Accurate and efficient early detection of pavement distress is crucial, as it prevents further deterioration, reduces repair costs, and enhances driving safety by mitigating hazards like potholes and cracks. However, automatically detecting pavement distress from image samples remains challenging due to complex asphalt textures, including variations in aggregate size, surface wear, and non-distress-related artifacts like shadows and tire marks. This study addresses these challenges by screening and optimizing texture descriptors to characterize local structures and patterns of image regions containing pavement distress. An automated detection system is developed using four categories of texture descriptors: (1) original image-based parameters, (2) multi-resolution analysis-based texture features, (3) local binary pattern (LBP) and their variants-based texture features, and (4) gray-level co-occurrence matrix (GLCM)-based texture features. Unlike previous studies that rely on a single type of descriptor, this integrated approach processes both two-dimensional (2D) intensity images and three-dimensional (3D) pavement surface models, capturing global and local texture patterns to improve robustness and feature discriminability for distress classification. A total of 196 optimal feature descriptors are selected to represent images from both imaging modalities, forming the basis of a distress classification model. A dataset comprising 4 932 samples across 8 distress categories, with moderate class imbalance, is constructed to compare the performance of random forest (RF), gradient boosting decision tree (GBDT), multi-layer perceptron (MLP), and support vector machine (SVM) in distress classification. Class-weighted training and balanced performance metrics were used to mitigate class imbalance effects and ensure robust model evaluation. Among these, the SVM classifier demonstrates the best predictive performance and robustness. Extensive experiments, including feature combination analysis and ablation studies, identify the optimal feature set comprising originatrunl image-based, multi-resolution analysis-based, and LBP variants-based descriptors. This feature set achieves a classification accuracy rate (CAR) of 0.896, an F1-score of 0.894, and an area under receiver operating characteristic curve (AUC) of 0.938 across all distress categories. The selected descriptors effectively differentiate pavement distress types, providing a reliable automated classification system. (c) 2026 Tongji University and Tongji University Press. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Accurate road alignment information is essential for safety analysis and digital infrastructure modeling, and map-based applications. This paper proposes a coarse-to-refined framework for extracting and parameterizing horizontal road alignment from unstructured mobile laser scanning (MLS) point clouds. An adaptive Bezier vectorization method transforms dense, noisy curb points into smooth, continuous geometric representations. A hybrid alignment identification strategy combining geometry-based segmentation and model fitting then identifies and parameterizes straight lines, arcs, and clothoids. The method reduces threshold dependence and enhances robustness. Experiments on real-world data achieve over 99.4% correctness and completeness at a 50 cm buffer width, with multi-scale evaluation across 10-50 cm confirming robustness. Tests on simulated data yield 96.92% mean Intersection over Union (IoU) and parameter errors below 0.401%. Runtime evaluation confirms its efficiency for offline processing, averaging 245 s/km for geometry extraction and 21 s/km for alignment identification on a single-core CPU. The framework provides an effective solution for infrastructure modeling using costeffective MLS systems.

This study aimed to investigate the contributing factors to high-speed railway (HSR) delays and their interdependency in China. A total of 420 records of high-speed railway delays in China were collected, and 15 risk factors related to high-speed railway delays were extracted. Descriptive statistics were used to illustrate the causes of HSR delays in terms of device, personnel and environmental conditions. The association rule mining technique was further applied to explore contributing factors that cause HSR delays, revealing the underlying mechanisms of delay occurrence. The results show that HSR delays of more than 1 hour are most likely caused by foreign objects hanging on the catenary. Moreover, 65% of HSR delays are within the range of 9–30 minutes. Among delay events with durations of 9 to 30 minutes, about 9% of are caused by the fault of on-board train control equipment, mainly the Automatic Train Protection (ATP). Regarding short delays of HSR within 8 minutes, the most likely cause is platform screen door faults, followed by traveller’s misconduct and train door faults. This study offers transportation agencies insights into HSR delay causes and aids in developing policies and engineering measures to reduce delays.

Accurate detection and quantification of pavement potholes are essential for ensuring driving safety, preventing vehicle damage, and supporting intelligent pavement condition assessment. Traditional methods, which rely on image processing techniques, often suffer from low accuracy and fail to consider the structural and three-dimensional (3D) characteristics of potholes. This study proposes an ensemble solution that combines deep learning with an emerging 3D triangulation system for automated pothole detection, segmentation, and multi-dimensional indicator extraction. The YOLOX-PTSNet two-step model is employed for pothole detection and precise segmentation. The Pyramid Transformer Module (PTM) is designed to achieve a global receptive field by replacing pooling operations with pixel block fusion and adaptive dilation rate window division, thereby minimizing information loss. The Pyramid Transformer Segmentation Network (PTSNet), entirely based on Transformer mechanisms, overcomes the limitation of CNNs, whose receptive fields are constrained by convolutional kernel size, and demonstrates strong global feature extraction capabilities and robustness. The method achieved an F1 score of 91.74% and an IoU of 0.847 on the test set, showing superior performance compared to the models tested in this study. Additionally, a calculation model was developed for extracting multi-dimensional pothole indicators using pixel-level segmentation and millimeter-level 3D elevation data through 3D reconstruction. Field validation on roads in Shanghai’s Jiading District demonstrated high accuracy and consistency, with average accuracies for maximum depth, area, and volume exceeding 92%, and R2 values of 0.919, 0.986, and 0.991, respectively. This study provides an integrated sensing-and-quantification framework that enables reliable, high-resolution 3D pothole characterization to support intelligent pavement condition assessment and maintenance planning.

Urban transportation, a vital artery for production and daily life, often suffers severe disruption during extreme weather events. Specifically, coastal cities often face the brunt of extreme weather events, including typhoons, heavy rainfall, and other extreme conditions, which can cause significant damage to urban transportation networks. These events often result in acute traffic congestion, with the distribution of bottleneck road sections differing from those under normal weather conditions. Most existing methodologies for identifying urban traffic bottlenecks predominantly utilize data obtained under normal weather conditions, with scant consideration for analysis under extreme weather scenarios. In this study, we invoke percolation theory from statistical physics, introducing the traffic percolation threshold (qc) as an indicator of network analysis and scrutinizing the traffic connectivity under extreme weather conditions. Our statistical analysis of bottleneck road section distribution reveals that the location distribution of bottleneck roads within the city under extreme weather differs from that under normal weather conditions. Three roads are identified as bottleneck sections under both weather conditions. The method proposed herein offers a valuable reference for urban traffic management and the prevention of traffic paralysis under extreme weather conditions.

Safe lane changing is critical for the automated driving technology. Current research on lane-changing trajectory planning exhibits drawbacks such as inaccurate environmental perception, insufficient system robustness, causing risks in the lane-changing process of autonomous vehicles. This paper proposes a dynamic assessment dual planning framework to enhance lane-change safety. Firstly, the Highway Drone (HighD) dataset is utilised to extract lane-changing events, and the iTransformer model is employed to predict the future trajectories of surrounding vehicles. Secondly, a spatiotemporal risk assessment method is introduced, incorporating accident probability, severity, and interaction intensity to weight inter-vehicle risks. Finally, a dual-planning module integrates risk factors with comfort, safety, and dynamic constraints using a quintic polynomial trajectory model optimized by particle swarm optimization. The model is validated on HighD dataset, showing its effectiveness in reducing lane-change risk and enhancing autonomous driving safety and efficiency.

Accurate identification of pavement types, particularly distinguishing asphalt from rigid concrete, is essential for reliable distress detection and the automation of 3D triangulation systems. However, visual similarities under varying environmental and construction conditions make this task challenging. This study proposes FPTCNet, a multi-feature fusion network for robust pavement type recognition. FPTCNet integrates frequency-domain features via Fourier transform, multi-scale features through wavelet transform, spatial texture patterns using gray-level co-occurrence matrices (GLCM), and implicit features extracted by a ResNet backbone. These heterogeneous descriptors are fused via a multilayer perceptron (MLP) to jointly represent 2D and 3D pavement characteristics. The model was trained and evaluated on a cross-regional dataset of 38,514 images from five Chinese provinces, achieving an F1-score of 0.9763, outperforming classical machine learning methods (0.8604) and advanced deep learning baselines, including Vision Transformer (0.9145) and ResNet (0.9549). Validation on an independent dataset from five additional provinces (2021-2022) confirmed 98.87% accuracy and real-time inference (12.42 ms per image). Typical errors, such as asphalt overlays visually resembling concrete or degraded rural concrete, highlight the importance of combining 2D, 3D, and front-view information. These results demonstrate that multi-feature fusion ensures high accuracy, adaptability, and efficiency, enablingdeployment in large-scale pavement evaluation applications.

This paper presents a coarse-to-refined method for segmenting road curbs from Mobile Laser Scanning (MLS) point clouds, encompassing both organized and unorganized data. The proposed method consists of four parts: point cloud pre-processing, coarse curb segmentation, refined curb segmentation, and global curb verification. Part I aims to reduce the number of irrelevant points and establish appropriate local coordinate systems for subsequent curb segmentation tasks. Part II extracts two-dimensional (2D) projection features to obtain coarse curbs. Part III introduces the universal conic section model and proposes k-neighborhood straight-line features for refined curb segmentation. Part IV develops the curb matching and segmentation rationality judgment rules to establish correspondences between left and right curbs, thereby accurately extracting and distinguishing curbs on both sides of the road. Experimental results demonstrate that the proposed method achieves desirable curb segmentation within a broad range of parameter settings in both typical and complex road scenarios.

Effective distress detection and quantitative analysis play a crucial role in road maintenance and driving safety. The Pavement distress segmentation network (PDSNet) is designed to combine the pyramid scene parsing network (PSPNet) and U-Net, providing both prior global information and local features that can overcome the common detection issues on the pavement data set faced by a single network. This paper proposes an efficient and improved architecture of PDSNet called PDSNet II for enhanced global modeling and retrieving fine details capacities. The proposed PDSNet II represents two major modifications on the original PDSNet. Firstly, a shifted window based on fully connected conditional random fields (FC-CRFs) layer is purposefully introduced to provide connections among consecutive self-attention layers that significantly enhance modeling power. Secondly, PDSNet II adopts multiple-head attention mechanisms to capture diverse interaction information across multiple projection spaces. Consequently, the output maps from the pyramid pooling module (PPM) head and the U-Net tail are fed into a neural window FC-CRFs layer. PDSNet II was trained using a data set consisting of 12,648 two-dimensional (2D) intensity and three-dimensional (3D) range images depicting various pavement conditions. The experimental results demonstrate that PDSNet II outperforms the original PDSNet in terms of F1-score and intersection over union (IoU). Compared with state-of-the-art networks, PDSNet II exhibits superior performance in detecting complex distress patterns, while effectively reducing noise and maintaining robustness. Overall, the proposed PDSNet II framework shows promising results in pavement distress segmentation, highlighting its potential for practical applications.

Road curb extraction is a critical component of road environment perception, being essential for calculating road geometry parameters and ensuring the safe navigation of autonomous vehicles. The existing research primarily focuses on extracting curbs from ordered point clouds, which are constrained by their structure of point cloud organization, making it difficult to apply them to unordered point cloud data and making them susceptible to interference from obstacles. To overcome these limitations, a multi-feature-filtering-based method for curb extraction from unordered point clouds is proposed. This method integrates several techniques, including the grid height difference, normal vectors, clustering, an alpha-shape algorithm based on point cloud density, and the MSAC (M-Estimate Sample Consensus) algorithm for multi-frame fitting. The multi-frame fitting approach addresses the limitations of traditional single-frame methods by fitting the curb contour every five frames, ensuring more accurate contour extraction while preserving local curb features. Based on our self-developed dataset and the Toronto dataset, these methods are integrated to create a robust filter capable of accurately identifying curbs in various complex scenarios. Optimal threshold values were determined through sensitivity analysis and applied to enhance curb extraction performance under diverse conditions. Experimental results demonstrate that the proposed method accurately and comprehensively extracts curb points in different road environments, proving its effectiveness and robustness. Specifically, the average curb segmentation precision, recall, and F1 score values across scenarios A, B (intersections), C (straight road), and scenarios D and E (curved roads and ghosting) are 0.9365, 0.782, and 0.8523, respectively.

Roads are the foundation of intelligent transportation systems, yet cracks are widely present in roads and seriously affect system performance. Cracks not repaired promptly can develop into severe road defects, significantly increasing the risk of traffic accidents. Researchers in the community have started to focus on the automatic sensing of cracks in asphalt pavements, while it is still a challenging task on concrete pavements. Cracks in the concrete pavement are easily recognized as interrupted segments rather than a continuous whole due to the interference of the surface texture. This mistake can seriously mislead the judgment of cracks and subsequent road repair. In this paper, we aim to solve the challenge by enhancing contextual information about cracks within the images. We first extract the information from the local and global representations using image information and then fuse it into complete contextual information by a designed multilayer perceptron. Finally, we use the discriminative loss to constrain the edges of cracks and backgrounds using complete crack contextual information. We have collected and annotated several images of concrete pavements from several significant provinces in China. Experiments show that our method achieves the best performance compared to state-of-the-art methods, especially in edge determination.

Automatic detection and segmentation of potholes and raveling are crucial for preventive maintenance and ensuring roads structural health. However, the extraction of pavement potholes was mainly based on traditional image processing methods, which have proven to be ineffective and inaccurate. Additionally, the absence of a unified pothole and raveling dataset has resulted in the lack of benchmarks for evaluating various methods. This work makes three contributions to address these problems. Firstly, we have curated the pavement pothole and raveling distress detection and segmentation datasets. Secondly, a two-step pavement pothole and raveling detection and segmentation method was proposed. In the initial step, an automated pavement pothole and raveling detection model was developed using the modified YOLOX. Subsequently, the segmentation model, named dual self-attention segmentation network (DSASNet), was proposed to segment distress by extracting mode-sensitive features from intensity and range images using two parallel Twins-SVT self-attention branches. Moreover, we design a mid-fusion module to adaptively fuse mode-specific and scale-specific features. Finally, a pyramid pooling module (PPM) is connected to further enhance the segmentation capability for potholes and raveling of various sizes and shapes. The F1-score and Intersection over union (IoU) of the proposed DSASNet on the test set are 93.65% and 0.881, respectively, outperforming other baseline methods. Furthermore, we conduct an experiment to quantitatively compare the two-step method with the one-step method using only a single semantic segmentation model. The results demonstrated clear advantages of the proposed two-step method in terms of accuracy and efficiency for pavement pothole and raveling segmentation.

Road surface deterioration, such as cracks and potholes, poses a significant threat to both road safety and infrastructure longevity. Swift and accurate detection of these issues is crucial for timely maintenance and user security. However, current techniques often overlook the unique characteristics of pavement images, where the small distressed areas are vastly outnumbered by the background. In response, we propose an innovative road distress classification model that capitalizes on sparse perception. Our method introduces a sparse feature extraction module using dilated convolution, tailored to capture and combine sparse features of different scales from the image. To further enhance our model, we design a specialized loss function rooted in domain-specific knowledge about pavement distress. This loss function enforces sparsity during feature extraction, guiding the model to align precisely with the sparse distribution of target features. We validate the strength and effectiveness of our model through comprehensive evaluations of a diverse dataset of road images containing various distress types and conditions. Our approach exhibits significant potential in advancing traffic safety by enabling more efficient and accurate detection and classification of road distress.

3D road reconstruction plays an essential role in road information extraction; however, the existing 3D point cloud registration methods cannot provide high accuracy and robustness for 3D road reconstruction. We propose a multi-direction registration framework for 3D road reconstruction using vehicle-borne LiDAR and integrated navigation system (INS) data. The framework includes five modules. Module I acquires road point clouds with rough geographical locations. Module II presents a novel key point matching algorithm to reduce the displacement deviation along the travel direction. Module III constructs a dual-plane sliding window to extract curb and pavement point clouds. Module IV develops a plane-to-plane registration algorithm for rough pavement registration. Module V extracts the intersection line between the pavement and curb for curb registration. Experimental results demonstrate that our method can achieve state-of-the-art performance in reconstruction accuracy and technical robustness.

Recently, the rapid development of the bike-sharing system (BSS) has dramatically influenced passengers’ travel modes. However, whether the relationship between the BSS and public transit is competitive or complementary remains unclear. In this paper, a difference-in-differences (DID) model is proposed to figure out the impact of the dockless BSS (DBSS) on bus ridership. The data was collected from Shanghai, China, which includes data from automatic fare collection (AFC) systems, automatic vehicle location (AVL) systems, DBSS transaction data, and point-of-interest (POI) data. The research is based on the route-level, and the results indicate that shared bikes have a substitution impact on bus ridership. Regarding all the travel distance, each shared bike along the route leads to a 0.39 decrease in daily bus ridership on the weekdays, and a 0.17 decrease in daily bus ridership on the weekends, respectively, indicating that dockless shared bikes lead to a stronger decrease in bus ridership on weekends compared to weekdays. Additionally, the substitution effects of shared bikes on bus ridership gradually decays from 0.104 to 0.016 in daily bus ridership on weekends, respectively, with the increase in the travel distance within 0–3 km. This paper reveals that the travel distance of passengers greatly influences the relationship between the DBSS and public transit on the route level.

Automated pavement crack detection is crucial to supporting fine pavement maintenance and ensuring safety for road facilities. Due to the complex pavement condition and crack features, it is still a critical challenge in intelligent pavement surveys. This paper proposed a novel pixel-level pavement crack segmentation network, PCSNet, to provide a solution to this challenge. The network has richer attention and hybrid pyramid structures, which implement full-process crack feature fusion and enhancement. The richer attention module consists of cascaded self-attention and attention gate modules. It captures the crack spatial dependence information and prunes the feature response. The hybrid pyramid structures consist of a multistage convolutional pyramid module and a pyramid pooling module. It integrates contextual information at multiple receptive field scales to enhance the potential crack feature representation. The proposed structure enriches the crack details and optimises the scene parsing on the global geometry of the cracks. A sizeable 3D pavement crack dataset is built for training and testing. The proposed network exhibited the best performance, achieving F1-score, mean intersection of union, and mean pixel accuracy of 81.21%, 77.13%, and 87.17%, respectively. The network can reconstruct the complete crack geometry, preserve the crack edges well, and optimises the detection of shallow and complex cracks. The method exhibits superior and robust performance, facilitating accurate pavement technical condition assessment and maintenance decisions.

The bridgehead settlement problem continues to be one of the most chronic issues affecting long-term bridge performance. In addition, the magnitude of non-staggered-step settlement across the bridge approach transition has not been quantified. Non-contact measurement is considered an alternative to manual inspection, enabling automated damage evaluation for structural maintenance. This paper proposes an inexpensive automatic system using an inertial navigation sensor and a line scanning camera to evaluate the non-staggered-step bridgehead settlement with acceptable accuracy. By analyzing road longitudinal slope data, driving distance, and pavement images, this paper established a calculation model and algorithm of non-staggered-step bridgehead settlement, in which case, a new calculation index named the settlement point ratio (SPR) was proposed. Moreover, the effect of the vehicular detection system and the distance gradient tested at three speeds were measured. The results illustrate that the system has a good performance in longitudinal slope data with an absolute error of less than 1.5%. In addition, 31 bridges in China, Ningbo city, were selected. Combined with the test data, 50 groups of SPR were output using the established model and algorithm. By validating the system's output with the standard measurement method, correlation, and regression analysis were carried out in order to verify the SPR model's reliability. The correlation coefficient is 0.934, and the determination coefficient of the regression model is 0.872, which confirms its capability for accurate data collection and settlement measurement. Therefore, the proposed method is scientific and reasonable for detecting and quantifying non-staggered-step bridgehead settlement, effectively completing the research blank of bridgehead settlement detection.