测量路面三维纹理的激光约束双目鲁棒算法

doi:10.3969/j.issn.1674-0696.2024.02.04

重庆交通大学学报（自然科学版） ›› 2024, Vol. 43 ›› Issue (2): 24-29.DOI: 10.3969/j.issn.1674-0696.2024.02.04

• 交通基础设施工程 • 上一篇

测量路面三维纹理的激光约束双目鲁棒算法

王元元1，李仁杰2，刘燕燕3，张亨通2，刘德政2

（1.湖北文理学院纯电动汽车动力系统设计与测试湖北省重点实验室,湖北襄阳441053；2.湖北文理学院机械工程学院, 湖北襄阳441053；3.重庆交通大学材料科学与工程学院，重庆400074）

收稿日期:2022-08-04 修回日期:2023-11-09 发布日期:2024-03-01
作者简介:王元元（1987—），男，湖北襄阳人，博士，教授，主要从事交通基础设施数字化方面的研究。E-mail:wangyuanyuan8710@163.com 通信作者：李仁杰（1998—），男，湖北孝感人，硕士，主要从事机器视觉试验研究方面的研究。E-mail:reajay0001@163.com
基金资助:
国家自然科学基金项目（52178422,51808084）

Laser-Constrained Binocular Robust Algorithm for Measuring 3D Texture of Pavement

WANG Yuanyuan1,LI Renjie2,LIU Yanyan3,ZHANG Hengtong2,LIU Dezheng2

(1.Hubei key Laboratory of Power System Design and Test for Electrical Vehicle, Hubei University of Arts and Science, Xiangyang 441053, Hubei, China; 2.School of Mechanical Engineering, Hubei University of Arts and Science, Xiangyang 441053, Hubei, China; 3.School of Materials Science and Engineering, Chongqing Jiaotong University, Chongqing 400074, China)

Received:2022-08-04 Revised:2023-11-09 Published:2024-03-01

摘要/Abstract

摘要： 为提升激光约束双目重构算法测量路面三维纹理的普适性，增强算法的抗光干扰能力，解决激光约束目标在变化光照下的鲁棒提取问题，基于Faster-RCNN(faster-region based convolutional neural network)构建激光约束识别模型，提取5~1 050 lux光照范围内人为构造的激光约束目标，对待测路面的约束目标开展子区域分割，进而实现子区域分割立体匹配下双目三维重构。结果表明：改进算法在1 050 lux光照强度下的平均构造深度测量值与铺砂法的拟合系数为0.987，优于原激光约束双目重构算法的0.887；在5~1 050 lux光照范围内，改进算法最大波动误差、平均波动误差分别为0.141、0.084 mm，较原激光约束双目重构算法分别降低70.20%和76.83%。

关键词: 道路工程；路面三维纹理；激光约束；双目重构；Faster-RCNN

Abstract: In order to improve the universality of the laser-constrained binocular reconstruction algorithm to measure the three-dimensional texture of the road surface, enhance the anti-light interference ability of the algorithm, and solve the problem of robust extraction under the changing illumination of the laser-constrained target, a laser-constraint recognition model was established based on Faster-RCNN (faster-region based convolutional neural network). The artificially constructed laser-constraint targets were extracted within the illumination range of 5~1 050 lux, the sub-region segmentation of the constrained target to be tested was carried out, and the binocular 3D reconstruction under the stereo matching of sub-region segmentation was realized. The results show that the fitting coefficient of the mean texture depth measurement values of the improved algorithm and the sand paving method under the illumination intensity of 1 050 lux is 0.987, which is better than 0.887 of the original laser-constrained binocular reconstruction algorithm. The maximum fluctuation error and the average fluctuation error of the improved algorithm in the illumination range of 5~1 050 lux are 0.141 mm and 0.084 mm respectively, which are 70.20% and 76.83% lower than the original laser-constrained binocular reconstruction algorithm respectively.

Key words: road engineering; three-dimensional texture of pavement; laser constraint; binocular reconstruction; Faster-RCNN

中图分类号:

U416.217

王元元1，李仁杰2，刘燕燕3，张亨通2，刘德政2. 测量路面三维纹理的激光约束双目鲁棒算法[J]. 重庆交通大学学报（自然科学版）, 2024, 43(2): 24-29.

WANG Yuanyuan1,LI Renjie2,LIU Yanyan3,ZHANG Hengtong2,LIU Dezheng2. Laser-Constrained Binocular Robust Algorithm for Measuring 3D Texture of Pavement[J]. Journal of Chongqing Jiaotong University(Natural Science), 2024, 43(2): 24-29.

参考文献

［1］黄晓明, 郑彬双. 沥青路面抗滑性能研究现状与展望［J］. 中国公路学报, 2019, 32(4): 32-49.
HUANG Xiaoming, ZHENG Binshuang. Research status and progress for skid resistance performance of asphalt pavements［J］. China Journal of Highway and Transport, 2019, 32(4): 32-49.
［2］王元元. 沥青路面抗滑特性与其表面粗糙特性之关系研究［D］. 南京: 东南大学, 2017.
WANG Yuanyuan.Study on the Relationship between Sliding Resistance of Asphalt Pavement and Its Surface Rough Characteristics［D］. Nanjing: Southeast University, 2017.
［3］ SUN Lu, WANG Yuanyuan. Three-dimensional reconstruction of macrotexture and microtexture morphology of pavement surface using six light sources-based photometric stereo with low-rank approximation［J］. Journal of Computing in Civil Engineering, 2017, 31(2): 4016054.1.
［4］ HONG P N, AHN C W. Stereo matching methods for imperfectly rectified stereo images［J］. Symmetry, 2019, 11(4): 570.
［5］王元元, 孙璐, 刘卫东, 等. 测量路面三维纹理双目重构算法的约束改进［J］. 吉林大学学报(工学版), 2021, 51(4): 1342-1348.
WANG Yuanyuan, SUN Lu, LIU Weidong, et al. Constraint improvement of binocular reconstruction algorithm used to measure pavement three-dimensional texture［J］. Journal of Jilin University (Engineering and Technology Edition), 2021, 51(4): 1342-1348.
［6］ LIU Yanyan, WANG Yuanyuan, CAI Xinyi, et al. The detection effect of pavement 3D texture morphology using improved binocular reconstruction algorithm with laser line constraint［J］. Measurement, 2020, 157: 107638.
［7］马永杰, 程时升, 马芸婷, 等. 卷积神经网络及其在智能交通系统中的应用综述［J］. 交通运输工程学报, 2021, 21(4): 48-71.
MA Yongjie, CHENG Shisheng, MA Yunting,et al. Review of convolutional neural network and its application in intelligent transportation system［J］. Journal of Traffic and Transportation Engineering, 2021, 21(4): 48-71.
［8］ NGUYEN T H, NGUYEN T L, SIDOROV D N, et al. Machine learning algorithms application to road defects classification［J］. Intelligent Decision Technologies, 2018, 12(1): 59-66.
［9］ HOANG N D, NGUYEN Q L. A novel method for asphalt pavement crack classification based on image processing and machine learning［J］. Engineering with Computers, 2019, 35(2): 487-498.
［10］彭博,黄大荣,郭黎,等. 基于像素-亚像素级形态分析的路面三维图像裂缝自动识别算法［J］.重庆交通大学学报(自然科学版),2018,37(9):34-42.
PENG Bo, HUANG Darong, GUO Li, et al. Automatic crack detection algorithm from 3D pavement images based on shape analysis at pixel-subpixel level［J］. Journal of Chongqing Jiaotong University (Natural Science),2018,37(9):34-42.
［11］姚万业, 李金平. 基于改进Faster R-CNN的行人检测算法［J］. 科学技术与工程, 2020, 20(4): 1498-1503.
YAO Wanye, LI Jinping. Pedestrian detection algorithm based on improved Faster R-CNN［J］. Science Technology and Engineering, 2020, 20(4): 1498-1503.
［12］赵立新, 邢润哲, 白银光, 等. 深度学习在目标检测的研究综述［J］. 科学技术与工程, 2021, 21(30): 12787-12795.
ZHAO Lixin, XING Runzhe, BAI Yinguang,et al. Review on survey of deep learning in target detection［J］. Science Technology and Engineering, 2021, 21(30): 12787-12795.

[1]	李骏1，何超1，刘震2，3，顾兴宇2，3. 基于多尺度加载试验的沥青路面结构参数设计方法[J]. 重庆交通大学学报（自然科学版）, 2024, 43(1): 31-38.
[2]	李志勇, 朱尚书. 明色化铺装材料的路用性能研究[J]. 重庆交通大学学报（自然科学版）, 2023, 42(2): 37-43.
[3]	王民1,2，尚飞1，肖丽1，包广志3. 钢桥面浇注式沥青铺装复合梁疲劳损伤规律分析[J]. 重庆交通大学学报（自然科学版）, 2021, 40(03): 84-88.
[4]	李升连1,梁乃兴2,曾晟2. 采集方法对数字图像沥青混合料离析评价影响研究[J]. 重庆交通大学学报（自然科学版）, 2021, 40(03): 103-107.
[5]	周建昆1，曹源文2，温永杰2，赵江1，白丽萍3. 不同高度下摊铺路面的数字图像差异性研究[J]. 重庆交通大学学报（自然科学版）, 2021, 40(02): 89-94.
[6]	刘洪辉，李晓娟. 水性环氧乳化沥青高温流变性能研究[J]. 重庆交通大学学报（自然科学版）, 2020, 39(10): 67-73.
[7]	李海莲1，2，林梦凯1，2，王起才1，2. 高速公路沥青路面使用性能模糊区间评价方法研究[J]. 重庆交通大学学报（自然科学版）, 2020, 39(09): 80-87.
[8]	马丽英，李茂其，曹源文,董琴琴. 沥青桥面压实均匀性实时检测方法研究[J]. 重庆交通大学学报（自然科学版）, 2020, 39(02): 69-74.
[9]	吴国雄1,2，叶新雨1，余苗1，杨丹彤3. 基于黏弹性的橡胶改性沥青路面抗滑性能研究[J]. 重庆交通大学学报（自然科学版）, 2020, 39(02): 81-86.
[10]	付军1,3，刘智鸿1,2，左雪娜3，赖泽涵1，王学智1，丁庆军4. 区域气候变化对沥青路面夏季温度效应的影响分析[J]. 重庆交通大学学报（自然科学版）, 2020, 39(02): 87-94.
[11]	高建红，许有俊，杨圣春. 粒料类基层沥青路面动力响应研究[J]. 重庆交通大学学报（自然科学版）, 2019, 38(12): 57-62.
[12]	张东长1，宁斌权2，伍杰1，陈飞1. 渗透式薄层抗滑材料性能研究[J]. 重庆交通大学学报（自然科学版）, 2019, 38(10): 61-67.
[13]	李浩1,2，许新权2，刘锋2，龚侥斌3. 四种均布荷载接触形式的路面力学响应对比分析[J]. 重庆交通大学学报（自然科学版）, 2019, 38(08): 53-58.
[14]	周兴林，李庆丰，肖神清. 基于多重分形的沥青混合料空隙分布特征分析[J]. 重庆交通大学学报（自然科学版）, 2018, 37(12): 29-35.
[15]	高嫄嫄1,2，王维玉2，赵庆新1. 基于解析方法的半刚性基层沥青路面裂缝反射问题研究[J]. 重庆交通大学学报（自然科学版）, 2018, 37(12): 49-54.

测量路面三维纹理的激光约束双目鲁棒算法

Laser-Constrained Binocular Robust Algorithm for Measuring 3D Texture of Pavement

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics