Road Classification Research Articles

RDD2022: A multi‐national image dataset for automatic road damage detection

AbstractThe data article describes the Road Damage Dataset, RDD2022, encompassing of 47,420 road images from majorly six countries, Japan, India, the Czech Republic, Norway, the United States, and China. The dataset incorporates over 55,000 instances of road damage, specifically longitudinal cracks, transverse cracks, alligator cracks, and potholes. Designed to facilitate the development of deep learning methodologies for automated road damage detection and classification, RDD2022 was unveiled as part of the Crowd sensing‐based Road Damage Detection Challenge (CRDDC'2022), with a major contribution from the challenge winners. This challenge garnered global participation, urging researchers to propose solutions for automatic road damage detection in multiple countries. A noteworthy outcome of CRDDC'2022 was the emergence of a top‐performing model achieving a remarkable F1 Score of 76.9% for road damage detection in all six countries using RDD2022. This success underscores the dataset's practical applicability for municipalities and road agencies, enabling low‐cost, automatic monitoring of road conditions. Beyond its immediate utility, RDD2022 stands as a valuable benchmark for researchers in computer vision, geoscience, and machine learning, offering a rich resource for algorithmic evaluation in diverse image‐based applications, including classification and object detection. The latest big data cup, Optimized Road Damage Detection Challenge (ORDDC'2024), is also based on RDD2022, underscoring its continued relevance and pivotal role in current research and development endeavors.

Stabilization of Fibrous Peat Soils with Addition Palm Shell Ash Waste

South Sumatra Province has a peat area of 1.4 million Ha. The distribution of peat soil in Ogan Ilir is 23,687.91 Ha. Peat soil has a low bearing capacity. This research aims to explain the effect of stabilization on peat soil characteristics. Peat soil samples were taken using the Block Sampling method. The research locations are Parit Village and Lorok Village, Ogan Ilir Regency. Peat soil needs better properties and is unsuitable for foundation soil for civil construction. To overcome this, one method of soil improvement is required, namely the chemical stabilization method: changing the chemical properties of the soil by adding a mixture. The mixture used is palm shell ash waste with variations of 0%, 10%, 15%, 20%, 25%. Soil properties (physical and chemical), SEM, EDS, PTS, and CBR tests were carried out to determine the effect of this mixture. Soil properties test results: water content () in Parit Village 226.39%, and Lorok Village 252.39%. The fiber content (FC) test results for Parit Village were 25.18% and for Lorok Village 28.01%. Peat soil is classified as fibrous peat soil. The CBR value for Parit Village was 4.60%, and Lorok Village was 4.16%. The results of the immersion CBR test showed that Parit Village had a curing period of 7 days, a variation of 5%, namely 4.68%. Look Village obtained the most outstanding results: a curing period of 14 days, a variation of 25%, and 4.76%. The CBR value obtained in this study is 3%-5% (average) when used for subgrade strength with compaction conditions depending on the road category.

Contextualizing urban road network hierarchy and its role for sustainable transport futures: A systematic literature review using bibliometric analysis and content analysis tools

AbstractUrban road networks play a crucial role in transport and urban planning and have the potential to contribute to more sustainable futures if their hierarchy is properly understood. However, the concept of the urban road network hierarchy, which refers to street classification and prioritization, is not well defined within the domain of transport engineering management, leaving many questions unanswered. Is it simply a planning tool, or does it extend to defining the essence of cities? Is it a qualitative or quantitative concept? Does it emerge organically or require proactive planning? Given the lack of comprehensive answers to these questions, this research aims to provide a contextual understanding of the urban road network hierarchy through the lens of sustainable transport futures. To this purpose, we conducted a systematic literature review, which is an effective method for consolidating knowledge on a specific topic. A total of 42 articles were analyzed using both quantitative bibliometric analysis and qualitative content analysis. Our work demonstrates that the road network hierarchy consists of 16 sub-concepts. Four main research trends were identified and discussed: a) road morphology and structure, b) advanced algorithms for street classification, c) integrated street classification planning, and d) the social dimension of street classification. Recent literature indicates a shift toward alternative road network hierarchy approaches that prioritize sustainable mobility over car-centric models. In conclusion, our analysis reveals that the urban road network hierarchy is a multifaceted yet under researched “vehicle for change,” which, if utilized effectively, offers opportunities to reimagine urban road environments.

Exploring road safety using alignment perspective features in real driving images: A case study on mountain freeways.

While driving, drivers frequently adapt their driving behaviors according to their perception of the road's alignment features. However, traditional two-dimensional alignment methods lack the ability to capture these features from the driver's perspective. This study introduces a novel method for road alignment recognition, employing image recognition technology to extract alignment perspective features, namely alignment perspective skewness (APS) and alignment perspective kurtosis (APK), from in-real driving images. Subsequently, the K-means clustering algorithm is utilized for road segment classification based on APS and APK indicators. Various sliding step length for clustering are employed, with step length ranging from 100m to 400m. Furthermore, the accident rates for different segment clusters are analyzed to explore the relationship between alignment perspective features and traffic safety. A 150 km mountain road section of the Erlianhaote-Guangzhou freewway from Huaiji to Sihui is selected as a case study. The results demonstrate that using alignment perspective features as classification criteria produces favorable clustering outcomes, with superior clustering performance achieved using shorter segment lengths and fewer cluster centers. The road segment classification based on alignment perspective features reveals notable differences in accident rates across categories; while traditional two-dimensional parameters-based classification methods fail to capture these differences. The most significant differences in accident rates across categories are observed with segment length of 100m, with the significance gradually diminishing as segment length increases and disappearing entirely when the length exceeds 300m. These findings validate the reliability of using alignment perspective features (APS and APK) for road alignment classification and road safety analysis, providing valuable insights for road safety management.

On the Structure of Set-Theoretic Polygon Equations

Polygon equations generalize the prominent pentagon equation in very much the same way as simplex equations generalize the famous Yang-Baxter equation. In particular, they appeared as ''cocycle equations'' in Street's category theory associated with oriented simplices. Whereas the $(N-1)$-simplex equation can be regarded as a realization of the higher Bruhat order $B(N,N-2)$, the $N$-gon equation is a realization of the higher Tamari order $T(N,N-2)$. The latter and its dual $\tilde T(N,N-2)$, associated with which is the dual $N$-gon equation, have been shown to arise as suborders of $B(N,N-2)$ via a ''three-color decomposition''. There are two different reductions of $T(N,N-2)$ and $\tilde T(N,N-2)$, to ${T(N-1,N-3)}$, respectively $\tilde T(N-1,N-3)$. In this work, we explore the corresponding reductions of (dual) polygon equations, which lead to relations between solutions of neighboring (dual) polygon equations. We also elaborate (dual) polygon equations in this respect explicitly up to the octagon equation.

Detection and Classification of Lanes Using Reannotated Lane Dataset to Aid Autonomous Driving

The core functionality of advanced driver assistance systems and self-driving cars depends on the ability to recognize drivable road areas. As there are many different categories of useful markings incorporated within the road area, lane detection and classification form a major step for taking appropriate actions leading to truly autonomous driving. Existing datasets do not provide ample classification of lane types and adequate granularity for precise localization of lane markings. A new dedicated dataset for semantic segmentation of 11 varied lane types obtained by reannotating the BDD100K dataset is presented. The reannotation process involves pixel-level lane markings of 76,000 image instances of the BDD100K dataset, which were originally represented as a sequence of coordinate points. This opens up the possibility of high resolution both spatially and semantically in the context of lane understanding for autonomous driving. Baseline results on the proposed dataset based on the Bilateral Segmentation Network (BiSeNetV2), which considers the spatial data and the categorical semantics distinctly, are presented. The performance is pinned at 85% accuracy during testing using BiSeNetV2 architecture. The dataset is expected to open up new directions of research to address problems such as severe class imbalance, segmentation of multiple classes of texture-less and eccentric features (lanes), and so forth. As a result, applications such as lane-centric activity interpretation, future event prediction, and continuous learning are expected.

Exploring the relationship between the International Roughness Index (IRI) and the ISO 8608 roughness coefficient (Gd (n 0))

ABSTRACT Globally, diverse agencies employ varied metrics to gauge road roughness, complicating the exchange of information for pavement maintenance. This study addresses this challenge by establishing a theoretical connection between the International Roughness Index (IRI) and the ISO 8608 Power Spectral Density (PSD)-based roughness coefficient, namely Gd (n 0). Based on random vibration theory, this paper first deduces a closed-form explicit solution for IRI related to Gd (n 0), solidifying a theoretical link. Next, the variability and convergence of the proposed solution are examined through Monte Carlo simulations using a quarter-car vehicle model with the Golden Car parameters. Then, we compare computed results from this study with those existing solutions from past studies, verifying that the proposed solution captures the curve trend well and agrees with the average value of existing solutions based on field measurement data. Finally, this paper proposes the IRI values corresponding to various ISO road class thresholds and a new Gd (n 0) scale for ride quality assessment of highways.

Evaluating the influence of volumetric properties on back-calculated asphalt layer moduli using falling weight deflectometer data

The back-calculation process performed in pavement systems is the numerical analysis of captured deflections to estimate layer stiffness parameters. Prediction of fatigue performance in terms of back-calculated asphalt layer moduli by using a Falling Weight Deflectometer (FWD) has specific challenges in terms of testing protocol, skillset, and complex back-calculation analysis. The performance of the asphalt layer is primarily governed by extrinsic parameters such as temperature, vehicular transient loading characteristics, moisture content, and intrinsic parameters such as binder properties and aggregate mix properties. The role of volumetric properties of asphalt mixes contributes significantly to the back-calculated asphalt layer moduli in terms of the overall life of the structure. The asphalt layer moduli value is dependent on the traditional volumetric properties of asphalt mixes such as air voids in the mix, voids in mineral aggregate, and the percentage of bitumen mix. In this study, a total of 60 in-service pavement sections are identified from three different categories of roads to perform FWD tests and collection of asphalt layer core samples. A detailed laboratory investigation has been carried out to estimate the volumetric properties of different core samples. This study uses field investigations to determine the degree of interdependency between the volumetric characteristics of asphalt mixtures and temperature on the back-calculated layer moduli. Furthermore, the findings from this study are utilized to establish several correlations at the aggregate level, demonstrating strong relationships with R2 values ranging from 0.84 to 0.875. The developed model is validated and depicted in good agreement with the actual values.

Joint effects of traffic-related air pollution and hypertensive disorders of pregnancy on maternal postpartum depressive and anxiety symptoms.

Ambient air pollution has been linked to postpartum depression. However, few studies have investigated the effects of traffic-related NOx on postpartum depression and whether any pregnancy-related factors might increase susceptibility. To evaluate the association between traffic-related NOx and postpartum depressive and anxiety symptoms, and effect modification by pregnancy-related hypertension. This study included 453 predominantly low-income Hispanic/Latina women in the MADRES cohort. Daily traffic-related NOx concentrations by road class were estimated using the California LINE-source dispersion model (CALINE4) at participants' residential locations and averaged across pregnancy. Postpartum depressive and anxiety symptoms were evaluated by a validated questionnaire (Postpartum Distress Measure, PDM) at 1, 3, 6 and 12 months postpartum. Multivariate linear regressions were performed to estimate the associations at each timepoint. Interaction terms were added to the linear models to assess effect modification by hypertensive disorders of pregnancy (HDPs). Repeated measurement analyses were conducted by using mixed effect models. We found prenatal traffic-related NOx was associated with increased PDM scores. Specifically, mothers exposed to an IQR (0.22 ppb) increase in NOx from major roads had 3.78% (95% CI: 0.53-7.14%) and 5.27% (95% CI: 0.33-10.45%) significantly higher 3-month and 12-month PDM scores, respectively. Similarly, in repeated measurement analyses, higher NOx from major roads was associated with 3.06% (95% CI: 0.43-5.76%) significantly higher PDM scores across the first year postpartum. Effect modification by HDPs was observed: higher freeway/highway and total NOx among mothers with HDPs were associated with significantly higher PDM scores at 12 months postpartum compared to those without HDPs. This study shows that prenatal traffic-related air pollution was associated with postpartum depressive and anxiety symptoms. The study also found novel evidence of greater susceptibility among women with HDPs, which advances the understanding of the relationships between air pollution, maternal cardiometabolic health during pregnancy and postpartum mental health. Our study has potential implications for clinical intervention to mitigate the effects of traffic-related pollution on postpartum mental health disorders. The findings can also offer valuable insights into urban planning strategies concerning the implementation of emission control measures and the creation of green spaces.

Road Accident Analysis and Classification Using Machine Learning Algorithms

Abstract: India is facing a severe public health crisis characterized by a significant number of road accidents resulting in fatalities and life-altering injuries. This study undertakes a detailed analysis of road accidents in India using multi-year accident data to unveil recurring patterns and identify high-risk demographics. The research meticulously examines the primary causes of these accidents, including driver behavior such as speeding, and inadequate road infrastructure such as poor lighting, signage, potholes, and vehicle malfunctions. Moreover, the paper proposes a comprehensive approach to prevent these tragedies. It suggests strategies for enhancing road design and maintenance, enforcing traffic regulations more rigorously, and conducting targeted public awareness campaigns to promote responsible driving practices. By highlighting the alarming increase in fatalities and the substantial socio-economic costs associated with these accidents, this research emphasizes the critical need for immediate and effective interventions to establish a safer road environment for all citizens in India.

Examining the relationship between land use and childhood leukemia and lymphoma in Tehran

We conducted a hospital-based case–control study to explore the association between proximity to various land use types and childhood leukemia and lymphoma. This research involved 428 cases of childhood leukemia and lymphoma (2016–2021), along with a control group of 428 children aged 1–15 in Tehran. We analyzed the risk of childhood cancer associated with land use by employing logistic regression adjusted for confounding factors such as parental smoking and family history. The odds ratio (OR) for children with leukemia and lymphoma residing within 100 m of the nearest highway was 1.87 (95% CI = 1.00–3.49) and 1.71 (95% CI = 1.00–2.93), respectively, in comparison to those living at a distance of 1000 m or more from a highway. The OR for leukemia with exposure to petrol stations within 100 m was 2.15 (95% CI = 1.00–4.63), and for lymphoma it was 1.09 (95% CI = 0.47–2.50). A significant association was observed near power lines (OR = 3.05; 95% CI = 0.97–9.55) within < 100 m for leukemia. However, no significant association was observed between power lines and the incidence of childhood lymphoma. There was no association between bus stations, major road class 2, and the incidence of childhood leukemia and lymphoma. In conclusion, our results suggest a possible association between the incidence of childhood leukemia and proximity to different urban land uses (i.e., highways and petrol stations). This study is the first step in understanding how urban land use affects childhood leukemia and lymphoma in Tehran. However, comprehensive studies considering individual-level data and specific pollutants are essential for a more nuanced understanding of these associations.

Road network model in Thailand for highly efficient transportation

This research presents a comprehensive model for enhancing the road network in Thailand to achieve high efficiency in transportation. The objective is to develop a systematic approach for categorizing roads that aligns with usage demands and responsible agencies. This alignment facilitates the creation of interconnected routes, which ensure clear responsibility demarcation and foster efficient budget allocation for road maintenance. The findings suggest that a well-structured road network, combined with advanced information and communication technology, can significantly enhance the economic competitiveness of Thailand. This model not only proposes a framework for effective road classification but also outlines strategic initiatives for leveraging technology to achieve transportation efficiency and safety.

Evaluation of data representation techniques for vibration based road surface condition classification

The accurate classification of road surface conditions plays a vital role in ensuring road safety and effective maintenance. Vibration-based techniques have shown promise in this domain, leveraging the unique vibration signatures generated by vehicles to identify different road conditions. In this study, we focus on utilizing vehicle-mounted vibration sensors to collect road surface vibrations and comparing various data representation techniques for classifying road surface conditions into four classes: normal road surface, potholes, bad road surface, and speedbumps. Our experimental results reveal that the combination of multiple data representation techniques results in higher performance, with an average accuracy of 93.4%. This suggests that the integration of deep neural networks and signal processing techniques can produce a high-level representation better suited for challenging multivariate time series classification issues.

Driving Domain Classification Based on Kernel Density Estimation of Urban Land Use and Road Network Scaling Models

Current research on automated driving systems focuses on Level 4 automated driving (AD) in specific operational design Domains (ODD). Measurement data from customer fleet operation are commonly used to extract scenarios and ODD features (road infrastructure, etc.) for the testing of AD functions. To ensure data relevance for the vehicle use case, driving domain classification of the data is required. Generally, classification into urban, extra-urban and highway domains provides data with similar ODD features. Highway classification can be implemented using global navigation satellite system coordinates of the driving route, map-matching algorithms, and road classes stored in digital maps. However, the distinction between urban and extra-urban driving domains is more complex, as settlement taxonomies and administrative-level hierarchies are not globally consistent. Therefore, this paper presents a map-based method for driving domain classification. First, potential urban areas (PUA) are identified based on urban land-use density, which is determined based on land-use categories from OpenStreetMap (OSM) and then spatially smoothed by kernel density estimation. Subsequently, two road network scaling models are used to distinguish between urban and extra-urban domains for the PUA. Finally, statistics of ODD feature distribution are analysed for the classified urban and extra-urban areas.

Multi‐object road waste detection and classification based on binocular vision

AbstractA road multi‐object detection algorithm is one of the core algorithms for intelligent road cleaning robots relying on machine vision. Most existing object detection algorithms analyse all image regions and finally calculate the category and location of each object. However, it is not necessary to analyse all areas of the image when detecting objects on the road surface where the background changes little, and the number of objects is small because there will be a lot of invalid calculations. If we can perform targeted local analysis on images instead of analysing all image regions, it will improve the detection efficiency. Therefore, this paper proposes a multi‐object detection method using a binocular camera and a convolutional neural network (CNN) that effectively reduces invalid calculations during the detection and improves detection efficiency. In the developed method, the binocular vision image acquired by the binocular camera is stereo matched and equalized, while linear regression and coordinate transformation eliminate the angle of the camera pair concerning the road surface. Then, the coordinates of the regions of interest (ROI) is calculated in the left vision image and the features within the ROI is extracted from the corresponding CNN's feature map. Next, ROI pooling resizes the extracted feature maps of different sizes to the same size, which are then input to the fully connected layers to output the results. The proposed binocular network and faster R‐CNN (VGG16) are trained and tested on a dataset involving 1000 road waste images. The experimental results demonstrate that the developed binocular network improves the detection accuracy and speed by 28.56% and 78.39%, respectively, compared with faster R‐CNN (VGG16), providing a reliable basis for a machine vision‐based intelligent road cleaning robot.

A new optimization approach based on neural architecture search to enhance deep U-Net for efficient road segmentation

Neural Architecture Search (NAS) has significantly improved the accuracy of image classification and segmentation. However, these methods concentrate on finding segmentation structures for natural or medical applications. In this study, we introduce a NAS approach based on gradient optimization to identify ideal cell designs for road segmentation. To the best of our knowledge, this work represents the first application of gradient-based NAS to road extraction. Taking insight from the U-Net model and its successful variations in different image segmentation tasks, we propose NAS-enhanced U-Net, illustrated by an equal number of cells in both encoder and decoder levels. While cross-entropy combined with dice loss is commonly used in many segmentation tasks, road extraction brings up a unique challenge due to class imbalance. To address this, we introduce a combination of loss function. This function merges cross-entropy with weighted Dice loss, focusing on elevating the importance of the road class by assigning it a weight (⍵), while background Dice values are disregarded. The results indicate that the optimal weight for the proposed model equals 2. Additionally, our work challenges the assumption that increased model parameters or depth inherently leads to improved performance. Therefore, we establish search spaces 2,3,4,5,6,7 and 8 to automatically choose the optimal depth for model. We present promising segmentation results for our proposed method, achieved without any pretraining on the Massachusetts road dataset. Furthermore, these results are compared with those of 14 models categorized into four groups: U-Net, Segnet, FCN8, and Nas-U-Net.

An Approach of Multi Criteria Decision Making in Analyzing Rural Road Safety in Nashik District

Abstract: Rural road safety evaluations play a significant role in the transportation industry, and most fatal accidents happen on these types of roads. Road Safety Analysis (RSA), a process to evaluate a road's safety criteria and support overall road management decision-making, ensures the avoidance of loss of life and property damage. The "Pradhan Mantri Gram Sadak Yojana," which is primarily utilized in rural areas, is the main emphasis of this. In order to identify the most vulnerable roadways and implement mitigation measures, roads must be prioritized depending on safety ratings. This document presents the results of a study conducted to rank the safety criteria of Pradhan Mantri Gram Sadak Yojana (PMGSY) roads, a particular category of rural roads. A questionnaire survey was used for the study's analysis, and three MCDM methods were used to examine the relationships and interactions between the various factors. The study's findings indicated that the based on input from PMGSY engineers and contractors. From the perspective of the engineers in the Nashik area, we can determine the safety impact factor that influences the safety assessment of the PMGSY route.

Use of linear features by mammal predators and prey in managed boreal forests

In managed boreal forests, logging operations maintain high levels of anthropogenic disturbance in the ecosystem. The establishment of permanent anthropogenic linear features such as logging roads in the landscape may be a major factor in the predator-prey system. Logging roads may potentially improve the numerical and functional response of predators. Using camera traps, our objective was to explain according to local and landscape factors how the number of uses by wolves, black bears, lynx and moose, varies along different natural and anthropogenic linear features during the snow-free season. In western Quebec (Canada), the managed forest south of Val-d’Or encloses an isolated caribou population facing extinction that requires active restoration of their habitat. In this site, we used stratified random selection of gravel forest roads (n = 33), winter forest roads (n = 28) and riparian areas (n = 19) to compare their characteristics and number of uses by the four species. For wolves, black bears, and lynx, positive differences in lateral cover between the surroundings and the linear feature mainly explained their number of uses. Number of uses by wolves and lynx were positively related to use by their respective prey species (moose and snowshoe hare). Gray wolf use was also positively affected by distance to a higher forest road class and negatively affected by distance to the nearest urban area. Gravel forest roads had the highest number of uses by all species, as they showed greater positive differences in lateral cover as compared to the surroundings area due to their limited vegetation growth and by frequent maintenance activities. We recommend that restoration efforts aimed at forest road closures should target roads with a high value of difference in lateral cover, which is particularly the case in most gravel roads. Lower lateral cover on these linear feature as compared to their surroundings area favors the movement of predators and alternative prey. Our results thus suggest that investing in gravel roads restoration can benefit conservation efforts in caribou habitat.

Transfer Function-Based Road Classification for Vehicles with Nonlinear Semi-Active Suspension

Ride comfort and handling are two important criteria regarding vehicle vibration control. For solving the inconsistency between ride comfort and handling, a semi-active suspension system equipped with a road classification system can be a suitable solution. Because the road condition varies during driving, the control gain of the semi-active suspension system should be adaptively changed according to the road level. In this paper, accelerometer data and a transfer function scheme will be used for road classification, and there is no need to measure the road directly with difficult and often expensive methods. In this approach, a transfer function that makes relevant the Power Spectral Density (PSD) of the road surface and the PSD of car acceleration is used. Road classification is investigated for a vehicle with a nonlinear, semi-active suspension system equipped with Continuous Damping Control (CDC) and nonlinear springs. To show the applicability of the proposed method in scenarios close to real situations, robustness analysis is done by considering vehicle model uncertainties and sensor noise. The simulation results show that the proposed method is robust against typical uncertainties and accelerometer noise and can classify the road level, which is used to tune the parameters of the nonlinear, semi-active suspension system.

Safety Assessment of Suburban-Type Arterial Roadways: New Findings Using Heterogeneity Models

This study investigates the safety performance of suburban-type roads (STRs), a category of medium-speed arterial roads characterized by high vehicular traffic and limited pedestrian and cyclist mobility. Such roadways present unique challenges in road safety, often overlooked in traditional road design focused predominantly on driver needs. By employing a correlated random parameter ordered probit model to analyze police-reported crash severity data from STRs, this research uncovers critical insights into the factors influencing crash injury severities on these roads. Our findings highlight the nuanced impact of various dimensions, such as road design, traffic volume, and environmental conditions, on STR safety; the results offer a detailed understanding of crash contexts on STRs and point toward targeted interventions for enhancing road safety. These interventions include the potential redesign of roadways to accommodate diverse user needs and the implementation of speed regulation measures tailored to specific road characteristics. This research contributes significantly to the existing body of knowledge on suburban road safety, providing evidence-based recommendations for improving the overall safety of these increasingly prevalent road types. The implications of these findings are far-reaching, offering a foundation for future studies to explore diverse methodologies in road safety analysis and to extend this research to other geographic contexts.