基于曲率相似性的路面连续纵长裂缝匹配方法

陈实; 黄玉春

doi:10.3963/j.jssn.1674-4861.2022.04.013

基于曲率相似性的路面连续纵长裂缝匹配方法

doi: 10.3963/j.jssn.1674-4861.2022.04.013

陈实,
黄玉春^,

武汉大学遥感信息工程学院武汉 430079

基金项目:

国家自然科学基金项目 41671419

详细信息

作者简介:
陈实（1996—），硕士研究生. 研究方向：交通遥感、移动测量. E-mail：2550686663@qq.com

通讯作者:
黄玉春（1977—），博士，副教授. 研究方向：交通遥感、移动测量. E-mail：hycwhu@whu.edu.cn

中图分类号: U491.5+4
计量
- 文章访问数: 667
- HTML全文浏览量: 223
- PDF下载量: 19
- 被引次数: 0
出版历程
- 收稿日期: 2022-04-21
- 网络出版日期: 2022-09-17

A Matching Method for Longitudinal Cracks Based on Curvature Similarity

CHEN Shi,
HUANG Yuchun^,

School of Remote Sensing and Information Engineering, Wuhan University, Wuhan 430079, China

摘要

摘要: 车载相机拍摄得到的路面裂缝形状分布随机，且由于视场角有限每次只能拍摄到道路上纵向长裂缝的一部分，导致纵长裂缝检测不完整。利用逆透视变换方法将车载相机采集的道路前方倾斜图像转化成正射图像，以去除纵长裂缝图像的透视变形；采用深度学习中的语义分割网络Deeplab V3+实现裂缝像素的提取；在此基础上，提出基于曲率相似性的由粗到精的两阶段路面连续纵长裂缝匹配方法。将待匹配的裂缝曲线分割为一连串相互重叠的子曲线序列，相互匹配的子曲线即为裂缝曲线相匹配的部分；利用曲率将子曲线局部形状与走势的特征表达为描述符，使用Kd-tree最邻近匹配算法对曲线描述符进行快速粗匹配。根据连续2张道路图像中纵长裂缝在空间位置分布上延续的特征，在裂缝曲线分割成子曲线时添加约束条件，前1张图像中裂缝曲线的起点和后1张图像中裂缝曲线的终点分别作为各自子曲线的1个端点；在粗匹配结果的基础上，逐步缩小分割曲线的间隔，迭代提高子曲线描述符间的归一化互相关系数，直至其大于等于阈值或者迭代次数超出最大迭代次数，实现对粗匹配结果的精调整。为验证算法精度，以武汉大学校园内路面不同类型的连续纵长裂缝为对象开展实验，匹配结果误差最小为0.688像素，精调整的误差比粗匹配平均减小24.19%。为进一步验证噪声下干扰的稳定性，仿真环境下增加了裂纹像素噪声；当高斯噪声的标准差从0增大到2像素时，匹配结果误差仅增大了1.083像素。将所提方法与SIFT算法进行对比，10组实验中，所提方法都能匹配成功；而SIFT算法在其中2组实验中匹配结果完全错误，表明所提算法有较好稳定性。
- 路面养护 /
- 纵长裂缝 /
- 透视变换 /
- 曲线匹配 /
- 曲率相似性
Abstract: Pavement cracks captured byon-board cameras are distributed randomly in shapes, and only a part of the longitudinal cracks on the roads can be captured each timedue to the limited field of view, resulting in incomplete detection of longitudinal cracks. The imagesacquired by the on-board cameras are transformed from oblique images intoorthographic images by using the inverse perspective transformation method, thus the perspective distortionof the longitudinal cracks are corrected. Then a deep learning based semantic segmentation network, Deeplab V3+, is used to extract the pixels of cracks. Based on curvature similarity, a two-stage method from coarse to fineis proposed for matching longitudinal cracks.The crack curve to be matched is divided into a sequence of overlapping sub-curves, which are characterized by descriptor of curvature, and the matched sub-curves are the matched parts of cracks. The curvature is used to express the local shape and trend features of sub-curvesas descriptors, then the Kd-tree nearest neighbor matching algorithm is used to perform coarse and fast matching of thedescriptors. According to the spatial distribution of longitudinal cracks in two consecutive road images, constraints are added when the crack curves are divided into sub-curves, the starting point of the crack curve in previousimage and the ending point in the next image areused asterminus of each respective sub-curve. Based on the results of coarse matching, the interval of segmentation curves is gradually reduced, and the normalized cross-correlation coefficient is iteratively improved until it is greater than or equal to the threshold or the number of iterations exceeds the maximum value to achieve fine adjustment of the results of coarse matching. To verify the accuracy of the algorithm, a case study is carried out with different types of continuous and longitudinal cracks on the campus roads of Wuhan University.The minimum error of the matching results can reach 0.688 pixels. Compared with the coarse matching, the error after fine adjustmentreduces by 24.19% on average. In order to further verify the stability of the algorithm under noise, crack pixel noise is added to the simulation environment.When the standard deviation of Gaussian noise increases from 0 to 2 pixels, the error of the matching results increases by only 1.083 pixels. Compared with the SIFT algorithm, the proposed method can achieve successful matching in all ten groups of experiments, while the matching results of the SIFT algorithm completely fails in two groups. It indicates that the algorithm proposed has better stability under normal and noise environment.
- pavement maintenance /
- longitudinal crack /
- perspective transformation /
- curve matching /
- curvature similarity

HTML全文

图 1 相机坐标系与世界坐标系

Figure 1. Camera coordinate system and world coordinate system

下载: 全尺寸图片幻灯片

图 2 曲线曲率特征

Figure 2. Curvature characteristics of curve

下载: 全尺寸图片幻灯片

图 3 重采样方法

Figure 3. Resampling method

下载: 全尺寸图片幻灯片

图 4 曲线匹配示意图

Figure 4. Diagram of curve matching

下载: 全尺寸图片幻灯片

图 5 基于K-d tree最邻近裂缝曲线粗匹配算法

Figure 5. Crack curve nearest matching algorithm based on K-d tree

下载: 全尺寸图片幻灯片

图 6 精调整起始位置流程

Figure 6. Process of fine adjust starting position

下载: 全尺寸图片幻灯片

图 7 粗匹配与精调整裂缝匹配误差

Figure 7. Matching error of coarse matching and fine adjustment

下载: 全尺寸图片幻灯片

图 8 简单裂缝实验结果

Figure 8. Results of simple crack

下载: 全尺寸图片幻灯片

图 9 复杂裂缝实验结果

Figure 9. Results of complex crack

下载: 全尺寸图片幻灯片

图 10 仿真实验曲线匹配误差

Figure 10. Error of curve matching in simulation experiment

下载: 全尺寸图片幻灯片

图 11 仿真实验结果

Figure 11. Results of simulation experimen

下载: 全尺寸图片幻灯片

图 12 SIFT算法匹配结果

Figure 12. Matching results of SIFT algorithm

下载: 全尺寸图片幻灯片

参考文献(22)

[1]	潘一凡, 张显峰, 童庆禧, 等. 公路路面质量遥感监测研究进展[J]. 遥感学报, 2017, 21(5): 796-811. https://www.cnki.com.cn/Article/CJFDTOTAL-YGXB201705014.htm PAN Y F, ZHANG X F, TONG Q X, et al. Research progress in remote sensing monitoring of highway pavement quality[J]. National Remote Sensing Bulletin, 2017, 21(5): 796-811. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YGXB201705014.htm
[2]	WOLFSON H J. On curve matching[J]. IEEE Transactions on PatternAnalysis & Machine Intelligence, 1990, 12(5): 483-489. doi: 10.1109/34.55108
[3]	KONG W, KIMIA B B. On solving 2D and 3D puzzles using curve matching[C]. IEEE Computer Society Conference on Computer Vision & Pattern Recognition, Kauai, USA: IEEE, 2001.
[4]	CUI M, FEMIANI J, HU J, et al. Curve matching for open 2D curves[J]. Pattern Recognition Letters, 2009, 30(1): 1-10. doi: 10.1016/j.patrec.2008.08.013
[5]	PETRAKIS E, DIPLAROS A, MILIOS E. Matching and retrieval of distorted and occluded shapes using dynamic programming[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(11): 1501-1516. doi: 10.1109/TPAMI.2002.1046166
[6]	MAI F, CHANG C Q, HUNG Y S. Affine-invariant shape matching and recognition under partial occlusion[C]. 17th International Conference on Image Processing, Hong Kong, China: IEEE, 2010.
[7]	ALAHI A, ORTIZ R, VANDERGHEYNST P. Freak: Fast retina keypoint[C]. IEEE Conference on Computer Vision & Pattern Recognition, Providence, USA: IEEE, 2012.
[8]	杨宇, 赵成星, 张晓玲. 基于SURF和改进RANSAC的图像拼接方法[J]. 激光杂志, 2021, 42(4): 105-108. https://www.cnki.com.cn/Article/CJFDTOTAL-JGZZ202104021.htm YANG Y, ZHAO C X, ZHANG X L. Image stitching method based on SURF and improved RANSAC[J]. Laser Journal, 2021, 42(4): 105-108. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JGZZ202104021.htm
[9]	DAVID L. Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision, 2004, 60(2): 91-110. doi: 10.1023/B:VISI.0000029664.99615.94
[10]	BAY H, ESS A, TUYTELAARS T, et al. Speeded-up robust features(SURF)[J]. Computer Vision and Image Understanding, 2008, 110(3): 346-359. doi: 10.1016/j.cviu.2007.09.014
[11]	SONG Z L, ZHANG J. Remote sensing image registration based on retrofitted SURF Algorithm and trajectories generated from lissajous figures[J]. IEEE Geoscienceand Remote Sensing Letters, 2010, 7(3): 491-495. doi: 10.1109/LGRS.2009.2039917
[12]	吕岩, 曲仕茹. 基于SIFT的路面裂缝配准及拼接算法[J]. 公路交通科技, 2012, 29(2): 23-28. doi: 10.3969/j.issn.1002-0268.2012.02.005 LYU Y, QU S R. An algorithm of pavement crack image registration and mosaic based on SIFT algorithm[J]. Journal of Highway and Transportation Research and Development, 2012, 29(2): 23-28. (in Chinese) doi: 10.3969/j.issn.1002-0268.2012.02.005
[13]	朱力强, 王春薇, 王耀东, 等. 基于特征点集距离描述的裂缝图像匹配算法研究[J]. 仪器仪表学报, 2016, 37(12): 2851-2857. https://www.cnki.com.cn/Article/CJFDTOTAL-YQXB201612027.htm ZHU L Q, WANG C W, WANG Y D, et al. Algorithm of crack images matching by feature points set distance description[J]. Chinese Journal of Scientific Instrument, 2016, 37 (12): 2851-2857. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YQXB201612027.htm
[14]	MOREL J M, YU G. ASIFT: A new framework for fully affine invariant image comparison[J]. SIAM Journal on Imaging Sciences, 2009, 2(2): 438-469. doi: 10.1137/080732730
[15]	胡钊政, 陶倩文, 黄刚, 等. "道路指纹"关键技术及其在智能车路系统中的应用[J]. 交通信息与安全, 2020, 38(5): 39-49. doi: 10.3963/j.jssn.1674-4861.2020.05.005 HU Z Z, TAO Q W, HUANG G, et al. A road fingerprint: Key technologies and applications for intelligent vehicles and infrastructure systems(IVIS)[J]. Journal of Transport Information and Safety, 2020, 38(5): 39-49. (in Chinese) doi: 10.3963/j.jssn.1674-4861.2020.05.005
[16]	李超峰. 路面裂缝图像拼接技术研究[D]. 西安: 西安电子科技大学, 2019. LI C F. Research on pavement crack image stitching technology[D]. Xi'an: Xidian University, 2019. (in Chinese)
[17]	RUBLEE E, RABAUD V, KONOLIGE K, et al. ORB: An efficient alternative to SIFT or SURF[C]. 2011 International Conference on Computer Vision, Barcelona, Spain: IEEE, 2011.
[18]	姜吉荣. 基于图像分析的路面裂缝检测方法与识别研究[D]. 南京: 南京邮电大学, 2016. JIANG J R. Research on pavement crack detection method and recognition based on image analysis[D]. Nanjing: Nanjing University of Posts and Telecommunications, 2016. (in Chinese)
[19]	瞿中, 林丽丹, 郭阳. 形态学与区域延伸相结合的图像裂缝检测算法研究[J]. 计算机科学, 2014, 41(11): 297-300. doi: 10.11896/j.issn.1002-137X.2014.11.058 QU Z, LIN L D, GUO Y. Algorithm of image crack detection based on morphology and region extends[J]. Computer Science, 2014, 41(11): 297-300. (in Chinese) doi: 10.11896/j.issn.1002-137X.2014.11.058
[20]	张娟, 沙爱民, 孙朝云, 等. 基于相位编组法的路面裂缝自动识别[J]. 中国公路学报, 2008, 21(2): 39-42. doi: 10.3321/j.issn:1001-7372.2008.02.008 ZHANG J, SHAA M, SUN CY, et al. Pavement crack automatic recognition based on phase-grouping method[J]. China Journal of Highway and Transport, 2008, 21 (2): 39-42(. inchinese doi: 10.3321/j.issn:1001-7372.2008.02.008
[21]	MIGUEL G, DAVID B, OSCAR M, et al. Adaptive road crack detection system by pavement classification[J]. Sensors, 2011, 11(10): 9628-9657. doi: 10.3390/s111009628
[22]	李清泉, 胡庆武. 基于图像自动匀光的路面裂缝图像分析方法[J]. 公路交通科技, 2010, 27(4): 1-5. doi: 10.3969/j.issn.1002-0268.2010.04.001 LI Q Q, HU Q W. A pavement crack image analysis approach based on automatic image dodging[J]. Journal of Highway and Transportation Research and Development, 2010, 27(4): 1-5. (in Chinese) doi: 10.3969/j.issn.1002-0268.2010.04.001