Road Inspection Research Articles

Exploring the Effects of Attention Mechanisms on Road Crack Detection and Classification

Pavement cracks are a prevalent form of roadway distress that pose significant safety hazards, necessitating prompt detection and repair. Due to the extensive road network, traditional image processing-based crack detection methods exhibit limitations in recognition accuracy and speed. To address these challenges, this paper proposes an efficient pavement crack detection model based on YOLOv8, termed YOLOv8-ACD (YOLOv8 - Attention for Cracks Detecting), which integrates a global attention mechanism. YOLOv8-ACD enhances detection efficiency and accuracy by focusing on crack identification while filtering out most irrelevant information. We evaluated YOLOv8-ACD on the RDD2022 dataset, and experimental results demonstrate significant improvements in key performance metrics, such as F1-Score and mean Average Precision (mAP), compared to the original YOLOv8 and other mainstream models. The real-time processing capability of this model makes it suitable for practical road inspection and maintenance, effectively reducing the workload of maintenance personnel and enhancing roadway safety.

DCUFormer: Enhancing pavement crack segmentation in complex environments with dual-cross/upsampling attention

Efficient road inspection and maintenance are essential to extend pavement lifespan and enhance safety. However, automated crack detection remains challenging due to varied environmental conditions and differences in image collection equipment, making robust algorithm development a critical need. Vision Transformers, with their capacity to capture long-range dependencies, offer significant advantages for crack detection in complex scenarios by effectively extracting global features. Nevertheless, existing Transformer-based methods encounter difficulties in boundary delineation due to decoder design limitations, which lead to suboptimal fusion of low-level and high-level features. To address this issue, we propose a comprehensive approach that integrates semantic preservation, detail refinement, and detail delineation. These concepts are realized through our novel Dual-Cross Attention Module (DCA) and Upsampling Attention Module (UA). The DCA module progressively filters redundant details from low-level feature layers using high-level semantic information, while preserving boundary details to refine high-level feature boundaries. In addition, the UA module employs progressive local cross-attention in upsampling, facilitating more precise boundary definitions and surpassing conventional dynamic upsampling methods. Our approach, utilizing both lightweight (MiT-B0, LVT) and middle-weight (Swin-T) backbones, demonstrates state-of-the-art performance on three diverse datasets—Crack500, CrackSC, and UAV-Crack500—highlighting its robustness across varied conditions. This work contributes to advancing Transformer-based architectures for defect segmentation in complex engineering contexts, underscoring the critical role of improved feature fusion in crack detection. The code is available at: https://github.com/SHAN-JH/DCUFormer.

Комплексне оцінювання стану дорожнього покриву

Introduction. This article deals with the issue of reviewing existing methods and approaches to assessing the condition of the road pavement and proposes a methodology for assessing the condition of the road pavement using a complex indicator. This indicator is used to analyze the defects and destruction of the pavement determined by the visual method. Problem. The need to improve the methodology for assessing the condition of the road surface (rigid, non-rigid) by calculating the coefficient by which the designer or balance holder can assess the condition of the existing pavement. The purpose of the article. To inform road industry specialists about new trends and developments in the field of road pavement inspection of roads and airfields.

Vehicle Localization Using Crowdsourced Data Collected on Urban Roads.

Vehicle localization using mounted sensors is an essential technology for various applications, including autonomous vehicles and road mapping. Achieving high positioning accuracy through the fusion of low-cost sensors is a topic of considerable interest. Recently, applications based on crowdsourced data from a large number of vehicles have received significant attention. Equipping standard vehicles with low-cost onboard sensors offers the advantage of collecting data from multiple drives over extensive road networks at a low operational cost. These vehicle trajectories and road observations can be utilized for traffic surveys, road inspections, and mapping. However, data obtained from low-cost devices are likely to be highly inaccurate. On urban roads, unlike highways, complex road structures and GNSS signal obstructions caused by buildings are common. This study proposes a reliable vehicle localization method using a large amount of crowdsourced data collected from urban roads. The proposed localization method is designed with consideration for the high inaccuracy of the data, the complexity of road structures, and the partial use of high-definition (HD) maps that account for environmental changes. The high inaccuracy of sensor data affects the reliability of localization. Therefore, the proposed method includes a reliability assessment of the localized vehicle poses. The performance of the proposed method was evaluated using data collected from buses operating in Seoul, Korea. The data used for the evaluation were collected 18 months after the creation of the HD maps.

Autonomous Crack Detection for Mountainous Roads Using UAV Inspection System.

Road cracks significantly affect the serviceability and safety of roadways, especially in mountainous terrain. Traditional inspection methods, such as manual detection, are excessively time-consuming, labor-intensive, and inefficient. Additionally, multi-function detection vehicles equipped with diverse sensors are costly and unsuitable for mountainous roads, primarily because of the challenging terrain conditions characterized by frequent bends in the road. To address these challenges, this study proposes a customized Unmanned Aerial Vehicle (UAV) inspection system designed for automatic crack detection. This system focuses on enhancing autonomous capabilities in mountainous terrains by incorporating embedded algorithms for route planning, autonomous navigation, and automatic crack detection. The slide window method (SWM) is proposed to enhance the autonomous navigation of UAV flights by generating path planning on mountainous roads. This method compensates for GPS/IMU positioning errors, particularly in GPS-denied or GPS-drift scenarios. Moreover, the improved MRC-YOLOv8 algorithm is presented to conduct autonomous crack detection from UAV imagery in an on/offboard module. To validate the performance of our UAV inspection system, we conducted multiple experiments to evaluate its accuracy, robustness, and efficiency. The results of the experiments on automatic navigation demonstrate that our fusion method, in conjunction with SWM, effectively enables real-time route planning in GPS-denied mountainous terrains. The proposed system displays an average localization drift of 2.75% and a per-point local scanning error of 0.33 m over a distance of 1.5 km. Moreover, the experimental results on the road crack detection reveal that the MRC-YOLOv8 algorithm achieves an F1-Score of 87.4% and a mAP of 92.3%, thus surpassing other state-of-the-art models like YOLOv5s, YOLOv8n, and YOLOv9 by 1.2%, 1.3%, and 3.0% in terms of mAP, respectively. Furthermore, the parameters of the MRC-YOLOv8 algorithm indicate a volume reduction of 0.19(×106) compared to the original YOLOv8 model, thus enhancing its lightweight nature. The UAV inspection system proposed in this study serves as a valuable tool and technological guidance for the routine inspection of mountainous roads.

Detecting and Classification of Complex Roads Defects Using a Deep Learning Approach: A survey

Road safety is one of the most important conditions for public driving safety around the world. This is done by monitoring the roads periodically to detect defects that can appear due to several factors affecting the asphalt, such as weather and accidents. Traditional road inspection methods are often time-consuming, labor-intensive and expensive due to the length of the methods. Deep learning (DL) has emerged as a promising technology for automated detection and classification of road defects, offering the potential for faster, more accurate, and less expensive inspections. This review paper provides a comprehensive overview of the state-of-the-art methods working on the principle of artificial intelligence for detecting defects on road surfaces. We will also discuss the different types of defect monitoring methods that differ from one method to another and furthermore, we will show the different datasets from oldest to newest with all their details that can be useful in DL. A review is then performed of various blind learning architectures and algorithms that have been used for defect detection and classification, incorporating generative adversarial networks (GANs), recurrent neural networks (RNNs), and convolutional neural networks (CNNs). Finally, we will provide a comparative analysis of the reviewed algorithms and methods by presenting the metrics used to choose the best technique, which is what these paper aimsto do.

ПАСПОРТИЗАЦИЯ АВТОМОБИЛЬНЫХ ДОРОГ С ИСПОЛЬЗОВАНИЕМ ЦИФРОВЫХ ТЕХНОЛОГИЙ И ДВОЙНИКОВ

The large scale of the Russian road sector requires the use of digital technologies, the creation of digital twins to improve the quality of management, while limited resources should be taken into account. The purpose of the study is to develop and test an approach to the certification of highways based on the creation of digital twins, taking into account Russian regulatory requirements. In this work, digital twins of roads were obtained based on photographic data and laser scanning using a mobile road laboratory. The quality, accuracy, and completeness of digital twin information meet a Russian regulatory requirement, which makes it possible to generate technical passports for registering objects for cadastral registration. The use of digital twins, compared to field inspection of roads, has a 1.5-2.0 times lower cost and reduces the time required for completing work by 2.5-3.0 times.

Raster-Based Point Cloud Mapping of Defective Road Marking: Toward Automated Road Inspection via Airborne LiDAR

Raster-Based Point Cloud Mapping of Defective Road Marking: Toward Automated Road Inspection via Airborne LiDAR

Roadside inspections, traffic events and traffic accidents from 2010–2022 in the experience of the Institute of Forensic Research as a consulting body in light of the proposed thresholds for the states of ‘after use’ and ‘under the influence’ for THC = Kontrole, zdarzenia i wypadki drogowe z lat 2010–2022 w praktyce opiniodawczej Instytutu Ekspertyz Sądowych w świetle propozycji progów „po

Road traffic safety depends largely on the psychophysical fitness of drivers. Apart from ethyl alcohol, this efficiency is influenced by many other factors, including: presence in the blood of the so-called substances that act similarly to alcohol, and cannabinols constitute one of the most popular groups of these substances. The paper presents statistics on delta-9-tetrahydrocannabinol (THC) – the main psychoactive ingredient of cannabis detected in the blood of drivers subjected to 7269 road inspections and involved in 4697 road events and accidents in the years 2010–2022 throughout Poland. Based on the results obtained, it was found that THC is present in the blood of 25–61% of roadside inspection cases, and in the case of road incidents and accidents, 5–15% of drivers have THC in a concentration equal to or higher than 1 ng/ml. The data in the charts are given in relation to two THC concentration ranges: 1.0–2.5 ng/ml and above 2.5 ng/ml (the value proposed as the limit of the state “after use” and “under the influence” of THC) and in relation to concentration ranges based on the THC elimination curve in human serum. The study did not distinguish between chronic or acute participants and occasional participants. It was based on THC and THCCOOH concentrations. The analyzes were performed using the enzyme-linked immunosorbent method (ELISA) and confirmatory methods such as liquid and gas chromatography coupled with mass spectrometry, in which the limit of quantification of 1 ng/ml was in accordance with the applicable requirement contained in the Regulation of the Minister of Health of July 16, 2014 on the list of similarly acting agents to alcohol and the conditions and method of conducting tests for their presence in the body.

Roadway Inspection System

Traditional road inspections are manual processes, prone to human error and inefficiencies. This paper presents a novel approach for automated roadway inspection using a Convolutional Neural Network (CNN) model. Our system leverages computer vision techniques to detect potholes and speed breakers on road surfaces from images. We developed a CNN model trained on a comprehensive dataset of road images containing various pothole and speed breaker types, lighting conditions, and road backgrounds. The model achieved an accuracy of 93% in detecting these road defects, demonstrating the effectiveness of deep learning for automated roadway inspections. This system has the potential to significantly improve the efficiency and objectivity of road inspections, leading to faster repairs and improved road safety

Automatic Detection of Pavement Marking Defects in Road Inspection Images Using Deep Learning

Automatic Detection of Pavement Marking Defects in Road Inspection Images Using Deep Learning

Integrative Approach for High-Speed Road Surface Monitoring: A Convergence of Robotics, Edge Computing, and Advanced Object Detection

To ensure precise and real-time perception of high-speed roadway conditions and minimize the potential threats to traffic safety posed by road debris and defects, this study designed a real-time monitoring and early warning system for high-speed road surface anomalies. Initially, an autonomous mobile intelligent road inspection robot, mountable on highway guardrails, along with a corresponding cloud-based warning platform, was developed. Subsequently, an enhanced target detection algorithm, YOLOv5s-L-OTA, was proposed. Incorporating GSConv for lightweight improvements to standard convolutions and employing the optimal transport assignment for object detection (OTA) strategy, the algorithm’s robustness in multi-object label assignment was enhanced, significantly improving both model accuracy and processing speed. Ultimately, this refined algorithm was deployed on the intelligent inspection robot and validated in real-road environments. The experimental results demonstrated the algorithm’s effectiveness, significantly boosting the capability for real-time, precise detection of high-speed road surface anomalies, thereby ensuring highway safety and substantially reducing the risk of liability disputes and personal injuries.

Development of a Network-Level Road Safety Assessment Procedure Based on Human Factors Principles

Road safety is a central issue in the management and development of a road network. Road agencies must try to identify the most dangerous sections of their network and act on them to improve safety. The most used procedure for this purpose is about considering the indicators based on crashes. However, a mature road safety management system must be able to assess the safety of a road section before accidents occur. The European community is moving in this direction with the update of Directive 2008/96/EC (Directive 1936/2019). This paper proposes a new methodology for carrying out a network-wide road safety assessment on rural single-carriageways and two-lane two-way roads. This procedure accounts for the influence of road characteristics on drivers’ perceptions. The methodology has been developed based on the human factors concepts from PIARC, and it includes a series of checklists that guide an inspector in carrying out a visual inspection of single-carriageway roads. The results from the checklist are then processed into an algorithm, and the level of risk in the analyzed section is provided. The objectives of the procedure are (a) to account for the perceptive aspects that are one of the major causes of road accidents, (b) to provide a proactive procedure in line with the requirements of the European Directive, and (c) to provide a useful instrument that can be easily implemented by road agencies and integrated with other analysis procedures. The procedure has been applied and tested on a case study of six different stretches of two-lane, two-way rural highways in Italy, Germany, and Slovenia (about 65 km). The results show a high degree of concordance with a risk classification based on the accident rate, mainly considering high-risk sections. Therefore, the procedure demonstrated its potential to be a useful instrument to be included in network safety assessments. Road agencies should consider the use of this procedure in their network safety analysis and ranking.

THE BENEFIT EVALUATION FRAMEWORK OF USING PERFORMANCE-BASED CONTRACTING IN URBAN ROAD MAINTENANCE

Performance-based contracts (PBC) have been adopted increasingly in various countries. However, establishing an objective and convincing benefit evaluation framework to facilitate promotion remains a significant challenge for those who wish to use PBCs. A recent study introduced PBC in the context of daily inspection and repair of urban roads. The pilot program in that study successfully addressed the needs of road users, road maintenance authorities, and contractors. Consequently, this study develops the benefit evaluation framework for this type of PBC for broader applications. The proposed benefit evaluation framework, consisting of one qualitative and eight quantitative performanceevaluation items, calculates the effectiveness of introducing PBC to verify its practicality. The validation results show that qualitative benefit evaluation using the Pavement Condition Index achieves a “satisfactory” rating. For the quantitative benefit evaluation, an overall benefit-cost ratio (BCR) of 5.35 suggests that it is economically viable. The developed benefit evaluation framework can be used as a simple and easy-to-use evaluation tool for other organizations to assess whether to introduce PBC. This study fills this research gap by addressing the use of PBCs in the maintenance of all categories of toll-free roads, and contributes to expanding the application of PBCs in road maintenance services.

Data Intelligence for UAV-Assisted Road Inspection in Post-Disaster Scenarios

Data Intelligence for UAV-Assisted Road Inspection in Post-Disaster Scenarios

Cross-Light Contrastive Learning of Pavement Crack Segmentation in Low-Light Conditions for Autonomous Road Inspection Vehicles

Cross-Light Contrastive Learning of Pavement Crack Segmentation in Low-Light Conditions for Autonomous Road Inspection Vehicles

Social-Cultural Factors and Performance of Roads Construction Projects in Nairobi City County, Kenya

The Kenyan Government is on the spot over stalled and poorly done roads construction projects especially in the Nairobi City County. Successful road construction is an impetus to economic development for Kenya as enumerated in the Kenya Vision 2030. However, there has been need for these road construction projects to update their systems due to the ever changing environment factors. Thisstudy therefore sought to determine the relationship between social-cultural factors and the performance of roads construction projectsin Nairobi City County, Kenya. The underpinning theory was the open system theory. Thisstudy used a positivist research philosophy. This study used the descriptive and explanatory research design. In this study, Nairobi City County refers to the five counties in the Nairobi Metropolitan Area namely, Nairobi County, Kiambu County, Kajiado County, Machakos County and Murang'a County. The target population of this research was the 176 completed roads construction projects in Nairobi City County by Kenya Rural Roads Authority (KERRA). The unit of observation was the road engineers, project planners and directors (KERRA), road supervisors, road inspectors, road surveyors, contractors, and project implementation teams’ members (KERRA). The unit of analysis wasthe completed roads construction projectsinNairobi City County. Proportional stratified sampling was used to derive a sample size of 253 respondents. A structured questionnaire was used to collect primary data. Data that was collected from the field was filtered, sorted and cleaned in line with research objectives. This study adopted both descriptive and inferential statistics. Descriptive statisticsincluding frequencies, percentages, mean scores, and standard deviation was produced for all the quantitative data. The results were presented using tables. Inferential statistics were done using the multiple linear regression. The study established a positive significant relationship between socio-cultural factors (t=2.417, p < 0.05) and project performance. The study concludes that human relationship influences project performance since projects operate within the society. Project culture is developed by communicating priority, status, and the alignment of official and operational rules. The study recommends that project leaders should be sensitive and aware of multicultural preferences guided by individual identity and role within the project versus family of origin and community, verbal and emotional expressiveness, relationship expectations, style of communication, language, personal priorities, values, and beliefs and time orientation.

Automatic road crack detection and analysis system based on deep feature fusion and edge structure extraction

ABSTRACT In order to solve the problem that the traditional crack image segmentation method cannot accurately extract the edge structure of road cracks, a flexible crack detection and identification system are provided. Firstly, a road crack dataset is built by processing the acquired images using the sliding window technique. A model called R2-AENet is proposed for training and testing. The model structure uses the recurrent residual convolutional neural network (RRCNN) to improve the convolutional structure and drive the accumulation of fracture features. The model decoder combines the attention-guided filtering module and the edge residuals module (ERB) better to extract the edge features of the feature map. As a result, it improves the crack segmentation performance. Finally, the research results are validated on the upgraded software platform. The experimental results demonstrate that the R2-AENet model achieves better segmentation results compared to the popular deep learning algorithm models. Evidently, the model achieves accuracy and recall of 0.982 and 0.849 on the collected dataset. Besides, the area under the curve (AUC) metric value can reach 0.992. In addition, the detection results in different data sets perform well, verifying that the model has exceptional robustness and can be applied to various environmental road inspection and maintenance projects.

Dark-Channel Soft-Constrained and Object-Perception-Enhanced Deep Dehazing Networks Used for Road Inspection Images.

Haze seriously affects the visual quality of road inspection images and contaminates the discrimination of key road objects, which thus hinders the execution of road inspection work. The basic assumptions of the classical dark-channel prior are not suitable for road images containing light-colored lane lines and vehicles, while typical deep dehazing networks lack physical model interpretability, and they focus on global dehazing effects, neglecting the preservation of object features. For this reason, this paper proposes a Dark-Channel Soft-Constrained and Object-Perception-Enhanced Deep Dehazing Network (DCSC-OPE-Net) for the information recovery of road inspection images. The network is divided into two modules: a dark-channel soft-constrained dehazing module and a near-view object-perception-enhanced module. Unlike the traditional dark-channel algorithms that impose strong constraints on dark pixels, a dark-channel soft-constrained loss function is constructed to ensure that the features of light-colored vehicles and lane lines are effectively maintained. To avoid resolution loss due to patch-based dark-channel processing for image dehazing, a resolution enhancement module is used to strengthen the contrast of the dehazed image. To autonomously perceive and enhance key road features to support road inspection, edge enhancement loss combined with a transmission map is embedded into the network to autonomously discover near-view objects and enhance their key features. The experiments utilize public datasets and real road inspection datasets to validate the performance of the proposed DCSC-OPE-Net compared with typical networks using dehazing evaluation metrics and road object recognition metrics. The experimental results demonstrate that the proposed DCSC-OPE-Net can obtain the best dehazing performance, with an NIQE score of 4.5 and a BRISQUE score of 18.67, and obtain the best road object recognition results (i.e., 83.67%) among the comparison methods.

Vision Image Monitoring on Transportation Infrastructures: A Lightweight Transfer Learning Approach

Vision monitoring of distress has emerged as a new trend in intelligent transportation infrastructure systems, including roads and bridges. Recently, transfer learning methods and lightweight networks have been used to realize efficient distress identification without large amount of human-labor work. This paper proposed an engineering approach that integrated transfer learning with lightweight models to classify and detect concrete bridge distresses. Two datasets named Distress Dataset of Asphalt Pavement (DDAP) with 2500 asphalt pavement distresses images and Distress Dataset of Concrete Bridge (DDCB) with 906 concrete bridge distresses images were used. The lightweight models MobileNet and MobileNet-SSD were employed to conduct 6 comparative experiments for exploring the model performance with different transfer learning modes (Mode I and II) and procedures (one-step and two-step procedure) in classification and detection tasks. Based on the comparison of the results, the optimum model for two tasks was respectively recognized. For classification task, the model directly transferred specific parameters of partial convolution layers from ImageNet-based model achieved the highest accuracy of 97.8%. For detection task, the two-step transfer learning model using an intermediate transfer learning step trained by DDAP reached a mean average precision of 87.16%. The proposed approach has the application potential for practical road inspection work in intelligent transportation infrastructure maintenance.