基于法线引导的激光中心线提取方法

刘天赐; 刘桂华; 胡莉; 黄会明

doi:10.5768/JAO202344.0107003

基于法线引导的激光中心线提取方法

doi: 10.5768/JAO202344.0107003

西南科技大学信息工程学院，四川绵阳 621010

基金项目: 四川省科技厅重点研发项目（2021YFG0380）；四川省卫生和计划生育科研课题(17PJ207)

详细信息

作者简介:
刘天赐（1997—），男，硕士研究生，主要从事三维重建、图像处理方面的研究。E-mail：1057568474@qq.com

通讯作者:
刘桂华（1972—），女，博士，教授，主要从事图像处理、机器学习方面的研究。E-mail：564879375@qq.com

中图分类号: TN209;TP391
计量
- 文章访问数: 82
- HTML全文浏览量: 33
- PDF下载量: 9
- 被引次数: 0
出版历程
- 收稿日期: 2022-03-30
- 修回日期: 2022-05-25
- 网络出版日期: 2022-08-04
- 刊出日期: 2023-01-17

Laser center-line extraction method based on normal guidance

School of Information Engineering, Southwest University of Science and Technology, Mianyang 621010, China

摘要

摘要: 在线结构光三维测量系统中，高精度激光条纹中心线提取是提高测量精度的关键。针对现有激光中心线存在提取精度不高、保留细节差等问题，提出了一种基于法线引导的激光中心线提取算法。该算法具体实现步骤为：首先，对图像进行预处理，结合边缘检测和几何中心法对激光线初步提取；然后，用主成分分析法( principal component analysis，PCA)求取其法线，在激光中心点处划分角度八邻域，通过法线角度引导搜寻有效点集；最后，利用灰度重心法对点集进行亚像素提取。实验结果表明：该算法均方根误差与灰度重心法相比提高了0.2339 像素，比Steger算法、方向模板法更好地保留了光条细节，可以更精确地提取光条中心，达到亚像素级的精度。
- 线结构光 /
- 法线引导 /
- 几何中心 /
- 主成分分析 /
- 八邻域 /
- 灰度重心
Abstract: In the line structured light three-dimensional measurement system, the high-precision laser stripe center-line extraction is the key to improve the measurement accuracy. Aiming at the existing laser center-line problems such as low extraction accuracy and poor retention of details, a laser center-line extraction algorithm based on normal guidance was proposed. The specific implementation steps of the algorithm were as follows: Firstly, the image was pre-processed, and the laser line was preliminarily extracted by combining edge detection and geometric center method. Then, the principal component analysis (PCA) was used to obtain its normal line, divided the angle eight neighborhood at the laser center point, and searched for the effective point set through the normal angle guidance. Finally, the gray centroid method was used to extract the sub-pixels from the point set. The experimental results show that the root mean square error of the algorithm is improved by 0.233 9 pixel compared with that of the gray centroid method, which can retain the light strip details better than the Steger algorithm and the directional template method, and can extract the light strip center more accurately and achieve the sub-pixel level accuracy.
- line structured light /
- normal guidance /
- geometric center /
- principal component analysis /
- eight neighborhood /
- gray centroid

HTML全文

图 1 修补断线过程示意图

Fig. 1 Schematic diagram of repairing broken wires

下载: 全尺寸图片幻灯片

图 2 几何中心法求取中心线

Fig. 2 Center line obtained by geometric center method

下载: 全尺寸图片幻灯片

图 3 法线方向示意图

Fig. 3 Schematic diagram of normal direction

下载: 全尺寸图片幻灯片

图 4 激光条纹灰度分布特性曲线图

Fig. 4 Curves of gray distribution properties of laser stripes

下载: 全尺寸图片幻灯片

图 5 8角度划分以及寻找有效点示意图

Fig. 5 Schematic diagram of eight-angle division and searching for effective points

下载: 全尺寸图片幻灯片

图 6 有效点集示意图

Fig. 6 Schematic diagram of effective point set

下载: 全尺寸图片幻灯片

图 7 激光原图及算法提取效果对比图

Fig. 7 Comparison of original laser image and extraction effect of algorithm

下载: 全尺寸图片幻灯片

图 8 各种算法提取效果图

Fig. 8 Effect diagram extracted by various algorithms

下载: 全尺寸图片幻灯片

图 9 激光线提取示意图

Fig. 9 Schematic diagram of laser line extraction

下载: 全尺寸图片幻灯片

表 1 光条中心提取结果及拟合直线偏离误差

Table 1 Extraction results of light strip center and deviation error of fitting straight line

激光编号	拟合方法	拟合直线	像素平均偏差/像素
①	Steger算法	y=0.340 6x+225.068 2	0.251 73
	灰度重心法	y=0.340 2x+223.746 3	0.089 17
	本文算法	y=0.340 2x+223.757 7	0.057 77
②	Steger算法	y=0.340 6x+256.920 6	0.258 59
	灰度重心法	y=0.340 3x+255.531 3	0.096 13
	本文算法	y=0.340 4x+255.543 9	0.060 73
③	Steger算法	y=0.341 4x+384.114 8	0.284 14
	灰度重心法	y=0.341 0x+382.974 8	0.091 04
	本文算法	y=0.340 9x+383.008 0	0.055 32

下载: 导出CSV

表 2 各种算法的均方根误差

Table 2 Root mean square error of various algorithms

提取方法	RMSE/像素
方向模板法	0.351 2
Steger算法	0.355 3
灰度重心法	0.595 9
文献[13]	0.389 2
本文算法	0.362 0

下载: 导出CSV

参考文献(17)

[1]	WANG S C, HAN Q, WANG H, et al. Extraction method of laser stripe center of rail profile in drivingenvironment[J]. Acta Optica Sinica,2019,39(2):167-176.
[2]	HE Leiying, WU Shanshan, WU Chuanyu. Robust laser stripe extraction for three-dimensional reconstruction based on a cross-structured light sensor[J]. Applied Optics,2017,56(4):823-832. doi: 10.1364/AO.56.000823
[3]	LUO Wenting, WANG K C P, LI Lin, et al. Surface drainage evaluation for rigid pavements using an inertial measurement unit and 1 mm three-dimensional texture data[J]. Transportation Research Record:Journal of the Transportation Research Board,2014,2457(1):121-128. doi: 10.3141/2457-13
[4]	LI Lin, WANG K C P, LI Q. Geometric texture indicators for safety on AC pavements with 1 mm 3D laser texture data[J]. International Journal of Pavement Research and Technology,2016,9(1):49-62. doi: 10.1016/j.ijprt.2016.01.004
[5]	刘智, 翟林培, 郝志航. 互补金属氧化物半导体图像传感器亚像元细分精度实验研究[J]. 中国激光,2007,34(1):118-124. doi: 10.3321/j.issn:0258-7025.2007.01.022 LIU Zhi, ZHAI Linpei, HAO Zhihang. Sub-pixel measurement accuracy experiment of complementary metal oxide semiconductor imager[J]. Chinese Journal of Lasers,2007,34(1):118-124. doi: 10.3321/j.issn:0258-7025.2007.01.022
[6]	STEGER C. An unbiased detector of curvilinear structures[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1998,20(2):113-125. doi: 10.1109/34.659930
[7]	潘硕, 邵新杰, 李晓磊, 等. 基于块匹配的激光条纹亚像素中心提取[J]. 激光与红外,2021,51(6):814-819. doi: 10.3969/j.issn.1001-5078.2021.06.020 PAN Shuo, SHAO Xinjie, LI Xiaolei, et al. Sub-pixel extraction of laser stripe center based on block-matching[J]. Laser & Infrared,2021,51(6):814-819. doi: 10.3969/j.issn.1001-5078.2021.06.020
[8]	王利, 陈念年, 巫玲, 等. 高噪声背景下激光条纹亚像素中心的提取[J]. 应用光学,2016,37(2):321-326. doi: 10.5768/JAO201637.0207004 WANG Li, CHEN Niannian, WU Ling, et al. Extraction of laser stripe sub-pixel center in high-noise background[J]. Journal of Applied Optics,2016,37(2):321-326. doi: 10.5768/JAO201637.0207004
[9]	胡斌, 李德华, 金刚, 等. 基于方向模板的结构光条纹中心检测方法[J]. 计算机工程与应用,2002,38(11):59-60. doi: 10.3321/j.issn:1002-8331.2002.11.021 HU Bin, LI Dehua, JIN Gang, et al. New method for obtaining the center of structured light stripe by directi on template[J]. Computer Engineering and Applications,2002,38(11):59-60. doi: 10.3321/j.issn:1002-8331.2002.11.021
[10]	蔡怀宇, 冯召东, 黄战华. 基于主成分分析的结构光条纹中心提取方法[J]. 中国激光,2015,42(3):278-283. CAI Huaiyu, FENG Zhaodong, HUANG Zhanhua. Centerline extraction of structured light stripe based on principal component analysis[J]. Chinese Journal of Lasers,2015,42(3):278-283.
[11]	周渊, 孟祥群, 江登表, 等. 复杂干扰情况下的结构光条纹中心提取方法[J]. 中国激光,2020,47(12):172-180. ZHOU Yuan, MENG Xiangqun, JIANG Dengbiao, et al. Centerline extraction of structured light stripe under complex interference[J]. Chinese Journal of Lasers,2020,47(12):172-180.
[12]	金俊, 李德华, 李和平. 结构光三维获取系统条纹中心线检测[J]. 计算机工程与应用,2006,42(4):42-44. doi: 10.3321/j.issn:1002-8331.2006.04.013 JIN Jun, LI Dehua, LI Heping. New method for obtaining the center of structured light stripe[J]. Computer Engineering and Applications,2006,42(4):42-44. doi: 10.3321/j.issn:1002-8331.2006.04.013
[13]	胡杨, 方素平. 线结构光条纹中心提取方法[J]. 激光与光电子学进展,2021,58(1):196-200. HU Yang, FANG Suping. Extraction method of light stripe center of linear structure[J]. Laser & Optoelectronics Progress,2021,58(1):196-200.
[14]	OTSU N. A Threshold selection method from gray-level histograms[J]. IEEE Transactions on Systems Man & Cybernetics,2007,9(1):62-66.
[15]	何新英, 王家忠, 孙晨霞, 等. 基于数学形态学和Canny算子的边缘提取方法[J]. 计算机应用,2008,28(2):477-478. doi: 10.3724/SP.J.1087.2008.00477 HE Xinying, WANG Jiazhong, SUN Chenxia, et al. Edge detection method based on mathematical morphology and canny algorithm[J]. Journal of Computer Applications,2008,28(2):477-478. doi: 10.3724/SP.J.1087.2008.00477
[16]	曾建新. 激光双目三维成像关键技术研究[D]. 成都: 电子科技大学, 2021. ZENG Jianxin. Research on key technologies of laser binocular 3D imaging[D]. Chengdu: University of Electronic Science and Technology of China, 2021.
[17]	杨毅, 闫兵, 董大伟, 等. 基于二次平滑算法的线结构光中心线提取方法[J]. 激光与光电子学进展,2020,57(10):309-315. YANG Yi, YAN Bing, DONG Dawei, et al. Method for extracting the centerline of line structured light based on quadratic smoothing algorithm[J]. Laser & Optoelectronics Progress,2020,57(10):309-315.