复杂背景下激光条纹中心亚像素提取方法

甘宏, 张超, 李林, 等. 复杂背景下激光条纹中心亚像素提取方法[J]. 光电工程, 2019, 46(2): 180457. doi: 10.12086/oee.2019.180457
引用本文: 甘宏, 张超, 李林, 等. 复杂背景下激光条纹中心亚像素提取方法[J]. 光电工程, 2019, 46(2): 180457. doi: 10.12086/oee.2019.180457
Gan Hong, Zhang Chao, Li Lin, et al. Sub-pixel extraction of laser stripe in complex background[J]. Opto-Electronic Engineering, 2019, 46(2): 180457. doi: 10.12086/oee.2019.180457
Citation: Gan Hong, Zhang Chao, Li Lin, et al. Sub-pixel extraction of laser stripe in complex background[J]. Opto-Electronic Engineering, 2019, 46(2): 180457. doi: 10.12086/oee.2019.180457

复杂背景下激光条纹中心亚像素提取方法

  • 基金项目:
    国家重点研发计划项目(2018YFB120181);国家自然科学基金项目(51608123);福建省自然科学基金面上项目(2017J01682, 2017J01475)
详细信息
    作者简介:
    通讯作者: 罗文婷(1983-),女,博士,讲师,主要从事交通工程、道路病害检测的研究。E-mail: luowenting531@gmail.com
  • 中图分类号: TP391.41; TB872

Sub-pixel extraction of laser stripe in complex background

  • Fund Project: Supported by the National Key Research and Development Program of China (2018YFB120181), National Natural Science Foundation Projects (51608123), and Fujian Natural Science Foundation (2017J01682, 2017J01475)
More Information
  • 针对激光条纹中心提取的复杂背景及噪声干扰问题,提出一种自适应双阈值分割方法及改进灰度权重模型。首先对光条图像的特征及噪声来源进行分析,并采用双边滤波进行图像预处理;然后通过图像灰度直方图计算双阈值,并采用双阈值分区域处理获取二值化图像,从而提取初始条纹中心及条纹宽度;最后利用改进灰度权重模型提取激光条纹亚像素中心。将双阈值分割方法及改进灰度权重模型与传统算法进行对比,结果表明:双阈值分割方法较极值法、大津法能更加准确完整地提取激光条纹区域。对比提取的亚像素中心残差值,改进灰度权重模型(0.23)较灰度重心法(0.71)、极值法(0.86)、高斯拟合法(0.86)具有更优结果。本文方法能有效避免复杂背景以及激光条纹法线方向噪声的影响,提高中心定位精度,具有较高的抗噪能力与稳健性,适用于复杂背景下快速、精确的光条中心提取。

  • Overview: Laser stripe center extraction is the key step of a structured light vision system, which determines the stability, real-time and accuracy of the system. The results of laser center extraction are affected by many factors, such as changing width of the laser stripe, the complex measuring environment, the optical properties of the measured surface, etc. In current studies, there are many methods presented to extract the laser stripe centre. The traditional extremum method and the geometric center method are low precise and insufficient in robustness. Direction template method can effectively eliminate noise, but it is not effective in extracting laser stripes with complex shapes and varying widths. Furthermore, all the gray information in the laser stripe are considered in the gray-gravity method. Although the accuracy of this method is much higher than other traditional methods, it is easily interfered by the noise with high frequency. Gaussian fitting is a popular method to detect the laser line center position, which eliminates the most noise of laser stripe. The Steger algorithm is precise in positioning and robust to noise. However, this method is complicated and cannot process in real-time.

    The improved gray weight model is proposed to make a balance between the robustness, calculation speeds and the intensive computation. First step is reducing the image noise. Secondly, the laser stripe and the complex background environment need to be divided accurately. Finally, a laser centre extraction method is proposed.

    The raw image contains amount of image noise, which affects the accuracy of the measurement of the structured light vision system. Firstly, the bilateral filter is applied to remove noise of raw images. Subsequently, the gray histogram of the laser image and the double threshold are computed. Based on the sub-regional processing, the initial stripe center and the stripe width in binary images are obtained. Then, a smoothing distance algorithm is used to obtain a continuous centre curve. Finally, sub-pixel center of the strip is extracted based on the proposed model. Double threshold segmentation method and the improved gray weight model are compared with traditional algorithms.

    The results show that adaptive double threshold method is more accurate on the laser stripe region extraction than the extreme value method and the Otsu method. Comparing with the residual value of sub-pixel center, the improved gray weight model (0.23) has the best result, and follows by gray-gravity method (0.71), extreme value method (0.86), and Gaussian fitting method (0.86). The algorithms proposed in this study avoid the impacts of complex background and laser stripe noise, increase the accuracy of the laser stripe center positioning and extract the stripe center extraction fast and accurately in complex backgrounds.

  • 加载中
  • 图 1  激光条纹原图及处理图像。(a)激光条纹原始图;(b)高斯滤波处理结果图;(c)双边滤波处理结果图

    Figure 1.  Original laser stripe image and processing image. (a) Original laser stripe image; (b) Result of bilateral filter; (c) Result of Gaussian filter

    图 2  不同阈值分割方法处理结果对比。(a)极值法处理结果图;(b)大津法处理结果图;(c)双阈值法处理结果图;(d)光条中心提取结果图

    Figure 2.  Comparison of different threshold segmentation methods. (a) Image binary based on extreme value threshold; (b) Image binary based on OTSU; (c) Image binary based on double threshold; (d) Result of laser stripe center extraction

    图 3  不同条件下采集的激光条纹图像。(a)亮条件下激光条纹图像;(b)暗条件下激光条纹图像

    Figure 3.  Laser stripe image captured by different conditions. (a) Laser tripe image under bright condition; (b) Laser tripe image under dark condition

    图 4  亮条件下残差对比图

    Figure 4.  Comparison of residuals under bright condition

    图 5  暗条件下残差对比图

    Figure 5.  Comparison of residuals under dark condition

    表 1  不同条件下各方法的光条中心拟合直线、相关系数及用时

    Table 1.  Different method fitting line, correlation of stripe center and run time under different conditions

    拟合方法 暗条件 亮条件 R2(亮) R2(暗) t/ms
    极值法 Y=-0.0139X+211.94 Y=-0.0117X+210.63 0.8465 0.8630 1.5
    高斯拟合法 Y=-0.0125X+211.11 Y=-0.0133X+210.39 0.8580 0.8630 69
    灰度重心法 Y=-0.0090X+212.42 Y=-0.0086X+211.63 0.8150 0.9662 3
    本文方法 Y=-0.0079X+211.85 Y=-0.0076X+210.54 0.9797 0.9977 10
    下载: 导出CSV
  • [1]

    陈远, 冯华君, 徐之海, 等.三维激光测量的投影算法[J].光电工程, 2007, 34(6):40-43, 48. doi: 10.3969/j.issn.1003-501X.2007.06.009

    Chen Y, Feng H J, Xu Z H, et al.Projection algorithm for laser three dimensional measurement[J].Opto-Electronic Engineering, 2007, 34(6):40-43, 48. doi: 10.3969/j.issn.1003-501X.2007.06.009

    [2]

    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. http://d.old.wanfangdata.com.cn/Periodical/yygx201602030

    Wang L, Chen N N, Wu L, et al. Extraction of laser stripe sub-pixel center in high-noise background[J].Journal of Applied Optics, 2016, 37(2):321-326. http://d.old.wanfangdata.com.cn/Periodical/yygx201602030

    [3]

    Li Q G, Yao M, Yao X, et al. A real-time 3D scanning system for pavement distortion inspection[J].Measurement Science and Technology, 2010, 21(1): 015702. doi: 10.1088/0957-0233/21/1/015702

    [4]

    Li L, Wang K C P, Li Q J, et al.Automated runway groove measurement and evaluation[J].KSCE Journal of Civil Engineering, 2017, 21(3): 758-765. doi: 10.1007/s12205-016-0789-6

    [5]

    Luo W T, Li L.Automatic geometry measurement for curved ramps using inertial measurement unit and 3D LiDAR system[J].Automation in Construction, 2018, 94:214-232. doi: 10.1016/j.autcon.2018.07.004

    [6]

    Li L, Luo W T, Wang K C P.Lane marking detection and reconstruction with line-scan imaging data[J].Sensors, 2018, 18(5):1635. doi: 10.3390/s18051635

    [7]

    He L Y, Wu S S, Wu C Y.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

    [8]

    Luo W T, Wang K C P, Li L, et al.Surface drainage evaluation for rigid pavements using an Inertial Measurement Unit and 1-mm 3D texture data[J].Transportation Research Record:Journal of the Transportation Research Board, 2014, 2457: 121-128. doi: 10.3141/2457-13

    [9]

    Li L, Wang K C P, Li Q J.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

    [10]

    Kokku R, Brooksby G.Improving 3D surface measurement accuracy on metallic surfaces[J].Proceedings of SPIE, 2005, 5856:618-624. doi: 10.1117/12.612243

    [11]

    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, 109. doi: 10.3321/j.issn:1002-8331.2002.11.021

    Hu B, Li D H, Jin G, et al. New method for obtaining the center of structured light stripe by direction template[J].Computer Engineering and Applications, 2002, 38(11):59-60, 109. doi: 10.3321/j.issn:1002-8331.2002.11.021

    [12]

    王泽浩, 张中炜.自适应方向模板线结构光条纹中心提取方法[J].激光杂志, 2017, 38(1): 60-64. http://d.old.wanfangdata.com.cn/Periodical/jgzz201701014

    Wang Z H, Zhang Z W.Adaptive direction template method to extract the center of structured light[J].Laser Journal, 2017, 38(1):60-64. http://d.old.wanfangdata.com.cn/Periodical/jgzz201701014

    [13]

    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 Z, Zhai L P, Hao Z H.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

    [14]

    Li Y H, Zhou J B, Huang F S, et al.Sub-pixel extraction of laser stripe center using an improved gray-gravity method[J].Sensors, 2017, 17(4):814. doi: 10.3390/s17040814

    [15]

    熊会元, 宗志坚, 高群, 等.精确提取线结构光条纹中心方法[J].计算机工程与应用, 2009, 45(10):235-237. doi: 10.3778/j.issn.1002-8331.2009.10.070

    Xiong H Y, Zong Z J, Gao Q, et al.Precise method for extracting center of structured light stripe[J].Computer Engineering and Applications, 2009, 45(10):235-237. doi: 10.3778/j.issn.1002-8331.2009.10.070

    [16]

    刘振, 李声, 冯常.基于互相关算法的激光条纹中心提取[J].中国激光, 2013, 40(5):0508004. http://d.old.wanfangdata.com.cn/Conference/8548497

    Liu Z, Li S, Feng C.Laser stripe center extraction based on cross-correlation algorithm[J].Chinese Journal of Lasers, 2013, 40(5):0508004. http://d.old.wanfangdata.com.cn/Conference/8548497

    [17]

    解则晓, 张成国, 张国雄.基于B样条迭代法的激光光条噪声去除技术研究[J].光学技术, 2005, 31(3):430-433. doi: 10.3321/j.issn:1002-1582.2005.03.009

    Xie Z X, Zhang C G, Zhang G X.Research on removing the noise on the laser stripe based on the iterative fitting of B-spline[J].Optical Technique, 2005, 31(3):430-433. doi: 10.3321/j.issn:1002-1582.2005.03.009

    [18]

    Xu G, Zhang X Y, Li X T, et al. Timed evaluation of the center extraction of a moving laser stripe on a vehicle body using the Sigmoid-Gaussian function and a tracking method[J].Optik, 2017, 130:1454-1461. doi: 10.1016/j.ijleo.2016.11.146

    [19]

    高世一, 杨凯珍.变边限高斯拟合提取激光条纹中心线方法的研究[J].仪器仪表学报, 2011, 32(5):1132-1137. http://d.old.wanfangdata.com.cn/Periodical/yqyb201105028

    Gao S Y, Yang K Z.Research on central position extraction of laser strip based on varied-boundary Gaussian fitting[J].Chinese Journal of Scientific Instrument, 2011, 32(5): 1132-1137. http://d.old.wanfangdata.com.cn/Periodical/yqyb201105028

    [20]

    Zhang Y, Liu W, Li X D, et al. Accuracy improvement in laser stripe extraction for large-scale triangulation scanning measurement system[J].Optical Engineering, 2015, 54(10):105108. doi: 10.1117/1.OE.54.10.105108

    [21]

    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

    [22]

    Tomasi C, Manduchi R.Bilateral filtering for gray and color images[C]//Proceedings of the 6th International Conference on Computer Vision, Bombay, India, 1998: 839-846.

    [23]

    钟权, 周进, 吴钦章, 等.基于Hough变换和边缘灰度直方图的直线跟踪算法[J].光电工程, 2014, 41(3):89-94. doi: 10.3969/j.issn.1003-501X.2014.03.014

    Zhong Q, Zhou J, Wu Q Z, et al. A method of line tracking based on Hough transforms and edge histogram[J].Opto-Electronic Engineering, 2014, 41(3):89-94. doi: 10.3969/j.issn.1003-501X.2014.03.014

    [24]

    马增强, 宋子彬, 王永胜.基于激光线的轮轨冲角检测新方法[J].光电工程, 2017, 44(8):818-825. doi: 10.3969/j.issn.1003-501X.2017.08.009

    Ma Z Q, Song Z B, Wang Y S.A method for detecting the wheel rail attack angle based on laser line detection[J].Opto-Electronic Engineering, 2017, 44(8): 818-825. doi: 10.3969/j.issn.1003-501X.2017.08.009

    [25]

    张槐祥, 刘怀广, 杨茂麟, 等.复杂光照环境下的线材缺陷自适应分割方法[J].光电工程, 2014, 41(2):40-46. doi: 10.3969/j.issn.1003-501X.2014.02.007

    Zhang H X, Liu H G, Yang M L, et al.A self-adaptive segmentation method of wire defects under complex illumination[J].Opto-Electronic Engineering, 2014, 41(2):40-46. doi: 10.3969/j.issn.1003-501X.2014.02.007

    [26]

    黎明, 冯华君, 徐之海, 等.利用光强信息的结构光图像轮廓提取修正方法[J].光电工程, 2005, 32(2):30-32. doi: 10.3969/j.issn.1003-501X.2005.02.009

    Li M, Feng H J, Xu Z H, et al. Profile extraction of structural light image and its correction based on light intensity[J].Opto-Electronic Engineering, 2005, 32(2):30-32. doi: 10.3969/j.issn.1003-501X.2005.02.009

    [27]

    杨宪铭, 贺俊吉, 张广军, 等.圆结构光光条中心亚像素级提取方法[J].光电工程, 2004, 31(4):46-49. doi: 10.3969/j.issn.1003-501X.2004.04.013

    Yang X M, He J J, Zhang G J, et al.A method of sub-pixel extraction from circular structured light stripes center[J].Opto-Electronic Engineering, 2004, 31(4):46-49. doi: 10.3969/j.issn.1003-501X.2004.04.013

    [28]

    Otsu N.A threshold selection method from gray-level histograms[J].Automatica, 1975, 11(285-296):23-27. http://cn.bing.com/academic/profile?id=dc5dbe2d8aeed5d9a6ffcfc529aeb02c&encoded=0&v=paper_preview&mkt=zh-cn

    [29]

    Sun Q C, Liu R Y, Yu F H.An extraction method of laser stripe centre based on Legendre moment[J].Optik, 2016, 127(2):912-915. doi: 10.1016/j.ijleo.2015.10.196

    [30]

    张瑞瑛, 周萍, 冯煦, 等.大视场下线结构光光条中心的快速提取[J].应用光学, 2010, 31(3):432-436. doi: 10.3969/j.issn.1002-2082.2010.03.019

    Zhang R Y, Zhou P, Feng X, et al.Rapid extraction of line-structured light stripe in large field of view[J].Journal of Applied Optics, 2010, 31(3):432-436. doi: 10.3969/j.issn.1002-2082.2010.03.019

  • 加载中

(5)

(1)

计量
  • 文章访问数:  10598
  • PDF下载数:  2958
  • 施引文献:  0
出版历程
收稿日期:  2018-09-05
修回日期:  2018-11-06
刊出日期:  2019-02-18

目录

/

返回文章
返回