于高分辨率图像的实时电子稳像算法[J].计算机技术与发展,2009,(03):9.
　ZHANG Yu,HUANG Ya-bo,JIAO Jian-bin.A Real Time Stabilization Algorithm for High Resolution Video[J].,2009,(02):9.
[4]刘忠艳 周波 车向前.一种高效的图像匹配算法[J].计算机技术与发展,2009,(04):45.
　LIU Zhong-yan,ZHOU Bo,CHE Xiang-qian.An Effective Algorithm for Image Registration[J].,2009,(02):45.
[5]何志明 马苗.基于灰色关联分析和人工蜂群算法的图像匹配方法[J].计算机技术与发展,2010,(10):78.
　HE Zhi-ming,MA Miao.Fast Image Matching Approach Based on Grey Relational Analysis and Artificial Bee Colony Algorithm[J].,2010,(02):78.
[6]于二丽 周宁宁.基于Hausdorff距离的图像匹配并行算法设计与实现[J].计算机技术与发展,2011,(09):28.
　YU Er-li,ZHOU Ning-ning.Design and Implementation of a Parallel Image Matching Algorithm Based on Hausdorff Distance[J].,2011,(02):28.
[7]陈晨,吴建国.基于WebRTC的残疾人鼠标研究与实现[J].计算机技术与发展,2013,(09):32.
　CHEN Chen,WU Jian-guo.Research and Implementation of Disabled Mouse Based on WebRTC[J].,2013,(02):32.
[8]汪昡紫,孙宪坤,高飞. 轨道表面图像处理算法研究[J].计算机技术与发展,2015,25(09):182.
　WANG Xuan-zi,SUN Xian-kun,GAO Fei. Research on Algorithm of Track Surface Image Processing[J].,2015,25(02):182.
[9]陈星[],徐迎晖[],肖青海[][]. QR码印刷品缺陷检测[J].计算机技术与发展,2015,25(10):191.
　CHEN Xing[],XU Ying-hui[],XIAO Qing-hai[][]. Defect Detection of Printed QR Code Image[J].,2015,25(02):191.
[10]李炀,翟社平. 改进的SIFT图像匹配算法[J].计算机技术与发展,2016,26(11):58.
　LI Yang,ZHAI She-ping. Improved SIFT Image Matching Algorithm[J].,2016,26(02):58.
[11]郑权,刘循,魏海明.基于学习的霍夫变换线段组物体检测算法[J].计算机技术与发展,2014,24(01):26.
　ZHENG Quan,LIU Xun,WEI Hai-ming.An Object Detection Algorithm of Hough Transform Line Segmentation Groups Based on Learning[J].,2014,24(02):26.
[12]刘加运,李玉惠,李勃,等. 一种多维特征融合的车辆对象同一性匹配方法[J].计算机技术与发展,2016,26(04):167.
　LIU Jia-yun,LI Yu-hui,LI Bo,et al. A Vehicle Object Identity Matching Method of Multidimensional Feature Combination[J].,2016,26(02):167.
[13]冯春,吴小锋,尹飞鸿,等. 基于局部特征匹配的双焦单目立体视觉深度估计[J].计算机技术与发展,2016,26(10):55.
　FENG Chun,WU Xiao-feng,YIN Fei-hong,et al. Depth Estimation for Bifocal Monocular Stereo Vision Based on Local Image Feature Descriptors Matching[J].,2016,26(02):55.

更新日期/Last Update: 2022-02-10