基于车载视频抖动矢量的路面平整性评估方法

陈子昂; 陈新; 曾宇同; 郭唐仪

doi:10.3963/j.jssn.1674-4861.2024.02.011

基于车载视频抖动矢量的路面平整性评估方法

doi: 10.3963/j.jssn.1674-4861.2024.02.011

1.
南京理工大学自动化学院南京 210094
2.
中国兵器工业集团北方光电股份有限公司西安 710043

基金项目:

国家重点研发计划-政府间国际科技创新合作项目 2019YFE0123800

南京市国际合作项目 202002013

详细信息

作者简介:
陈子昂（1999—），硕士研究生，研究方向：交通运输工程. E-mail：434241989@qq.com

通讯作者:
陈新（1970—），硕士，副教授. 研究方向：交通信息工程及控制. E-mail：6470959@qq.com

中图分类号: U416.2
计量
- 文章访问数: 121
- HTML全文浏览量: 83
- PDF下载量: 4
- 被引次数: 0
出版历程
- 收稿日期: 2023-06-08
- 网络出版日期: 2024-09-14

Assessing Pavement Rougness Using Jitter Vector from In-vehicle Camera Videos

1.
School of Automation, Nanjing University of Science and Technology, Nanjing 210094, China
2.
North Electro-Optic Co., Ltd, China North Industries Group Corporation Limited, Xi'an 710043, China

摘要

摘要: 针对路面平整性评估流程繁琐、效率低、周期长等问题，提出基于车载视频抖动矢量的路面平整性评估方法，实现常态化场景下对路面状态的初步快速筛选评估。使用车载采集设备获取的行车视频作为评估数据基础，对车载图像进行预处理，增强行车视频图像的对比度，降低行车环境变化对视频图像对比度的影响。利用分块灰度投影算法对视频图像进行相似性判定，去除大偏差的抖动矢量和运动目标干扰，提取行车视频的主要抖动矢量特征。采用粒子群优化算法改进投影相关性曲线的搜索模式，通过使用行（列）方向的灰度投影曲线相关性作为适应度函数来提高算法的搜索效率。建立基于遗传算法（genetic algorithm，GA）优化的K-means聚类分析算法，实现了自主采集路段中不同车速条件下的路面平整性分级评估。通过自主采集数据实验验证，基于粒子群优化的灰度投影算法在检测平整路面时，耗时0.148 s，算法效率比原算法提高了91.41%；在检测粗糙路面时，耗时0.123 s，算法效率比原算法法提高了87.58%，且检测出的抖动矢量数值一致。本文提出的基于车载视频抖动矢量的GA-K-means路面平整性分级评估方法能够有效降低初始聚类中心的干扰。
- 道路工程 /
- 车载视频抖动矢量 /
- 路面平整性评估 /
- 灰度投影法 /
- GA-K-means聚类算法
Abstract: The process of assessing pavement smoothness is cumbersome, inefficient and time-consuming. To address these issues, a pavement smoothness assessment method based on in-vehicle video jitter vectors is proposed. This method enables preliminary and rapid screening of pavement conditions under normal scenarios. It uses driving videos collected by onboard devices as the assessment data. Preprocessing enhances the contrast of driving video images and reduces the effect of changes in the driving environment on the contrast of video images. The video images then undergo block-wise grayscale projection and similarity determination to remove significant deviations in jitter vectors and interference from moving objects. This extracts the main jitter vectors from the driving videos. The particle swarm optimization (PSO) algorithm improves the search pattern of the projection correlation curve. Using the grayscale projection curve correlation formula as the fitness function in the row (or column) direction enhances search efficiency of the algorithm. A genetic algorithm (GA) optimized K-means clustering algorithm is established to autonomously assess road smoothness at different vehicle speeds by combining vehicle speed and video jitter vectors. Experimental validation shows that the PSO-based grayscale projection algorithm detects smooth road surfaces in 0.148 s, improving efficiency by 91.41% compared to the original algorithm. For rough road surfaces, detection takes 0.123 s, improving efficiency by 87.58%, and consistently detects jitter vector values. The GA-K-means algorithm effectively reduces interference from initial cluster centers, avoiding premature conver-gence.
- road engineering /
- jitter vector from in-vehicle camera videos /
- pavement smoothness assessing /
- gray-scale projection algorithm /
- GA-K-means algorithm

HTML全文

图 1 改进灰度投影算法流程图

Figure 1. Flow chart of the program to improve the gray scale projection algorithm

下载: 全尺寸图片幻灯片

图 2 第1分块的灰度投影相关曲线

Figure 2. Grey scale projection correlation curves for the first part

下载: 全尺寸图片幻灯片

图 3 行车视频抖动检测结果

Figure 3. Driving video jitter test results

下载: 全尺寸图片幻灯片

图 4 车辆经过不同状况路面的抖动检测

Figure 4. Detection of vehicle shaking over different road conditions

下载: 全尺寸图片幻灯片

图 5 利用遗传算法优化K-means聚类的流程图

Figure 5. Optimization of K-means clustering process using genetic algorithm

下载: 全尺寸图片幻灯片

图 6 K-means聚类和改进K-means聚类收敛曲线对比图

Figure 6. Comparison of convergence curves of K-means clustering and improved K-means clustering

下载: 全尺寸图片幻灯片

图 7 不同车速区间行车抖动量均值聚类分析结果

Figure 7. Cluster analysis results of the mean value of travel jitter in different speed zone

下载: 全尺寸图片幻灯片

表 1 数据基础

Table 1. Data foundation

数据采集日期	数据采集路段	车速（/km/h）	路面粗糙性能的当前状态
2021.5.10	南京市中山门大街、紫金东路、双麒路等	30~50	路面坑洞、破损、凹陷凸起
2021.8.29	学校一号路、二号路及附近市区道路路段	25~40	路面坑洞、破损、凹陷凸起、窨井盖及减速带
2021.9.16	南京市友谊路，学校二号路及中山门大街	25~40	路面坑洞、破损、凹陷凸起、窨井盖及减速带
2021.9.18	南京市中山门大街、紫金山路、光华路、金马路、双麒路等，高速公路省道和国道路段	40~50 70~110	路面坑洞、破损、凹陷凸起、窨井盖及减速带，高速公路路段的桥头跳车

下载: 导出CSV

表 2 不同算法的检测对比

Table 2. Comparison of detection with different algorithms

路面类型	方法对比	全局搜索时间/s	抖动矢量	提升效率/%
平整路面	改进算法	0.148	1	91.41
平整路面	传统算法	1.722	1	91.41
粗糙路面	改进算法	0.123	－5	87.58
粗糙路面	传统算法	0.990	－5	87.58

下载: 导出CSV

表 3 路面平整性等级划分

Table 3. Grading of road surface levelness

平整性等级	定性描述
第1级	路面平整舒适
第2级	路面较平整
第3级	路面出现粗糙
第4级	路面较粗糙
第5级	路面严重粗糙

下载: 导出CSV

表 4 路面平整性的分级评估实例

Table 4. Example of grading assessment of road surface levelness

数据采集路段	车速/(km/h)	粗糙性能当前状态	行车抖动量均值	粗糙性等级
南京市中山门大街、紫金东路、双麒路等	40~50	路面坑洞、破损、凹陷凸起	0.677	第2级
学校一号路、二号路及附近市区道路路段	30~40	路面坑洞、破损、凹陷凸起、窨井盖及减速带	2.321	第3级
南京市中山门大街、紫金山路、光华路、金马路、双麒路等，高速公路省道和国道路段	40~50	路面坑洞、破损、凹陷凸起、窨井盖及减速带，高速公路路段的桥头跳车	1.886	第4级

下载: 导出CSV

参考文献(25)

[1]	惠记庄, 张泽宇, 叶敏, 等. 公路建养装备数字孪生技术综述[J]. 交通运输工程学报, 2023, 23(4): 23-44. HUI J Z, ZHANG Z Y, YE M, et al. A review of digital twin technology for highway construction and maintenance equipment[J]. Journal of Transportation Engineering, 2023, 23(4): 23-44. (in Chinese)
[2]	何洪文, 孙逢春, 李梦林. 我国综合交通工程科技现状及未来发展[J]. 中国工程科学, 2023, 25(6): 202-211. HE H W, SUN F C, LI M L. Current status and future development of comprehensive transportation engineering science and technology in China[J]. China Engineering Science, 2023, 25(6): 202-211. (in Chinese)
[3]	KALOOP M R, El-BADAWY S M, HU J W, et al. International roughness index prediction for flexible pavements using novel machine learning techniques[J]. Engineering Applications of Artificial Intelligence, 2023, 122: 106007. doi: 10.1016/j.engappai.2023.106007
[4]	LIU J, LIU F, ZHENG C, et al. Improving asphalt mix design considering international roughness index of asphalt pavement predicted using autoencoders and machine learning[J]. Construction and Building Materials, 2022, 360: 129439. doi: 10.1016/j.conbuildmat.2022.129439
[5]	LIU C, WU D, LI Y, et al. Large-scale pavement roughness measurements with vehicle crowdsourced data using semi-supervised learning[J]. Transportation Research Part C: Emerging Technologies, 2021, 125: 103048. doi: 10.1016/j.trc.2021.103048
[6]	AUGUSTAUSKAS R, LIPNICKAS A. Improved pixel-level pavement-defect segmentation using a deep autoencoder[J]. Sensors, 2020, 20(9): 2557. doi: 10.3390/s20092557
[7]	张金喜, 王琳, 周同举, 等. 基于行车振动的路面平整性智能检测方法研究[J]. 中外公路, 2020, 40(1): 31-36. ZHANG J X, WANG L, ZHOU T J, et al. Study on Intelligent detection method of pavement roughness based on vibration of moving vehicles[J]. Journal of China & Foreign Highway, 2020, 40(1): 31-36. (in Chinese)
[8]	CERECEDA D, MEDEL-VERA C, ORTIZ M, et al. Roughness and condition prediction models for airfield pavements using digital image processing[J]. Automation in Construction, 2022, 139: 104325. doi: 10.1016/j.autcon.2022.104325
[9]	BASHAR M Z, TORRES-MACHI C. Deep learning for estimating pavement roughness using synthetic aperture radar data[J]. Automation in Construction, 2022, 142: 104504. doi: 10.1016/j.autcon.2022.104504
[10]	TIAN C, ZHENG M, ZUO W, et al. Multi-stage image denoising with the wavelet transform[J]. Pattern Recognition, 2023, 134: 109050. doi: 10.1016/j.patcog.2022.109050
[11]	李之红, 申天宇, 文琰杰, 等. 基于混合机器学习框架的网约车订单需求预测与异常点识别[J]. 交通信息与安全, 2023, 41(3): 157-165, 174. LI Z H, SHEN T Y, WENG Y J, et al. Order demand prediction and anomaly-point identification for online car-hailing orders based on hybrid machine learning framework[J]. Journal of Transport Information and Safety, 2023, 41(3): 157-165, 174. (in Chinese)
[12]	KREMER T, IRONS T, M LLER-PETKE M, et al. Review of acquisition and signal processing methods for electromagnetic noise reduction and retrieval of surface nuclear magnet-ic resonance parameters[J]. Surveys in Geophysics, 2022, 43(4): 999-1053. doi: 10.1007/s10712-022-09695-3
[13]	GUO J, SI Z, XIANG J. A compound fault diagnosis method of rolling bearing based on wavelet scattering transform and improved soft threshold denoising algorithm[J]. Measurement, 2022, 196: 111276. doi: 10.1016/j.measurement.2022.111276
[14]	张玺君, 袁占亭, 张红, 等. 交通轨迹大数据预处理方法研究[J]. 计算机工程, 2019, 45(6): 26-31. ZHANG X J, YUAN Z T, ZHANG H, et al. Research on preprocessing method for traffic trajectory big data[J]. Computer Engineering, 2019, 45(6): 26-31. (in Chinese)
[15]	GOLDBERG P W, KATZMAN M J. Lower bounds for the query complexity of equilibria in Lipschitz games[J]. Theoretical Computer Science, 2023, 962: 113931.
[16]	LI H, CAO Y, WAN Y, et al. An improved temporal phase unwrapping based on super-grayscale multi-frequency grating projection[J]. Optics and Lasers in Engineering, 2022, 153: 106990.
[17]	FENG Y, DANG Y, WANG J, et al. A novel grey projection incidence model for assessing the relationships between cardiovascular diseases and air pollutants[J]. ISA Transactions, 2023, 135: 398-409. .
[18]	党媛媛, 陈兆学. 基于灰度积分投影与霍夫圆变换算法的人眼瞳距自动检测[J]. 计算机系统应用, 2022, 31(7): 259-264. DANG Y Y, CHEN Z X. Automatic detection of human eye pupil distance based on gray integral projection and hough circle transform algorithm[J]. Computer Systems Application, 2022, 31(7): 259-264. (in Chinese)
[19]	胡建祥, 侯毅男. 基于多元灰度投影的无人艇电子稳像方法[J]. 弹箭与制导学报, 2021, 41(4): 65-68, 73. HU J X, HOU Y N. USV video stabilization algorithm based on multilayer gray projection[J]. Journal of Projectiles, Rockets, Missiles and Guidance, 2021, 41(4): 65-68, 73. (in Chinese)
[20]	胡常俊, 张著洪. 基于视频抖动的灰度投影稳像算法[J]. 贵州大学学报(自然科学版), 2019, 36(1): 82-86. HU C J, ZHANG Z H. Video jitter-based image stabilization algorithm for grayscale projection[J]. Journal of Guizhou University(Natural Sciences), 2019, 36(1): 82-86. (in Chinese)
[21]	高玮宁, 马善涛, 何勇军, 等. 图像灰度投影的聚焦窗口选择方法[J]. 哈尔滨理工大学学报, 2021, 26(5): 60-67. GAO W N, MA S T, HE Y J, et al. A focusing window selection based on gray-scale projection[J]. Journal of harbin university of science and technology, 2021, 26(5): 60-67. (in Chinese)
[22]	GUPTA G. Algorithm for image processing using improved median filter and comparison of mean, median and improved median filter[J]. International Journal of Soft Computing and Engineering(IJSCE), 2011, 1(5): 304-311.
[23]	常振廷, 肖智豪, 张文军, 等. 基于网格分类与纵横向注意力的城市道路车道线检测方法[J]. 交通信息与安全, 2023, 41(3): 92-102, 110 CHANG Z T, XIAO Z H, ZHANG W J, et al. Lane line detection method for urban roads based on grid classification and vertical and horizontal attention[J]. Journal of Transport Information and Safety, 2023, 41(3): 92-102, 110. (in Chinese)
[24]	LI H, WANG J. Collaborative annealing power k-means ++ clustering[J]. Knowledge-Based Systems, 2022, 255: 109593.
[25]	KORDOS M, BLACHNIK M, SCHERER R. Fuzzy clustering decomposition of genetic algorithm-based instance selection for regression problems[J]. Information Sciences, 2022, 587: 23-40.