Automated Pavement Condition Index Assessment with Deep Learning and Image Analysis: An End-to-End Approach.

The degradation of road pavements due to environmental factors is a pressing issue in infrastructure maintenance, necessitating precise identification of pavement distresses. The pavement condition index (PCI) serves as a critical metric for evaluating pavement conditions, essential for effective budget allocation and performance tracking. Traditional manual PCI assessment methods are limited by labor intensity, subjectivity, and susceptibility to human error. Addressing these challenges, this paper presents a novel, end-to-end automated method for PCI calculation, integrating deep learning and image processing technologies. The first stage employs a deep learning algorithm for accurate detection of pavement cracks, followed by the application of a segmentation-based skeleton algorithm in image processing to estimate crack width precisely. This integrated approach enhances the assessment process, providing a more comprehensive evaluation of pavement integrity. The validation results demonstrate a 95% accuracy in crack detection and 90% accuracy in crack width estimation. Leveraging these results, the automated PCI rating is achieved, aligned with standards, showcasing significant improvements in the efficiency and reliability of PCI evaluations. This method offers advancements in pavement maintenance strategies and potential applications in broader road infrastructure management.

Mapping nationwide concentrations of sulfate and nitrate in ambient PM2.5 in South Korea using machine learning with ground observation data.

Particulate matter (PM) is a major air pollutant in Northeast Asia, with frequent high PM episodes. To investigate the nationwide spatial distribution maps of PM2.5 and secondary inorganic aerosols in South Korea, prediction models for mapping SO42- and NO3- concentrations in PM2.5 were developed using machine learning with ground-based observation data. Specifically, the random forest algorithm was used in this study to predict the SO42- and NO3- concentrations at 548 air quality monitoring stations located within the representative radii of eight intensive air quality monitoring stations. The average concentrations of PM2.5, SO42-, and NO3- across the entire nation were 17.2 ± 2.8, 3.0 ± 0.6, and 3.4 ± 1.2 µg/m3, respectively. The spatial distributions of SO42- and NO3- concentrations in 2021 revealed elevated concentrations in both the western and central regions of South Korea. This result suggests that SO42- concentrations were primarily influenced by industrial activities rather than vehicle emissions, whereas NO3- concentrations were more associated with vehicle emissions. During a high PM2.5 event (November 19-21, 2021), the concentration of SO42- was primarily influenced by SOX emissions from China, while the concentration of NO3- was affected by NOX emissions from both China and Korea. The methodology developed in this study can be used to explore the chemical characteristics of PM2.5 with high spatiotemporal resolution. It can also provide valuable insights for the nationwide mitigation of secondary PM2.5 pollution.

Special Issue on "Smart City and Smart Infrastructure".

Recent developments in sensor technologies and data-driven approaches have been recognized as the main enablers of smart cities [...].

Semi-Precise Analytical Method for Investigating the Liftoff Variation on the Hall Sensor in Metal Defect Sensing.

Hall-effect sensors are used to detect metal surface defects both experimentally and numerically. The gap between the specimen and the sensor, called the liftoff, is assumed to remain constant, while a slight misplacement of a sample may lead to incorrect measurements by the Hall-effect sensor. This paper proposes a numerical simulation method to mitigate the liftoff issue. Owing to the complexity of conducting precise finite-element analysis, rather than obtaining the induced current in the Hall sensor, only the magnetic flux leakage is obtained. Thus, to achieve a better approximation, a numerical method capable of obtaining the induced current density in the circumferential direction in terms of the inspection direction is also proposed. Signals of the conventional and proposed approximate numerical methods affected by the sensor liftoff variation were obtained and compared. For small liftoffs, both conventional and proposed numerical methods could identify notch defects, while as the liftoff increased, no defect could be identified using the conventional numerical method. Furthermore, experiments were performed using a variety of liftoff configurations. Based on the results, considering the threshold of the conventional numerical method, defects were detected for greater liftoffs, but misdetection did not occur.

Development of inspection robots for bridge cables.

This paper presents the bridge cable inspection robot developed in Korea. Two types of the cable inspection robots were developed for cable-suspension bridges and cable-stayed bridge. The design of the robot system and performance of the NDT techniques associated with the cable inspection robot are discussed. A review on recent advances in emerging robot-based inspection technologies for bridge cables and current bridge cable inspection methods is also presented.

A synchronized multipoint vision-based system for displacement measurement of civil infrastructures.

This study presents an advanced multipoint vision-based system for dynamic displacement measurement of civil infrastructures. The proposed system consists of commercial camcorders, frame grabbers, low-cost PCs, and a wireless LAN access point. The images of target panels attached to a structure are captured by camcorders and streamed into the PC via frame grabbers. Then the displacements of targets are calculated using image processing techniques with premeasured calibration parameters. This system can simultaneously support two camcorders at the subsystem level for dynamic real-time displacement measurement. The data of each subsystem including system time are wirelessly transferred from the subsystem PCs to master PC and vice versa. Furthermore, synchronization process is implemented to ensure the time synchronization between the master PC and subsystem PCs. Several shaking table tests were conducted to verify the effectiveness of the proposed system, and the results showed very good agreement with those from a conventional sensor with an error of less than 2%.

A vision-based dynamic rotational angle measurement system for large civil structures.

In this paper, we propose a vision-based rotational angle measurement system for large-scale civil structures. Despite the fact that during the last decade several rotation angle measurement systems were introduced, they however often required complex and expensive equipment. Therefore, alternative effective solutions with high resolution are in great demand. The proposed system consists of commercial PCs, commercial camcorders, low-cost frame grabbers, and a wireless LAN router. The calculation of rotation angle is obtained by using image processing techniques with pre-measured calibration parameters. Several laboratory tests were conducted to verify the performance of the proposed system. Compared with the commercial rotation angle measurement, the results of the system showed very good agreement with an error of less than 1.0% in all test cases. Furthermore, several tests were conducted on the five-story modal testing tower with a hybrid mass damper to experimentally verify the feasibility of the proposed system.

Sulfur deposition simulations over China, Japan, and Korea: a model intercomparison study for abating sulfur emission.

In response to increasing trends in sulfur deposition in Northeast Asia, three countries in the region (China, Japan, and Korea) agreed to devise abatement strategies. The concepts of critical loads and source-receptor (S-R) relationships provide guidance for formulating such strategies. Based on the Long-range Transboundary Air Pollutants in Northeast Asia (LTP) project, this study analyzes sulfur deposition data in order to optimize acidic loads over the three countries. The three groups involved in this study carried out a full year (2002) of sulfur deposition modeling over the geographic region spanning the three countries, using three air quality models: MM5-CMAQ, MM5-RAQM, and RAMS-CADM, employed by Chinese, Japanese, and Korean modeling groups, respectively. Each model employed its own meteorological numerical model and model parameters. Only the emission rates for SO(2) and NO(x) obtained from the LTP project were the common parameter used in the three models. Three models revealed some bias from dry to wet deposition, particularly the latter because of the bias in annual precipitation. This finding points to the need for further sensitivity tests of the wet removal rates in association with underlying cloud-precipitation physics and parameterizations. Despite this bias, the annual total (dry plus wet) sulfur deposition predicted by the models were surprisingly very similar. The ensemble average annual total deposition was 7,203.6 ± 370 kt S with a minimal mean fractional error (MFE) of 8.95 ± 5.24 % and a pattern correlation (PC) of 0.89-0.93 between the models. This exercise revealed that despite rather poor error scores in comparison with observations, these consistent total deposition values across the three models, based on LTP group's input data assumptions, suggest a plausible S-R relationship that can be applied to the next task of designing cost-effective emission abatement strategies.