Spatiotemporal matching method for tracking pavement distress using high-frequency detection data

Various algorithms based on deep learning have achieved promising results in pavement distress detection. However, the detected distresses are not tracked throughout the life cycle. In long-term application scenarios, pavement distresses may take on different forms due to image acquisition mode, distress development, and environmental change, which make tracking distresses a tough question. We present in this study a spatiotemporal matching method based on high-frequency real pavement distress datasets. Pavement distresses of fixed routes were collected 30 times over 5 months, and distresses with spatiotemporal information were obtained at time series. We apply image rectification, stitching and distress class, and bounding box generation algorithms for pre-processing to align the collected images to the same-detail level and angle. A four-step spatiotemporal matching module is designed, including global positioning system (GPS) filtering, class filtering, relative position filtering, and distress feature filtering. The results reveal that the comprehensive rank-3 hit rate of the matching method reaches 88.73%, and the method is robust to environmental factors, which helps show performance decay of distresses and the effect of maintenance operations. It is concluded that the spatiotemporal matching method is convenient to operate, and it lays the foundation for an agency to track distress evolution and make timely treatment of distresses in the life cycle.     

An Unmanned Aerial Vehicle‐Based Imaging System for 3D Measurement of Unpaved Road Surface Distresses1

Abstract: Road condition data are important in transportation management systems. Over the last decades, significant progress has been made and new approaches have been proposed for efficient collection of pavement condition data. However, the assessment of unpaved road conditions has been rarely addressed in transportation research. Unpaved roads constitute approximately 40% of the U.S. road network, and are the lifeline in rural areas. Thus, it is important for timely identification and rectification of deformation on such roads. This article introduces an innovative Unmanned Aerial Vehicle (UAV)‐based digital imaging system focusing on efficient collection of surface condition data over rural roads. In contrast to other approaches, aerial assessment is proposed by exploring aerial imagery acquired from an unpiloted platform to derive a three‐dimensional (3D) surface model over a road distress area for distress measurement. The system consists of a low‐cost model helicopter equipped with a digital camera, a Global Positioning System (GPS) receiver and an Inertial Navigation System (INS), and a geomagnetic sensor. A set of image processing algorithms has been developed for precise orientation of the acquired images, and generation of 3D road surface models and orthoimages, which allows for accurate measurement of the size and the dimension of the road surface distresses. The developed system has been tested over several test sites with roads of various surface distresses. The experiments show that the system is capable for providing 3D information of surface distresses for road condition assessment. Experiment results demonstrate that the system is very promising and provides high accuracy and reliable results. Evaluation of the system using 2D and 3D models with known dimensions shows that subcentimeter measurement accuracy is readily achieved. The comparison of the derived 3D information with the onsite manual measurements of the road distresses reveals differences of 0.50 cm, demonstrating the potential of the presented system for future practice.     

沥青路面预防性养护时机探讨

从分析沥青路面损坏速度和使用性能的衰变行为出发,说明高速公路沥青路面预防性养护的重要性,并提出在路面破坏由慢转快的转折点处可确定为路面预防性养护时机.     

Fuzzy Representation of Pavement Condition for Efficient Pavement Management

Abstract: Many government agencies and private consulting companies manage large pavement networks in terms of infrastructure condition assessment and maintenance planning. Efficient pavement management is supported by pavement management systems (PMSs), which includes models for pavement condition assessments considered “valuable” by agency's engineers. The objective of this article is to define a pavement condition model able to overcome surveyors’ subjectivity in rating distresses and thus provide meaningful pavement conditions for the agencies to employ in project planning. The article proposes a fuzzy inference model for calculating pavement condition ratio (PCR) specifically tailored on the Alabama Department of Transportation Pavement (ALDOT) guidelines and policies. Applied to several surveyors’ ratings, the proposed model has the ability to smooth distress extent differences among surveyors producing PCR values within acceptable range of variability. The proposed approach has the intention of not only enhancing pavement condition characterization but also to exploit the opportunity made available by automation in the collection and interpretation of pavement data which are anyway characterized by an inherent subjectivity.     

Image-based 3D reconstruction using traditional and UAV datasets for analysis of road pavement distress

On local and urban networks, the enduring issue of scarce resources for Maintenance, Rehabilitation, and Reconstruction strategies (MR&R) has led, in many cases, to using unadjusted or poor techniques for road pavement distress detection and analysis, yielding ineffective or even counterproductive results. Therefore, it is necessary to have tools that can carry out quick, reliable and low-cost assessment surveys. This paper aims at validating the use of innovative and low-cost technologies for road pavement analysis, assessing their potentialities for improving the automation and reliability of distress detection. A Structure from Motion (SfM) technique is analyzed at different altitudes. The technique was applied on a distressed road pavement inside the University Campus in Palermo. The models obtained were compared with a terrestrial laser scanned 3D model to analyze the technique's metric accuracy and reliability. The results have shown that the technique accurately replicates pavement distresses, inciting an integrated approach to optimize pavement management strategies.     

Fuzzy multi-attribute decision making in pavement management

A methodology for ranking pavement sections according to maintenance urgency has been developed using fuzzy multi-attribute decision-making concepts. First. the pavement sections in a highway network are grouped according to their maintenance needs, using fuzzy pavement serviceability indices. Then, the attributes relevant to each category of maintenance are identified, and an expert knowledge base containing priority values for selected combinations of attribute values is formed in collaboration with decisionmakers. Finally, it is shown how the fuzzy attribute values for each pavement section can interact with the expert knowledge base to produce a unique set of rankings for these sections. This ranking scheme is capable of handling possible fuzziness in the priorities assigned by the decision makers.     

Simulating pavement structural condition using artificial neural networks

This article aims to investigate the feasibility of incorporating of an artificial neural network (ANN) as an innovative technique for modelling the pavement structural condition, into pavement management systems. For the development of the ANN, strain assessment criteria are set in order to characterise the structural condition of flexible asphalt pavements with regards to fatigue failure. This initial task is directly followed with the development of an ANN model for the prediction of strains primarily based on in situ field gathered data and not through the usage of synthetic databases. For this purpose, falling weight deflectometer (FWD) measurements were systematically conducted on a highway network, with ground-penetrating radar providing the required pavement thickness data. The FWD data (i.e. deflections) were back-analysed in order to assess strains that would be utilised as output data in the process of developing the ANN model. A paper exercise demonstrates how the developed ANN model combined with the suggested conceptual approach for characterising pavement structural condition with regard to strain assessment could make provisions for pavement management activities, categorising network pavement sections according to the need for maintenance or rehabilitation. Preliminary results indicate that the ANN technique could help assist policy decision makers in deriving optimum strategies for the planning of pavement infrastructure maintenance.     

Pavement Smoothness Prediction Based on Fuzzy and Gray Theories

Abstract: Pavement smoothness has been recognized as one of the measures of pavement performance. In the Mechanistic‐Empirical Pavement Design Guide (MEPDG), pavement smoothness indicated by the International Roughness Index (IRI) was predicted based on various distresses using traditional regression analysis approaches. Recognizing the limitations of linear regression method, a Gray Theory‐based technique was previously proposed by the authors for the development of pavement smoothness prediction models. In this article, instead of using the conventional least squares method to determine the coefficients for gray prediction models, fuzzy regression method is proposed to solve this gray problem. With pavement IRI and distresses data exported from the Long Term Pavement Performance (LTPP) database, Fuzzy and Gray Model (FGM)‐based smoothness predictions are established using influencing factors similar to those in MEPDG. Based on the comparisons among results originated from MEPDG models, conventional GM models, FGM models, and actual LTPP data, it is shown that the Gray Theory‐based prediction methods with fuzzy regression for estimating model coefficients provide promising results and are useful for modeling pavement performance.     

Pavement Distress Analysis Using Image Processing Techniques

Abstract: We demonstrate the feasibility of applying image processing techniques to the analysis of pavement distress due to cracking. Pavement image samples were obtained using a custom-designed data acquisition system called the Automatic Crack Monitor (ACM). The image samples conteining pavement cracks are parameters, are extracted using techniquesw described in this paper. The crack parameters are necessary measures used in calculations of the Pavement Serviceability Index (PSI), which is used by highway maintenance engineers to decide whether a certain pavement section needs to be repaired. Experimental results are shown and the potential harware implementation of the developed techniques is also discussed.     

Nonlinear properties analysis on rutting behaviour of bituminous materials with different air void contents

Rutting is one of the major distresses of flexible pavement. It is defined as the formation of longitudinal depressions along the wheel paths caused by the progressive movement of materials under traffic loading in the asphalt pavement layers or in the underlying base through consolidation or plastic flow. This structural damage has a negative financial impact to the economy. In this study, the rutting behaviour of bituminous materials with different air void contents was investigated. The dynamic cyclic compression testing was carried out to establish nonlinear material models with multiple regression technique. With the specified material models, finite element analysis was carried out to study the rutting behaviour of the wearing course materials with different air void contents in a flexible pavement structure. The simulation result shows that the rutting depth is small at the air void contents of 4.5–8% for wearing course materials. However, for the air void contents above or below this range, the rutting resistance reduces, and thus the rutting depth increases. To verify this simulation result, wheel tracking tests were performed to obtain laboratory data, and the test data was found to be very close to the simulated one. This proved that the developed nonlinear model is applicable to simulate the rutting performance of bituminous mixture and it is a convenient and economical method to be used for the design of bituminous mixtures for both new and rehabilitated pavements.     

沥青路面裂缝填封修补的再破损分析及其防治

通过对再破损的产生原因分析能够掌握其防治与改善途径,针对沥青路面裂缝修补的填封修补再破损,根据其内涵确定了三种表现类型,并具体分析了各自可能产生的原因,结合裂缝填封的受力变形分析,提出了室内试验选材控制要求方面及防治裂缝填封再破损的技术措施。     

Grouping Pavement Condition Variables for Performance Modeling Using Self-Organizing Maps

Different modeling techniques have been employed for the evaluation of pavement performance, determination of structural capacity, and performance predictions. The evaluation of performance involves the functional analysis of pavements based on the history of the riding quality. The riding comfort and pavement performance can be conveniently defined in terms of roughness and pavement distresses. Thus different models have been developed relating roughness with distresses to predict pavement performance. These models are too complex and require parsimonious equations involving fewer variables. Artificial neural networks have been used successfully in the development of performance-prediction models. This article demonstrates the use of an artificial intelligence neural networks self-organizing maps for the grouping of pavement condition variables in developing pavement performance models to evaluate pavement conditions on the basis of pavement distresses.     

A Decision Support System for Flexible Pavement Treatment Selection

The current decision–making process of the Thailand Department of Highways (DOH) for pavement treatment selection is based on the engineering judgment of DOH practitioners, which causes inconsistency and ineffectiveness in treatment selections. In order to address these problems, Thailand Pavement Maintenance Decision Support System (TPMDSS) is developed in this study to help provide an effective guideline for flexible pavement treatment selection. TPMDSS integrates both the database management systems (DBMS) and the treatment decision model. A treatment decision model employs the causebased strategy, which follows a logical progression by synthesizing relevant pavement–related data to help identify pavement problems and recommend feasible treatments. The main components of TPMDSS are: (1) TPMS inventory module, (2) field inspection module, (3) pavement section analysis module, (4) distress explanation facility, and (5) reporting module. The developed system is verified and validated against the experts' judgment and actual test cases.     

An expert system to estimate highway pavement routine maintenance work load

An expert system was developed that can be used to estimate highway pavement routine maintenance needs at a subdistrict level. The knowledge base was prepared by extracting the experience and judgement of unit foremen. The expert system is written in LISP and is interactive in nature. It requires the user to input information about the general features of the highway section and the observed distress data. The package gives explicit recommendations as to the type and amount of activities to be performed along with the information on expected costs.     

Strategies for autonomous robots to inspect pavement distresses

The distress survey is an important task for pavement maintenance and rehabilitation (M&R) activities. As distress surveys require tremendous human resources, many investigators have begun to develop automatic inspection methods with the aim of increasing the efficiency and accuracy of inspections. After assessment of distress surveys on pavements using an autonomous robot (P3-AT), this research aims at developing motion strategies for executing distress surveys using robots under project-level practices. Three motion strategies were specifically developed: (1) Strategy I: random survey (R); (2) Strategy II: random survey with map recording (R + M); (3) Strategy III: random survey with map recording and vision guidance (R + M + V). To validate these three strategies, we developed a test field in a virtual environment. The test field included five distress types, including an alligator crack, a small patching, a pothole, a rectangular manhole and a circular manhole. We also developed a virtual robot to navigate the test field autonomously. The three survey strategies were then implemented by the virtual robot and their performances were compared with the current traffic-directional survey strategy.     

An Expert System to Evaluate Concrete Pavements

"Pavement Expert" is an expert system developed to aid in the evaluation of concrete pavements. The system is operational on an IBM microcomputer. It guides the engineer through the evaluation procedure, ensures that it is carried out adequately, and that it can be independently repeated by other engineers. "Pavement Expert" operates in two modes: dialogue and data rogging modes. The dialogue mode is controlled by a Savoir shell which controls the dialogue between the user and the system. The system makes decisions "intelligently" concerning the length of the section to evaluate, extent and severity of the observed distresses, and calculates and stores the final indices. The data logging mode acts as an intelligent data logger recording specific observed distresses. This information is presented graphically on the screen, allowing for alterations to be made. The system consists of five stages ranging from identifying the road to providing a full report on the general condition of the pavement.     

A Pavement Crack Detection Method Combining 2D with 3D Information Based on Dempster‐Shafer Theory

Pavement cracking is one of the main distresses presented in the road surface. Objective and accurate detection or evaluation for these cracks is an important task in the pavement maintenance and management. In this work, a new pavement crack detection method is proposed by combining two‐dimensional (2D) gray‐scale images and three‐dimensional (3D) laser scanning data based on Dempster‐Shafer (D‐S) theory. In this proposed method, 2D gray‐scale image and 3D laser scanning data are modeled as a mass function in evidence theory, and 2D and 3D detection results for pavement cracks are fused at decision‐making level. The experimental results show that the proposed method takes advantage of the respective merits of 2D images and 3D laser scanning data and therefore improves the pavement crack detection accuracy and reduces recognition error rate compared to 2D image intensity‐based methods.     

Alternative modeling framework for pervious concrete pavement condition analysis

Pervious concrete pavement (PCP) is listed as one of the best stormwater management practices by the Environmental Protection Agency (EPA) due to its capacity to reduce impervious areas, capability to improve water quality, reduce runoff quantity, eliminate the need for detention basins, protect downstream channels, and reduce flooding (save lives and properties). Moreover, it is designated a “Green Asset” since it allows stormwater to seep through pavement to recharge ground water level. PCP may potentially improve skid resistance, reduce noise, decrease heat, and reduce water contamination. However, since PCP must have adequate permeability (porous structure), it can potentially be susceptible to freeze–thaw damage particularly in cold climates. To deploy the PCP in a transportation network, its performance should be thoroughly evaluated. For this purpose, the condition of PCP has to be defined and determined over time. Since PCP has not been widely investigated in cold climates such as Canada, no condition indices have been defined for it. This paper proposes a combined condition index for PCP to ensure that it will perform adequately over time. To develop a combined condition index, both a panel rating approach and objective measurements are employed to assess surface distresses and permeability rates of specific PCP sections. The panel rating method is applied due to the lack of long-term performance data to evaluate PCP sections’ conditions with regards to surface distress (e.g., raveling, spalling, and cracking) and functional performance (i.e., permeability rate). The objective measurements of surface distress and permeability rate (functional performance) of the same PCP sections are performed through the application of the adjusted Ministry of Transportation of Ontario (MTO) protocol and a permeameter, respectively. Regression analysis is used to relate surface objective measurements to the surface panel rating. As a result of statistical analyses, the panel rating method is successfully applied only in surface distress rating. Accordingly, a combined condition index is developed using surface distress ratings and objective functional performance measures.     

The development of a machine vision-assisted,teleoperated pavement crack sealer

Crack sealing is a maintenance technique commonly used to prevent water and debris penetration and reduce future degradation in pavement. The conventional crack sealing operations are, however, dangerous, costly, and labor-intensive. Labor turnover and training are also increasing problems related to crack sealing crews. Automating crack sealing will improve productivity and quality, and offer safety benefits by getting workers off the road. The reduction in crew size and the increase in productivity of the automated sealing process will be translated directly into significant potential cost savings. The main objective of this study is to develop an automated system for sealing cracks in pavement, and to validate the developed system through field trials. A machine vision algorithm, which is composed of noise elimination, crack network mapping and modeling, and path planning, was developed to operate the proposed automated system effectively. Extension of the algorithms and tools presented in the study to other applications is also recommended for future studies.     

A Spatial‐Bayesian Technique for Imputing Pavement Network Repair Data

Abstract: Pavement construction and repair history is necessary for several pavement management functions such as developing pavement condition prediction models and developing maintenance and rehabilitation (M&R) trigger values based on past repair frequencies. It is often difficult to integrate M&R data with condition data since these data are often stored in disparate heterogeneous databases. This article provides a computational technique for estimating construction and M&R history of a pavement network from the spatiotemporal patterns of its condition data. The technique is founded on Bayesian and spatial statistics and searches pavement condition data in groups of adjacent pavement sections for evidence of repair. The developed technique was applied to a pavement network in Texas and has been found to have a 74% precision and a 95% accuracy in estimating repair history data.