连接向量特征匹配的暗场图像配准方法

黄卓; 陈凤东; 刘国栋; 魏富鹏; 彭志涛; 唐军; 刘楠

doi:10.3788/IRLA201847.1126005

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

连接向量特征匹配的暗场图像配准方法

黄卓, 陈凤东, 刘国栋, 魏富鹏, 彭志涛, 唐军, 刘楠

文章导航 > 红外与激光工程 > 2018 > 47(11): 1126005-1126005(6)

黄卓, 陈凤东, 刘国栋, 魏富鹏, 彭志涛, 唐军, 刘楠. 连接向量特征匹配的暗场图像配准方法[J]. 红外与激光工程, 2018, 47(11): 1126005-1126005(6). doi: 10.3788/IRLA201847.1126005

引用本文:

Huang Zhuo, Chen Fengdong, Liu Guodong, Wei Fupeng, Peng Zhitao, Tang Jun, Liu Nan. Dark field image registration based on connection vector feature matching method[J]. Infrared and Laser Engineering, 2018, 47(11): 1126005-1126005(6). doi: 10.3788/IRLA201847.1126005

Citation:

Huang Zhuo, Chen Fengdong, Liu Guodong, Wei Fupeng, Peng Zhitao, Tang Jun, Liu Nan. Dark field image registration based on connection vector feature matching method[J]. Infrared and Laser Engineering, 2018, 47(11): 1126005-1126005(6). doi: 10.3788/IRLA201847.1126005

连接向量特征匹配的暗场图像配准方法

doi: 10.3788/IRLA201847.1126005

1.
哈尔滨工业大学电气工程及自动化学院,黑龙江哈尔滨 150001;
2.
中国工程物理研究院激光聚变研究中心,四川绵阳 621900

基金项目:

国家自然科学基金（51275120，61275096）;航空科学基金（20120177004）;中央高校基本科研业务费专项资金（HIT.NSRIF.2013012）

详细信息

作者简介:
黄卓(1994-),女,硕士生,主要从事计算机视觉、图像处理方面的研究。Email:hithuangzhuo@163.com

中图分类号: TP391

Dark field image registration based on connection vector feature matching method

1.
School of Electrical Engineering and Automation,Harbin Institute of Technology,Harbin 150001,China;
2.
The Laser Fusion Research Center,China Academy of Engineering Physics,Mianyang 621900,China

摘要: 利用损伤点之间的位置关系，根据光学元件损伤暗场图像的特点设计了一种基于连接向量特征匹配的配准方法。该方法首先对基准图像及待配准图像分别进行图像预处理，提取损伤点轮廓的中心坐标作为损伤点的位置值。然后构建损伤点连接向量，求出主方向并计算主方向下的连接向量特征，使用连接向量匹配获得匹配点对，最后利用RANSAC算法对匹配点对进行仿射变换参数计算。该方法具有旋转不变性，尺度不变性以及较高的配准准确度。实验对比分析了该方法与SIFT算法的计算效率及配准精度，结果表明在暗场图像条件下文中方法更有效且为背景单一，灰度信息较少同时要求较高的配准速度的场景下的图像配准问题提供了解决方案。
- 图像配准 /
- 随机抽样一致性检验 /
- 仿射变换 /
- 暗场成像
Abstract: Based on the the positional relationship between damage points and characteristics of optical damage dark field image, a registration method using connection vector feature matching was designed. The method firstly preprocessed the reference image and the image to be registered, and then extracted the center coordinates of the contour of the damage point as the position value of the damage point. Then the connection vector of the damage points was constructed, the main direction was obtained and the connection vector feature in the main direction was calculated, the exact match was achieved through the fine matches. Finally, the affine transformation parameter was calculated by using the RANSAC algorithm. The method had rotational invariance, scale invariance and high registration accuracy. The computational efficiency and registration precision of this method and SIFT algorithm were compared and analyzed. The experimental results show the more effectiveness of the proposed method in dark field image.
- image registration /
- random sample and consensus /
- affine transformation /
- dark-field image

[1]	Nostrand M C, Carr C W, Liao Z M, et al. Tools for predicting optical damage on inertial confinement fusion-class laser systems[Z]. LLNL-TR-464836, 2011:5.
[2]	Feng Bo, Liu Bingguo, Chen Fengdong, et al. Final optics damage online inspection system for ICF[J]. Infrared and Laser Engineering, 2013, 42(9):2519-2524.(in Chinese)
[3]	Feng Bo, Chen Fengdong, Liu Bingguo, et al. Online inspection of final optics based on image mosaic for ICF system[J]. Optics and Precision Engineering, 2014, 22(3):555-561. (in Chinese)
[4]	Du Siliang, Yang Bo, Wang Mi, et al. Stellar image registration based on mutual information in object area[J]. Optics and Precision Engineering, 2016, 24(2):406-412. (in Chinese)
[5]	Chen liang, Zhou mengzhe, Chen he. A method of image registration combined by edge region and cross correlation[J]. Transactions of Beijing Institute of Technology, 2016, 36(3):320-325. (in Chinese)
[6]	Kisaki M, Yamamura Y, Kim H, et al. High speed image registration of head CT and MR images based on Levenberg-Marquardt algorithms[C]//International Symposium on Soft Computing and Intelligent Systems. IEEE, 2015:1481-1485.
[7]	Patel M S, Patel N M, Holia M S. Feature based multi-view image registration using SURF[C]//International Symposium on Advanced Computing and Communication, IEEE, 2016:213-218.
[8]	Zou X, Lin G, Tang Y, et al. Sub-pixel object-image registration using improved iterative closest point method[J]. Journal of Computer-Aided Design Computer Graphics, 2016.
[9]	Wang Zhiqiang, Cheng Hong, Yang Guang, et al. Fast target location method of global image registration[J].Infrared and Laser Engineering, 2015, 44(S1):225-229. (in Chinese)
[10]	Lowe D G. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004, 60(2):91-110.
[11]	Cao Yu, Feng Ying, Yang Yuntao, et al. Application of line estimation algorithmbased on RANSAC inroad points cloud Optimization[J]. Infrared and Laser Engineering, 2012, 41(11):3108-3112. (in Chinese)
[12]	Litman R, Korman S, Bronstein A, et al. Inverting RANSAC:Global model detection via inlier rate estimation[C]//IEEE Conference on Computer Vision and Pattern Recognition, IEEE Computer Society, 2015:5243-5251.

[1]	王中军, 晁艳锋. 采用SURF特征和局部互相关信息的图像配准算法 . 红外与激光工程, 2022, 51(6): 20210950-1-20210950-6. doi: 10.3788/IRLA20210950
[2]	赵慧洁, 张晓元, 贾国瑞, 邱显斐, 翟亮. 运动补偿下双通道星载高光谱成像仪图像配准 . 红外与激光工程, 2021, 50(3): 20211022-1-20211022-9. doi: 10.3788/IRLA20211022
[3]	姜宇, 李娜, 孟令杰, 蔡辉, 巩学美, 赵慧洁. 基于误差分析的岩芯高光谱数据几何校正方法 . 红外与激光工程, 2018, 47(5): 526004-0526004(8). doi: 10.3788/IRLA201847.0526004
[4]	荆滢, 齐乃新, 杨小冈, 卢瑞涛. 基于LK和FAST的时间序列图像快速配准算法 . 红外与激光工程, 2018, 47(11): 1126006-1126006(9). doi: 10.3788/IRLA201847.1126006
[5]	王志强, 程红, 谭海峰, 杨明权, 滕飞. 基于侧抑制竞争的航空遥感影像配准算法 . 红外与激光工程, 2018, 47(S1): 206-212. doi: 10.3788/IRLA201847.S126005
[6]	赖丽君, 徐智勇, 张栩铫. 应用于稳像系统中的改进梯度光流法 . 红外与激光工程, 2016, 45(4): 428004-0428004(7). doi: 10.3788/IRLA201645.0428004
[7]	王贵圆, 常本康, 富容国, 冯澍, 孙斌, 李英杰. 分光轴式多源图像融合系统近距离配准方法研究 . 红外与激光工程, 2016, 45(11): 1104003-1104003(7). doi: 10.3788/IRLA201645.1104003
[8]	李英杰, 张俊举, 常本康, 钱芸生, 刘磊. 远距离多波段红外图像融合系统及配准方法 . 红外与激光工程, 2016, 45(5): 526002-0526002(6). doi: 10.3788/IRLA201645.0526002
[9]	武奕楠, 李国宁, 张柯, 张宇, 金龙旭. 基于同名点追踪的空间相机成像拼接配准模型 . 红外与激光工程, 2016, 45(3): 326002-0326002(7). doi: 10.3788/IRLA201645.0326002
[10]	王志强, 程红, 杨桄, 李成, 吴迪. 全局图像配准的目标快速定位方法 . 红外与激光工程, 2015, 44(S1): 225-229.
[11]	张宏伟, 樊祥, 朱斌, 施展. 引入外点剔除机制的双波段红外图像的配准 . 红外与激光工程, 2015, 44(S1): 23-28.
[12]	韩艳丽, 刘峰. 基于三角形匹配的空间小目标检测算法 . 红外与激光工程, 2014, 43(9): 3134-3140.
[13]	刘晓诚, 薛模根, 黄勤超, 王峰. 基于矩阵恢复的红外偏振图像分区配准算法 . 红外与激光工程, 2014, 43(8): 2733-2739.
[14]	何林阳, 刘晶红, 李刚, 刘剑. 改进BRISK特征的快速图像配准算法 . 红外与激光工程, 2014, 43(8): 2722-2727.
[15]	王国刚, 史洪岩, 汪滢, 袁德成. 仿射不变子空间特征及其在图像匹配中的应用 . 红外与激光工程, 2014, 43(2): 659-664.
[16]	彭晨, 陈钱, 钱惟贤, 徐富元. 复杂地面场景下的红外运动目标跟踪 . 红外与激光工程, 2013, 42(6): 1410-1414.
[17]	吴明军, 许建铮, 周桢, 张亚涛. 针对运动摄像机的快速低存储开销运动目标检测算法 . 红外与激光工程, 2013, 42(8): 2275-2280.
[18]	吴泽鹏, 郭玲玲, 朱明超, 贾宏光, 宣明. 结合图像信息熵和特征点的图像配准方法 . 红外与激光工程, 2013, 42(10): 2846-2852.
[19]	杨新锋, 滕书华, 夏东. 基于空间变换迭代的SIFT特征图像匹配算法 . 红外与激光工程, 2013, 42(12): 3496-3501.
[20]	张德新, 马广富, 邵晓巍. 利用目标直线边缘交点的某型航侦CCD相机图像拼接 . 红外与激光工程, 2012, 41(1): 234-238.

点击查看大图

计量

文章访问数: 545
HTML全文浏览量: 129
PDF下载量: 51
被引次数: 0

连接向量特征匹配的暗场图像配准方法

doi: 10.3788/IRLA201847.1126005

1. 哈尔滨工业大学电气工程及自动化学院,黑龙江哈尔滨 150001;

2. 中国工程物理研究院激光聚变研究中心,四川绵阳 621900

作者简介:
黄卓(1994-),女,硕士生,主要从事计算机视觉、图像处理方面的研究。Email:hithuangzhuo@163.com

收稿日期: 2018-06-10
修回日期: 2018-07-20
刊出日期: 2018-11-25

基金项目:

国家自然科学基金（51275120，61275096）;航空科学基金（20120177004）;中央高校基本科研业务费专项资金（HIT.NSRIF.2013012）

中图分类号: TP391

关键词:

摘要: 利用损伤点之间的位置关系，根据光学元件损伤暗场图像的特点设计了一种基于连接向量特征匹配的配准方法。该方法首先对基准图像及待配准图像分别进行图像预处理，提取损伤点轮廓的中心坐标作为损伤点的位置值。然后构建损伤点连接向量，求出主方向并计算主方向下的连接向量特征，使用连接向量匹配获得匹配点对，最后利用RANSAC算法对匹配点对进行仿射变换参数计算。该方法具有旋转不变性，尺度不变性以及较高的配准准确度。实验对比分析了该方法与SIFT算法的计算效率及配准精度，结果表明在暗场图像条件下文中方法更有效且为背景单一，灰度信息较少同时要求较高的配准速度的场景下的图像配准问题提供了解决方案。

English Abstract

Dark field image registration based on connection vector feature matching method

1. School of Electrical Engineering and Automation,Harbin Institute of Technology,Harbin 150001,China;

2. The Laser Fusion Research Center,China Academy of Engineering Physics,Mianyang 621900,China

Received Date: 2018-06-10
Rev Recd Date: 2018-07-20
Publish Date: 2018-11-25

Keywords:

Abstract: Based on the the positional relationship between damage points and characteristics of optical damage dark field image, a registration method using connection vector feature matching was designed. The method firstly preprocessed the reference image and the image to be registered, and then extracted the center coordinates of the contour of the damage point as the position value of the damage point. Then the connection vector of the damage points was constructed, the main direction was obtained and the connection vector feature in the main direction was calculated, the exact match was achieved through the fine matches. Finally, the affine transformation parameter was calculated by using the RANSAC algorithm. The method had rotational invariance, scale invariance and high registration accuracy. The computational efficiency and registration precision of this method and SIFT algorithm were compared and analyzed. The experimental results show the more effectiveness of the proposed method in dark field image.