基于LK光流和网格运动统计的图像匹配改进算法

doi:10.16180/j.cnki.issn1007-7820.2022.05.001

电子科技 ›› 2022, Vol. 35 ›› Issue (5): 1-6.doi: 10.16180/j.cnki.issn1007-7820.2022.05.001

• • 下一篇

基于LK光流和网格运动统计的图像匹配改进算法

刘群坡,席秀蕾,杨凌霄

河南理工大学电气工程与自动化学院,河南焦作 454000

收稿日期:2020-12-26 出版日期:2022-05-25 发布日期:2022-05-27
作者简介:刘群坡(1978-),男,博士,副教授。研究方向:机器视觉、特殊环境机器人控制及智能装备。|席秀蕾(1995-),女,硕士研究生。研究方向:机器视觉。
基金资助:
国家重点研发计划(2016YFC0600906);河南省高校科技创新团队(20IRTSTHN019);河南省创新型科技人才队伍建设工程(CXTD2016054);河南理工大学博士基金(722103/001/070)

Improved Image Matching Algorithm Based on LK Optical Flow and Grid Motion Statistics

LIU Qunpo,XI Xiulei,YANG Lingxiao

School of Electrical Engineering and Automation,Henan Polytechnic University,Jiaozuo 454000,China

Received:2020-12-26 Online:2022-05-25 Published:2022-05-27
Supported by:
National Key R&D Program(2016YFC0600906);Henan University Science and Technology Innovation Team(20IRTSTHN019);Construction Project of Innovative Scientific and Technological Talents in Henan(CXTD2016054);Doctor Foundation of Henan Polytechnic University of Technology(722103/001/070)

摘要/Abstract

摘要：

针对AKAZE算法在玻璃封装电连接器图像匹配中准确率低及耗时长的问题,文中提出了基于LK光流和网格运动统计的图像匹配改进算法。首先采用AKAZE算法提取玻璃封装电连接器特征点,运用M-LDB描述子进行特征描述。然后利用LK光流法计算匹配区域进行条件约束,从而得到匹配点,并通过FLANN算法进行特征匹配。最后将玻璃封装电连接器图像划分为多个网格,计算FLANN匹配好的特征点邻域的正确匹配个数和阈值并剔除误匹配点。文中实验采用公开数据集和玻璃封装电连接器数据,从实时性和准确性两方面对算法性能进行验证分析。结果表明,改进的算法在处理模糊、亮度及旋转变化的玻璃封装电连接器图像对时,其匹配准确率均达到93%以上,耗时为0.4 s内,证明了算法的有效性。

关键词: 双目视觉定位, 玻璃封装电连接器, AKAZE算法, LK光流, 特征, 匹配, FLANN算法, 网格运动统计

Abstract:

In order to solve the low accuracy and time-consuming problem of AKAZE algorithm in the image matching of glass-encapsulated electrical connectors, improved image matching algorithm based on LK optical flow and grid motion statistics is proposed in this study. First, the AKAZE algorithm is used to extract feature points, and the M-LDB descriptor is used to describe the features. Then, LK optical flow method is used to calculate the matching area for conditional constraints to obtain matching points, and the FLANN algorithm is adopted for feature matching. Finally, the glass-encapsulated electrical connector image is divided into multiple grids, and the numbers and threshold values of correct matching points of the neighborhood of the feature points that are matched by FLANN are calculated to eliminate the wrong matching points. By using public datasets and glass-encapsulated electrical connector data,the performance of the algorithm is verified and analyzed from both real-time and accuracy aspects. The results show that the improved algorithm has a matching accuracy of more than 93% when dealing with image pairs of glass-encapsulated electrical connectors with blur, brightenss and rotation changes, and the time-consuming is within 0.4 s, which proves the effectiveness of the algorithm.

Key words: binocular vision positioning, glass-encapsulated electrical connector, AKAZE algorithm, LK optical flow, feature, matching, FLANN algorithm, grid motion statistics

中图分类号:

TP391

刘群坡,席秀蕾,杨凌霄. 基于LK光流和网格运动统计的图像匹配改进算法[J]. 电子科技, 2022, 35(5): 1-6.

LIU Qunpo,XI Xiulei,YANG Lingxiao. Improved Image Matching Algorithm Based on LK Optical Flow and Grid Motion Statistics[J]. Electronic Science and Technology, 2022, 35(5): 1-6.

图/表 9

图1

图2

图3

图4

图5

表1

图6

图7

图8

参考文献 19

[1]	Zhang J, Feng Z, Zhang J, et al. An improved randomized local binary features for keypoints recognition[J]. Sensors, 2018, 18(6):1937-1961. doi: 10.3390/s18061937
[2]	Lowe G. SIFT-the scale invariant feature transform[J]. International Journal, 2004(2):91-110.
[3]	Bay H, Tuytelaars T, Van Gool L. Surf: Speeded up robust features[C]. Heidelberg: Proceedings of the Ninth European Conference on Computer Vision, 2006.
[4]	Bay H, Ess A, Liu C, et al. Speeded-up robust features[J]. Computer Vision and Image Understanding, 2008, 110(3):346-359. doi: 10.1016/j.cviu.2007.09.014
[5]	Shao Z, Chen M, Liu C, et al. Feature matching for illumination variation images[J]. Journal of Electronic Imaging, 2015, 24(3):1-11.
[6]	邢凯盛, 凌有铸, 陈孟元, 等. ORB特征匹配的误匹配点剔除算法研究[J]. 电子测量与仪器学报, 2016, 30(8):1255-1262.
	Xing Kaisheng, Ling Youzhu, Chen Mengyuan, et al. Mismatching points elimination algorithm for ORB feature matching[J]. Journal of Electronic Measurement and Instrumentation, 2016, 30(8):1255-1262.
[7]	Lin W Y, Liu S, Jiang N, et al. RepMatch: robust feature matching and pose for reconstructing modern cities[C]. Amsterdam: Proceedings of the European Conference on Computer Vision, 2016.
[8]	Tang P, Peng Y. Exploiting distinctive topological constraint of local feature matching for logo image recognition[J]. Neurocomputing, 2017, 23(6):113-122.
[9]	Bian J W, Lin W Y, Matsushita Y, et al. GMS: grid-based motion statistics for fast, ultra-robust feature correspondence[C]. Honolulu: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017.
[10]	Sedaghat A, Mohammadi N. Uniform competency-based local feature extraction for remote sensing images[J]. Journal of Photogrammetry and Remote Sensing, 2018, 13(5):142-157.
[11]	Prakash C S, Panzade P P, Om H. Detection of copy-move forgery using AKAZE and SIFT key point extraction[J]. Multimedia Tools and Applications, 2019, 78(16):23535-23558. doi: 10.1007/s11042-019-7629-x
[12]	颜焕欢, 张培镇, 王伊侬, 等. 肌骨超声图像特征检测及拼接[J]. 中国图象图形学报, 2020, 25(5):1032-1042.
	Yan Huanhuan, Zhang Peizhen, Wang Yinong, et al. Feature detection algorithm of musculoskeletal ultrasound image and its application of image stitching[J]. Journal of Image and Graphics, 2020, 25(5):1032-1042.
[13]	徐睦浩, 衣禹桥, 李英群, 等. 基于边缘位置与颜色信息的不规则碎纸片拼接方法[J]. 陕西师范大学学报(自然科学版), 2020, 48(6):90-95.
	Xu Muhao, Yi Yuqiao, Li Yingqun, et al. A method of stitching irregular scrapped paper based on edge position and color information[J]. Journal of Shaanxi Normal University(Natural Science Edition), 2020, 48(6):90-95.
[14]	那彦, 廖萌萌. 基于傅里叶梅林变换的SURF算法[J]. 电子科技, 2016, 29(7):55-57.
	Na Yan, Liao Mengmeng.A SURF algorithm based on the Fourier merlin transform[J]. Electronic Science and Technology, 2016, 29(7):55-57.
[15]	严茂森, 陈家琪. 一种多环境适用的交通标志识别模型[J]. 电子科技, 2019, 32(4):68-76.
	Yan Maosen, Chen Jiaqi. A multi-environmental traffic sign recognition model[J]. Electronic Science and Technology, 2019, 32(4):68-76.
[16]	王招娣, 王贤立, 杨数强. 基于颜色和SURF特征的图像匹配算法[J]. 电子科技, 2017, 30(12):67-70.
	Wang Zhaodi, Wang Xianli, Yang Shuqiang. Image matching algorithm based on color and SURF feature[J]. Electronic Science and Technology, 2017, 30(12):67-70.
[17]	于雯越, 安博文, 赵明. 基于光流法与RPCA的红外运动目标检测[J]. 现代计算机, 2018(23):66-71.
	Yu Wenyue, An Bowen, Zhao Ming. Infrared moving target detection based on RPCA and optical flow algorithm[J]. Modern Computer, 2018(23):66-71.
[18]	程俊廷, 郭博洋, 田宽. 改进的LK光流法在SLAM中的应用[J]. 黑龙江科技大学学报, 2019, 29(6):736-740.
	Cheng Junting, Guo Boyang, Tian Kuan. Application of improved LK optical flow method in SLAM[J]. Journal of Heilongjiang University of Science and Technology, 2019, 29(6):736-740.
[19]	Yang X, Huang C, Cheng K T. libLDB:a library for extracting ultrafast and distinctive binary feature description[C]. Orlando: Proceedings of the Twenty-second ACM International Conference on Multimedia, 2014.

图像变化	算法	匹配点对数	正确点对数	CMR/%	时间/s
	SIFT	2 670	1 484	55.58	0.71
旋转变化	SURF	769	319	41.18	1.34
旋转变化	AKAZE	481	277	57.59	0.18
	本文算法	481	449	93.39	0.40
	SIFT	1 671	114	6.82	0.27
光照变化	SURF	1 507	312	20.70	1.31
光照变化	AKAZE	371	176	47.44	0.15
	本文算法	371	349	94.07	0.25
	SIFT	394	160	40.61	0.24
模糊变化	SURF	918	470	51.19	1.27
模糊变化	AKAZE	329	203	61.70	0.18
	本文算法	329	313	95.07	0.20

基于LK光流和网格运动统计的图像匹配改进算法

Improved Image Matching Algorithm Based on LK Optical Flow and Grid Motion Statistics

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 19

相关文章 15

Metrics

本文评价

推荐阅读 0

[1]	程顺达,程颖,孙士江. 基于机器学习的肿瘤智能辅助诊断方法[J]. 电子科技, 2022, 35(5): 56-59.
[2]	林潮威,李菲菲,陈虬. 基于深度卷积特征的场景全局与局部表示方法[J]. 电子科技, 2022, 35(4): 20-27.
[3]	陈汗青,李菲菲,陈虬. 基于三维卷积和哈希方法的视频检索算法[J]. 电子科技, 2022, 35(4): 35-39.
[4]	华莉,杨俭,袁天辰,宋瑞刚. 基于VMD和BP神经网络的轨道病害诊断方法[J]. 电子科技, 2022, 35(4): 40-46.
[5]	纪玲玉,高永彬,赵呈陆,汤先华,徐凯成,徐嘉诚. 基于3D全卷积网络的腹部动脉CTA分割算法[J]. 电子科技, 2022, 35(3): 38-44.
[6]	齐美义,廖爱华. 基于MATLAB App Designer的牵引电机滚动轴承可靠性评估系统设计[J]. 电子科技, 2022, 35(3): 79-86.
[7]	伍伟嘉,杨俭,袁天辰,邵志慧. 基于支持向量机的轨道结构病害识别研究[J]. 电子科技, 2022, 35(2): 27-33.
[8]	李辉,王一丞. 基于CNNCIFG-Attention模型的文本情感分类[J]. 电子科技, 2022, 35(2): 46-51.
[9]	邵志慧,杨俭,袁天辰,伍伟嘉. 基于排列熵和支持向量机的轨枕病害诊断[J]. 电子科技, 2022, 35(2): 52-58.
[10]	李刚. 频变耦合双通带滤波器的综合方法[J]. 电子科技, 2022, 35(1): 1-5.
[11]	刘轩,付东翔. 基于人脸姿态估计的多角度虚拟眼镜试戴技术[J]. 电子科技, 2021, 34(9): 58-65.
[12]	赵崇,迟蒙蒙,储聪,张鹏. 导盲犬行走机构运动仿真及其视觉识别算法研究[J]. 电子科技, 2021, 34(9): 66-72.
[13]	陆玮,刘翔,薛冕. 基于超像素分割的孪生网络双目立体匹配方法研究[J]. 电子科技, 2021, 34(9): 73-78.
[14]	王庆文,胡海洋. 基于时序行为检测的工作流识别[J]. 电子科技, 2021, 34(8): 14-18.
[15]	司明明,陈玮,胡春燕,尹钟. 融合Resnet50和U-Net的眼底彩色血管图像分割[J]. 电子科技, 2021, 34(8): 19-24.