Fast image matching algorithm with approximate affine and scale invariance

Juan YUE; Si-li GAO; Fan-ming LI; Neng-bin CAI

doi:10.37188/OPE.20202810.2349

您当前的位置：

首页 >

文章列表页 >

Fast image matching algorithm with approximate affine and scale invariance

Information Sciences | 更新时间：2020-11-16

- Fast image matching algorithm with approximate affine and scale invariance
- Optics and Precision Engineering Vol. 28, Issue 10, Pages: 2349-2359(2020)
- 作者机构：
  
  1.中国科学院上海技术物理研究所智能红外感知重点实验室, 上海 200083
  2.上海市现场物证重点试验室, 上海 200083
- 作者简介：
- 基金信息：
- DOI：10.37188/OPE.20202810.2349
  CLC： TP391.41
- Received：25 March 2020，
  
  Revised：17 May 2020，
  
  Accepted：17 May 2020，
  
  Published：25 October 2020
- 稿件说明：
移动端阅览
Juan YUE, Si-li GAO, Fan-ming LI, et al. Fast image matching algorithm with approximate affine and scale invariance[J]. Optics and precision engineering, 2020, 28(10): 2349-2359.
DOI：

Juan YUE, Si-li GAO, Fan-ming LI, et al. Fast image matching algorithm with approximate affine and scale invariance[J]. Optics and precision engineering, 2020, 28(10): 2349-2359. DOI： 10.37188/OPE.20202810.2349.

摘要

为了解决大仿射形变场景下，尺度不变特征变换（Scale Invariant Feature Transform，SIFT）算法的局限性以及仿射尺度不变特征变换（Affine-SIFT，ASIFT）算法运算量大的问题，提出了一种具有近似仿射尺度不变特征的快速图像匹配算法（Fast Approximate-Affine-SIFT，Fast-AASIFT）。该算法具有比ASIFT算法更清晰的物理意义，首先从逆仿射变换出发，对原图进行仿射形变纠正，估计出对应的正射图像；然后在正射图像上进行特征点提取及SIFT描述；最后进行SIFT优化匹配。实验结果表明：大仿射形变场景下，Fast-AASIFT算法依然能匹配到足够多的特征点，且峰值匹配误差 < 2.5 pixel，平均匹配误差 < 1.2 pixel，其抗仿射形变能力明显优于SIFT算法，与ASIFT算法相当；Fast-AASIFT算法耗时 < 0.3倍ASIFT，有效改善了ASIFT算法的耗时问题。可见，Fast-AASIFT算法既有效保证了算法抗仿射形变鲁棒性，又大幅提高了算法效率，对场景重构、场景识别等应用具有重要意义。

Abstract

To address the limitations of the scale invariant feature transform (SIFT) algorithm and reduce the computational burden of the Affine-SIFT (ASIFT) algorithm in scenes with large affine deformations

a fast image matching algorithm based on approximate-affine-SIFT (Fast-AASIFT) is proposed. Fast-AASIFT has a clearer physical meaning than the ASIFT algorithm. First

Fast-AASIFT recovers original images as rectified images by performing inverse affine transformations. Then

it performs feature point extraction and SIFT description on the rectified images. Finally

it performs SIFT optimization matching. The experimental results demonstrate that

in scenes with a large affine deformation

Fast-AASIFT can still match enough feature points

with a peak matching error of < 2.5 pixels and an average matching error of < 1.2 pixels. This proves that the anti-affine deformation ability of Fast-AASIFT is equivalent to that of the ASIFT algorithm

which is significantly better than that of the SIFT algorithm. Furthermore

the time consumed by Fast-AASIFT less than 30% of that consumed by the ASIFT algorithm; thus

it effectively addresses the time-consumption problem of the ASIFT algorithm. Obviously

Fast-AASIFT not only maintains good robustness against affine deformations but also greatly improves computational efficiency; consequently

it is of great value for applications such as scene reconstruction and recognition.

关键词

Keywords

references

张振超.多视角倾斜航空影像匹配技术研究[D].武汉: 解放军信息工程大学, 2015.

ZHANG ZH CH. Research on Multi-view Oblique Aerial Image Matching[D]. Wuhan: Graduate PLA Information Engineering University. (in Chinese)

潘九宝 , 王玮 , 黄明伟 , 等 . 采用倾斜摄影测量和几何造型方法的城市实景三维建模 . 测绘通报 , 2020 . ( 6 ): 121 - 124 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=chtb202006024 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=chtb202006024 .

J B PAN , W WANG , M W HUANG , 等 . Urban reality 3D model construction using inclined photogrammetry and geometric modeling . Bulletin of Surveying and Mapping , 2020 . ( 6 ): 121 - 124 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=chtb202006024 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=chtb202006024 .

李在林.基于多视图几何的三维重建技术研究[D].北京: 中国科学院大学, 2020.

LI Z L. A Research of 3D Reconstruction Technology Based on Multiple View Geometry[D]. Beijing: Graduate University of Chinese Academy of Sciences. (in Chinese)

张玉珍 , 侯守明 , 连盼盼 , 等 . 基于主方向重建的SURF多角度识别匹配算法优化 . 信息与控制 , 2019 . 48 ( 1 ): 42 - 49 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=xxykz201901007 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=xxykz201901007 .

Y ZH ZHANG , SH M HOU , P P LIAN , 等 . Multi-angle recognition and matching of SURF algorithm optimization based on principal direction reconstruction . Information and Control , 2019 . 48 ( 1 ): 42 - 49 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=xxykz201901007 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=xxykz201901007 .

肖雄武 , 李德仁 , 郭丙轩 , 等 . 一种具有视点不变性的倾斜影像快速匹配方法 . 武汉大学学报:信息科学版 , 2016 . 41 ( 9 ): 1151 - 1159 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=whchkjdxxb201609003 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=whchkjdxxb201609003 .

X W XIAO , D R LI , B X GUO , 等 . A robust and rapid viewpoint-invariant matching method for oblique images . Journal of Wuhan University:Geomatics and Information Science Edition , 2016 . 41 ( 9 ): 1151 - 1159 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=whchkjdxxb201609003 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=whchkjdxxb201609003 .

詹总谦 , 李一挥 , 王陈东 , 等 . 顾及局部相对几何变形改正的影像匹配和空三逐步精化方法 . 武汉大学学报:信息科学版 , 2018 . 43 ( 11 ): 1620 - 1627 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=whchkjdxxb201811003 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=whchkjdxxb201811003 .

Z Q ZHAN , Y H LI , CH D WANG , 等 . A stepwise refinement method for image matching and aerotriangulation using correction of local relative geometric distortions . Journal of Wuhan University:Geomatics and Information Science Edition , 2018 . 43 ( 11 ): 1620 - 1627 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=whchkjdxxb201811003 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=whchkjdxxb201811003 .

马国宝 , 俞友 . 多特征结合的倾斜无人机影像匹配方法 . 地理信息世界 , 2019 . 26 ( 2 ): 116 - 119 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dlxxsj201902022 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dlxxsj201902022 .

G B MA , Y YU . UAV tilted image matching algorithm with multi-feature combination . Geomatics World , 2019 . 26 ( 2 ): 116 - 119 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dlxxsj201902022 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dlxxsj201902022 .

J M MOREL , G YU . ASIFT:A new framework for fully affine invariant image comparison . SIAM journal on imaging sciences , 2009 . 2 ( 2 ): 438 - 469 . DOI: 10.1137/080732730 http://doi.org/10.1137/080732730 .

Y HUACHAO , Z SHUBI , W YONGBO . Robust and precise registration of oblique images based on scale-invariant feature transformation algorithm . IEEE Geoscience and Remote Sensing Letters , 2012 . 9 ( 4 ): 783 - 787 . DOI: 10.1109/LGRS.2011.2181485 http://doi.org/10.1109/LGRS.2011.2181485 .

肖雄武 , 郭丙轩 , 李德仁 , 等 . 一种具有仿射不变性的倾斜影像快速匹配方法 . 测绘学报 , 2015 . 44 ( 4 ): 414 - 421 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=chxb201504009 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=chxb201504009 .

X W XIAO , B X GUO , D R LI , 等 . A quick and affine invariance matching method for oblique images . Acta Geodaetica et Cartographica Sinica , 2015 . 44 ( 4 ): 414 - 421 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=chxb201504009 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=chxb201504009 .

杨晓敏 , 吴炜 , 卿粼波 , 等 . 图像特征点提取及匹配技术 . 光学精密工程 , 2009 . 17 ( 9 ): 2276 - 2282 . http://ope.lightpublishing.cn/thesisDetails?columnId=1750067&Fpath=&index=-1&l=zh http://ope.lightpublishing.cn/thesisDetails?columnId=1750067&Fpath=&index=-1&l=zh .

X M YANG , W WU , L B QIN , 等 . Image feature extraction and matching technology . Opt. Precision Eng. , 2009 . 17 ( 9 ): 2276 - 2282 . http://ope.lightpublishing.cn/thesisDetails?columnId=1750067&Fpath=&index=-1&l=zh http://ope.lightpublishing.cn/thesisDetails?columnId=1750067&Fpath=&index=-1&l=zh .

丁国绅 , 乔延利 , 易维宁 , 等 . 基于高光谱图像的改进SIFT特征提取与匹配 . 光学精密工程 , 2020 . 28 ( 4 ): 954 - 962 . http://ope.lightpublishing.cn/thesisDetails?columnId=1432597&Fpath=&index=-1&l=zh http://ope.lightpublishing.cn/thesisDetails?columnId=1432597&Fpath=&index=-1&l=zh .

G SH DING , Y L QIAO , W N YI , 等 . Improved SIFT feature extraction and matching technology based on hyperspectral image . Opt. Precision Eng. , 2020 . 28 ( 4 ): 954 - 962 . http://ope.lightpublishing.cn/thesisDetails?columnId=1432597&Fpath=&index=-1&l=zh http://ope.lightpublishing.cn/thesisDetails?columnId=1432597&Fpath=&index=-1&l=zh .

耿庆田 , 赵浩宇 , 王宇婷 , 等 . 基于改进SIFT特征提取的车标识别 . 光学精密工程 , 2018 . 26 ( 5 ): 1267 - 1274 . http://ope.lightpublishing.cn/thesisDetails?columnId=1415856&Fpath=&index=-1&l=zh http://ope.lightpublishing.cn/thesisDetails?columnId=1415856&Fpath=&index=-1&l=zh .

Q T GENG , H Y ZHAO , Y T WANG , 等 . A vehicle logo recognition algorithm based on the improved SIFT feature . Opt. Precision Eng. , 2018 . 26 ( 5 ): 1267 - 1274 . http://ope.lightpublishing.cn/thesisDetails?columnId=1415856&Fpath=&index=-1&l=zh http://ope.lightpublishing.cn/thesisDetails?columnId=1415856&Fpath=&index=-1&l=zh .

张昊骕 , 朱晓龙 , 胡新洲 , 等 . 基于SURF和SIFT特征的视频镜头分割算法 . 液晶与显示 , 2019 . 34 ( 5 ): 521 - 529 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=yjyxs201905012 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=yjyxs201905012 .

H S ZHANG , X L ZHU , X ZH HU , 等 . Shot segmentation technology based on SURF features and SIFT features . Chinese Journal of Liquid Crystals and Displays , 2019 . 34 ( 5 ): 521 - 529 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=yjyxs201905012 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=yjyxs201905012 .

刘凯 , 汪侃 , 杨晓梅 , 等 . 基于DoG检测图像特征点的快速二进制描述子 . 光学精密工程 , 2020 . 28 ( 2 ): 485 - 496 . http://ope.lightpublishing.cn/thesisDetails?columnId=2125983&Fpath=&index=-1&l=zh http://ope.lightpublishing.cn/thesisDetails?columnId=2125983&Fpath=&index=-1&l=zh .

K LIU , K WANG , X M YANG , 等 . DoG keypoint detection based fast binary descriptor . Opt. Precision Eng. , 2020 . 28 ( 2 ): 485 - 496 . http://ope.lightpublishing.cn/thesisDetails?columnId=2125983&Fpath=&index=-1&l=zh http://ope.lightpublishing.cn/thesisDetails?columnId=2125983&Fpath=&index=-1&l=zh .

樊彦国 , 柴江龙 , 许明明 , 等 . 基于ORB与RANSAC融合改进的图像配准 . 光学精密工程 , 2019 . 27 ( 3 ): 702 - 717 . http://ope.lightpublishing.cn/thesisDetails?columnId=1422165&Fpath=&index=-1&l=zh http://ope.lightpublishing.cn/thesisDetails?columnId=1422165&Fpath=&index=-1&l=zh .

Y G FAN , J L CHAI , M M XU , 等 . Improved fast Image registration algorithm based on ORB and RANSAC fusion . Opt. Precision Eng. , 2019 . 27 ( 3 ): 702 - 717 . http://ope.lightpublishing.cn/thesisDetails?columnId=1422165&Fpath=&index=-1&l=zh http://ope.lightpublishing.cn/thesisDetails?columnId=1422165&Fpath=&index=-1&l=zh .

Views

506

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Visible and infrared image matching based on CycleGAN-SIFT

Improved SIFT feature extraction and matching technology based on hyperspectral image

Visual measurement of high-temperature welding deformation for high-strength steel sheet

Improved hybrid spill-tree algorithm for fast target matching recognition of satellite images

Image mosaic algorithm based on area blocking and SIFT

Related Author

Feng WANG

Hu WEN

Tian HE

Xu MA

Shuai HAO

Yingqi WU

Li-li DU

Wei-ning YI

Related Institution

Institute of Physics and Electrical Engineering， Weinan Normal University

College of Safety Science and Engineering， Xi'an University of Science and Technology

College of Electrical and Control Engineering， Xi'an University of Science and Technology

Key Laboratory of Optical Calibration and Characterization, Anhui Institute of Optics and Fine Mechanics, Chinese Academy of Sciences

University of Science and Technology of China

AI问答

Address：No.3888 Dong Nanhu Road, Changchun, Jilin, China Postal code：130033
Tel：0431-86176855 Email：gxjmgc@ciomp.ac.cn
Technical support is provided by Beijing Founder electronics co., LTD 吉ICP备11002662号-17 京公网安备11010802024621
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰