-
摘要: 为了实现强噪声、弱光照、低对比度条件下的机场道面细小裂缝检测, 设计了基于深度图像的机场道面裂缝检测算法; 将采集到的深度图像划分成多个网格, 并对每个网格进行扩充, 获得了局部道面区域; 针对每个网格区域, 基于随机抽样一致算法进行局部三次曲面构建和优化估计; 在此基础上, 在全局尺度下融合全部网格区域的曲面模型, 生成整个图像采集区域道面的全局曲面模型; 利用全局曲面模型与原始深度图像之间的差值图像, 采用自适应阈值方法分割出候选裂缝像素, 并利用裂缝的像素总数、长度以及长宽比等多种形态学约束筛选候选裂缝像素, 去除错误的候选裂缝像素, 从而获得了最终的裂缝检测结果; 在机场道面深度图像数据集上进行了试验, 以人工标注结果作为真实值, 以准确率、召回率以及F值作为量化评估指标, 将提出的算法分别与4种有代表性的传统算法进行了对比。试验结果表明: 传统算法能够取得的最高准确率、召回率以及F值分别为77.05%、41.02%和50.02%, 提出的算法在准确率、召回率和F值3个指标上均有明显优势, 其均值分别为91.20%、97.99%和94.12%;提出的算法能够在分辨率为1 984像素×2 000像素的深度图像上检测出最小宽度为3 mm、最小长度为10 cm的裂缝, 实现了在复杂机场道面场景中识别细小裂缝的目标。Abstract: To detect small cracks in airport pavements under strong noise, weak illumination and low contrast, a crack detection algorithm for airport pavements based on depth images was designed. The collected depth image was divided into multiple grids, and each grid was expanded to obtain a local pavement region. For each grid region, the random sampling consensus algorithm was used to construct and optimally estimate the local cubic curved surface. On this basis, the global curved surface model of the entire image acquisition region of parement was generated by fusing the curved surface models of all grid regions under the global scale. Based on the difference image between the global curved surface model and the original depth image, candidate crack pixels were segmented by the adaptive threshold method, and various morphological constraints, such as the total number, length and length width ratio of crack pixels were used to screen the candidate crack pixels to eliminate the incorrect candidate crack pixels, so as to obtain the final crack detection results. The experiment was carried out on the airport pavement depth image datasets. The manual annotation results were taken as the ground truth. The accuracy, recall rate and F value were used as the quantitative evaluation indices. The proposed algorithm was compared with four representative traditional algorithms. Experiment result shows that the highest accuracy, recall rate and F value of the traditional algorithm are 77.05%, 41.02% and 50.02%, respectively. The proposed algorithm has obvious advantages in the accuracy, recall rate and F value, with average values of 91.20%, 97.99% and 94.12%, respectively. The proposed algorithm can detect the crack with the minimum width of 3 mm and the minimum length of 10 cm in the depth image with a resolution of 1 984×2 000, and realize the target of detecting small cracks in the complex airport pavement scene.
-
Key words:
- airport pavement /
- crack detection /
- depth image /
- multi-scale curved surface model
-
图 7 不同k时算法的平均准确率、召回率和F值
Figure 7. Average accuracies, recall rates and F values of algorithm with different k
图 8 不同k时算法的平均准确率、召回率、F值
Figure 8. Average accuracies, recall rates and F values of algorithm with different k
图 9 准确率、召回率、F值累计频率分布(机场道面)
Figure 9. Cumulative frequency distributions of accuracy, recall rate and F value (airport pavement)
图 10 提出的算法在公路数据集上的检测结果
Figure 10. Detection result of proposed algorithm on road pavement dataset
表 1 机场道面数据集检测结果
Table 1. Detection result of airport pavement dataset
% 方法 准确率 召回率 F值 CrackIT 11.63 8.74 6.24 平面拟合算法 6.25 70.60 9.93 Canny 59.70 2.40 4.50 Crack Forest 77.05 41.02 50.02 本文算法 91.20 97.99 94.12 
下载: 导出CSV
表 2 公路路面数据集检测结果
Table 2. Detection result of road pavement dataset
% 方法 准确率 召回率 F值 CrackIT 2.44 6.45 3.40 平面拟合算法 1.21 90.22 2.38 Canny 91.50 6.22 11.57 CrackForest 81.75 43.97 54.95 本文算法 83.26 97.91 89.76
下载: 导出CSV
-
[1] SONG Hong-xun, WANG Wei-xing, WANG Feng-ping, et al. Pavement crack detection by ridge detection on fractional calculus and dual-thresholds[J]. International Journal of Multimedia and Ubiquitous Engineering, 2015, 10(4): 19-30. doi: 10.14257/ijmue.2015.10.4.03 [2] 张德津, 李清泉, 陈颖, 等. 基于空间聚集特征的沥青路面裂缝检测方法[J]. 自动化学报, 2016, 42(3): 443-454. https://www.cnki.com.cn/Article/CJFDTOTAL-MOTO201603010.htmZHANG De-jin, LI Qing-quan, CHEN Ying, et al. Asphalt pavement crack detection based on spatial clustering featurel[J]. Acta Automatica Sinica, 2016, 42(3): 443-454. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-MOTO201603010.htm [3] OLIVEIRA H, CORREIA P L. Crack IT—an image processing toolbox for crack detection and characterization[C]∥DETLEV M. 2014 IEEE International Conference on Image Processing (ICIP). New York: IEEE, 2014: 798-802. [4] AMHAZ R, CHAMBON S, IDIER J, et al. Automatic crack detection on 2D pavement images: an algorithm based on minimal path selection[J]. IEEE Transactions on Intelligent Transportation Systems, 2016, 17(10): 2718-2729. doi: 10.1109/TITS.2015.2477675 [5] SHI Yong, CUI Li-meng, QI Zhi-quan, et al. Automatic road crack detection using random structured forests[J]. IEEE Transactions on Intelligent Transportation Systems, 2016, 17(12): 3434-3445. doi: 10.1109/TITS.2016.2552248 [6] 李良福, 马卫飞, 李丽, 等. 基于深度学习的桥梁裂缝检测算法研究[J]. 自动化学报, 2019, 45(9): 1727-1742. https://www.cnki.com.cn/Article/CJFDTOTAL-MOTO201909010.htmLI Liang-fu, MA Wei-fei, LI Li, et al. Research on detection algorithm for bridge cracks based on deep learning[J]. Acta Automatica Sinica, 2019, 45(9): 1727-1742. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-MOTO201909010.htm [7] 孙朝云, 马志丹, 李伟, 等. 基于深度卷积神经网络融合模型的路面裂缝识别方法[J]. 长安大学学报(自然科学版), 2020, 40(4): 1-13. https://www.cnki.com.cn/Article/CJFDTOTAL-XAGL202004002.htmSUN Zhao-yun, MA Zhi-dan, LI Wei, et al. Pavement crack identification method based on deep convolutional neural network fusion mode[J]. Journal of Chang'an University (Natural Science Edition), 2020, 40(4): 1-13. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-XAGL202004002.htm [8] KATAKAM N. Pavement crack detection system through localized thresholding[D]. Toledo: The University of Toledo, 2009. [9] 徐威, 唐振民, 吕建勇. 基于图像显著性的路面裂缝检测[J]. 中国图象图形学报, 2018, 18(1): 69-77. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGTB201301010.htmXU Wei, TANG Zhen-min, LYU Jian-yong. Pavement crack detection based on image saliency[J]. Journal of Image and Graphics, 2018, 18(1): 69-77. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-ZGTB201301010.htm [10] 瞿中, 鞠芳蓉, 陈思琪. 结构森林边缘检测与渗流模型相结合的混凝土表面裂缝检测[J]. 计算机科学, 2018, 45(11): 295-298, 311. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJA201811048.htmQU Zhong, JU Fang-rong, CHEN Si-qi. Concrete surface cracks detection combining structured forest edge detection and percolation model[J]. Computer Science, 2018, 45(11): 295-298, 311. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-JSJA201811048.htm [11] 柴雪松, 朱兴永, 李健超, 等. 基于深度卷积神经网络的隧道衬砌裂缝识别算法[J]. 铁道建筑, 2018, 58(6): 60-65. doi: 10.3969/j.issn.1003-1995.2018.06.16CHAI Xue-song, ZHU Xing-yong, LI Jian-chao, et al. Tunnel lining crack identification algorithm based on deep convolutional neural network[J]. Railway Engineering, 2018, 58(6): 60-65. (in Chinese). doi: 10.3969/j.issn.1003-1995.2018.06.16 [12] PRASANNA P, DANA K J, GUCUNSKI N, et al. Automated crack detection on concrete bridges[J]. IEEE Transactions on Automation Science and Engineering, 2014, 13(99): 1-9. [13] 曹锦纲, 杨国田, 杨锡运. 基于注意力机制的深度学习路面裂缝检测[J]. 计算机辅助设计与图形学学报, 2020, 32(8): 1324-1333. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJF202008014.htmCAO Jin-gang, YANG Guo-tian, YANG Xi-yun. Pavement crack detection with deep learning based on attention mechanism[J]. Journal of Computer-Aided Design and Computer Graphics, 2020, 32(8): 1324-1333. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-JSJF202008014.htm [14] RODRÍGUEZ-MARTÍN M, LAGÜELA S, GONZÁLEZ-AGUILERA D, et al. Thermographic test for the geometric characterization of cracks in welding using IR image rectification[J]. Automation in Construction, 2016, 61: 58-65. doi: 10.1016/j.autcon.2015.10.012 [15] XU Chang-hang, XIE Jing, CHEN Guo-ming, et al. An infrared thermal image processing framework based on superpixel algorithm to detect cracks on metal surface[J]. Infrared Physics and Technology, 2014, 67: 266-272. doi: 10.1016/j.infrared.2014.08.002 [16] PARK H, CHOI M, PARK J, et al. A study on detection of micro-cracks in the dissimilar metal weld through ultrasound infrared thermography[J]. Infrared Physics and Technology, 2014, 62: 124-131. doi: 10.1016/j.infrared.2013.10.006 [17] DHITAL D, LEE J R. A fully non-contact ultrasonic propagation imaging system for closed surface crack evaluation[J]. Experimental Mechanics, 2012, 52: 1111-1122. doi: 10.1007/s11340-011-9567-z [18] ILIOPOULOS S, AGGELIS D G, PYL L, et al. Detection and evaluation of cracks in the concrete buffer of the Belgian nuclear waste container using combined NDT techniques[J]. Construction and Building Materials, 2015, 78: 369-378. doi: 10.1016/j.conbuildmat.2014.12.036 [19] HUANG Yu-chun, TSAI Y. Crack fundamental element (CFE) for multi-scale crack classification[C]//SCARPAS A. 7th RILEM International Conference on Cracking in Pavements. Berlin: Springer, 2012: 419-428. [20] PENG Bo, WANG K C P, CHEN Cheng. Automatic crack detection by multi-seeding fusion on 1 mm resolution 3D pavement images[C]∥AMIY V. T & amp; amp; DI Congress 2014. Reston: ASCE, 2014: 543-552. [21] 李伟, 呼延菊, 沙爱民, 等. 基于3D数据和双尺度聚类算法的路面裂缝检测[J]. 华南理工大学学报(自然科学版), 2015, 43(8): 99-105. doi: 10.3969/j.issn.1000-565X.2015.08.015LI Wei, HU Yan-ju, SHA Ai-min, et al. Two-scale clustering algorithm road pavement crack detection technology based on 3D data[J]. Journal of South China University of Technology (Natural Science Edition), 2015, 43(8): 99-105. (in Chinese). doi: 10.3969/j.issn.1000-565X.2015.08.015 [22] CAO Ting, WANG Wei-xing, TIGHE S, et al. Crack image detection based on fractional differential and fractal dimension[J]. IET Computer Vision, 2019, 13(1): 79-85. doi: 10.1049/iet-cvi.2018.5337 [23] 宋葵阳, 徐贵力, 李开宇, 等. 基于结构光的路面裂缝检测方法研究[J]. 机械制造与自动化, 2016(6): 223-228. doi: 10.3969/j.issn.1671-5276.2016.06.060SONG Kui-yang, XU Gui-li, LI Kai-yu, et al. Research on method of pavement crack detection based on structured light[J]. Machine Building and Automation, 2016(6): 223-228. (in Chinese). doi: 10.3969/j.issn.1671-5276.2016.06.060 [24] MOHAMMAD R J, FARROKH J, SAMI F M, et al. Unsupervised approach for autonomous pavement-defect detection and quantification using an inexpensive depth sensor[J]. Journal of Computing in Civil Engineering, 2013, 27: 743-754. doi: 10.1061/(ASCE)CP.1943-5487.0000245 [25] OUYANG W, XU B. Pavement cracking measurements using 3D laser-scan images[J]. Measurement Science and Technology, 2013, 24(10): 105204-1-9. [26] 蒋彬, 金湘亮. 改进的TOF相机谐波和强度误差校正算法设计[J]. 光学学报, 2020, 40(1): 0111024-1-10. https://www.cnki.com.cn/Article/CJFDTOTAL-GXXB202001025.htmJIANG Bin, JIN Xiang-liang. Improved correction algorithm for harmonic- and intensity- related errors in time-of-flight cameras[J]. Acta Optica Sinica, 2020, 40(1): 0111024-1-10. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-GXXB202001025.htm [27] 李占利, 周康, 牟琦, 等. TOF相机实时高精度深度误差补偿方法[J]. 红外与激光工程, 2019, 48(12): 1213004-1-10. https://www.cnki.com.cn/Article/CJFDTOTAL-HWYJ201912035.htmLI Zhan-li, ZHOU Kang, MU Qi, et al. TOF camera real-time high precision depth error compensation method[J]. Infrared and Laser Engineering, 2019, 48(12): 1213004-1-10. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-HWYJ201912035.htm [28] 杜瑞建, 葛宝臻, 陈雷. 多视高分辨率纹理图像与双目三维点云的映射方法[J]. 中国光学, 2020, 13(5): 1055-1064. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA202005017.htmDU Rui-jian, GE Bao-zhen, CHEN Lei. Texture mapping of multi-view high-resolution images and binocular 3D point clouds[J]. Chinese Optics, 2020, 13(5): 1055-1064. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA202005017.htm [29] 丁莹, 范静涛, 宋天喻. 双目立体视觉检测系统正向最优化设计方法研究[J]. 仪器仪表学报, 2016, 37(3): 650-657. doi: 10.3969/j.issn.0254-3087.2016.03.023DING Ying, FAN Jing-tao, SONG Tian-yu. Optimal forward design method for the binocular stereo vision inspection system[J]. Chinese Journal of Scientific Instrument, 2016, 37(3): 650-657. (in Chinese). doi: 10.3969/j.issn.0254-3087.2016.03.023 [30] 徐志刚, 车艳丽, 李金龙, 等. 路面破损图像自动处理技术研究进展[J]. 交通运输工程学报, 2019, 19(1): 172-190. doi: 10.3969/j.issn.1671-1637.2019.01.017XU Zhi-gang, CHE Yan-li, LI Jin-long, et al. Research progress on automatic image processing technology for pavement distress[J]. Journal of Traffic and Transportation Engineering, 2019, 19(1): 172-190. (in Chinese). doi: 10.3969/j.issn.1671-1637.2019.01.017 [31] 刘飞, 何春桥, 申爱民, 等. 结构光饱和区域分区投射优化补偿方法[J]. 光学学报, 2018, 38(6): 0612001-1-8. https://www.cnki.com.cn/Article/CJFDTOTAL-GXXB201806023.htmLIU Fei, HE Chun-qiao, SHEN Ai-min, et al. Optimized compensation method of divisional projection for saturated region of structured light[J]. Acta Optica Sinica, 2018, 38(6): 0612001-1-8. (in Chinese). https://www.cnki.com.cn/Article/CJFDTOTAL-GXXB201806023.htm [32] CANNY J. A computational approach to edge detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1986, 8(6): 679-698. -
点击查看大图
图(11) / 表(2)
计量
- 文章访问数: 1084
- HTML全文浏览量: 309
- PDF下载量: 298
- 被引次数: 0
