<rss version="2.0" xmlns:atom="https://www.w3.org/2005/Atom">
  <channel>
    <title>Transport Research International Documentation (TRID)</title>
    <link>https://trid.trb.org/</link>
    <atom:link href="https://trid.trb.org/Record/RSS?s=PHNlYXJjaD48cGFyYW1zPjxwYXJhbSBuYW1lPSJkYXRlaW4iIHZhbHVlPSJhbGwiIC8+PHBhcmFtIG5hbWU9InN1YmplY3Rsb2dpYyIgdmFsdWU9Im9yIiAvPjxwYXJhbSBuYW1lPSJ0ZXJtc2xvZ2ljIiB2YWx1ZT0ib3IiIC8+PHBhcmFtIG5hbWU9ImxvY2F0aW9uIiB2YWx1ZT0iMCIgLz48L3BhcmFtcz48ZmlsdGVycz48ZmlsdGVyIGZpZWxkPSJpbmRleHRlcm1zIiB2YWx1ZT0iJnF1b3Q7U2V3ZXImcXVvdDsiIG9yaWdpbmFsX3ZhbHVlPSImcXVvdDtTZXdlciZxdW90OyIgLz48L2ZpbHRlcnM+PHJhbmdlcyAvPjxzb3J0cz48c29ydCBmaWVsZD0icHVibGlzaGVkIiBvcmRlcj0iZGVzYyIgLz48L3NvcnRzPjxwZXJzaXN0cz48cGVyc2lzdCBuYW1lPSJyYW5nZXR5cGUiIHZhbHVlPSJwdWJsaXNoZWRkYXRlIiAvPjwvcGVyc2lzdHM+PC9zZWFyY2g+" rel="self" type="application/rss+xml" />
    <description></description>
    <language>en-us</language>
    <copyright>Copyright © 2026. National Academy of Sciences. All rights reserved.</copyright>
    <docs>http://blogs.law.harvard.edu/tech/rss</docs>
    <managingEditor>tris-trb@nas.edu (Bill McLeod)</managingEditor>
    <webMaster>tris-trb@nas.edu (Bill McLeod)</webMaster>
    <image>
      <title>Transport Research International Documentation (TRID)</title>
      <url>https://trid.trb.org/Images/PageHeader-wTitle.jpg</url>
      <link>https://trid.trb.org/</link>
    </image>
    <item>
      <title>Leveraging Bayesian Inference and Prior Knowledge to Predict Concrete Wall Loss Rate in Reinforced-Concrete Sanitary Sewer Pipes</title>
      <link>https://trid.trb.org/View/2606238</link>
      <description><![CDATA[AbstractThe deterioration of concrete/reinforced-concrete sanitary sewer pipes (RCSSPs), primarily due to microbially induced corrosion (MIC), introduces significant uncertainty and affects their serviceability. Nondestructive condition assessment tools such as laser-based technologies [laser profiling or light detection and ranging (LiDAR)] facilitate quantitative evaluation of concrete wall loss after subsequent postprocessing; however, accurately assessing the rate of wall-thickness reduction for a specific pipe requires sufficient data over time, which may be challenging to collect due to limited resources. The purpose of this study is to demonstrate how a Bayesian inference framework can integrate prior knowledge with limited observational data to improve the estimation of the concrete wall loss rate in RCSSPs. This framework, grounded in Bayes’ rule, utilizes limited concrete wall loss data from targeted pipes as evidence, while incorporating data from other pipes with similar operational characteristics to develop prior knowledge. This prior knowledge is then used to update key parameters that describe the progression of wall-thickness reduction. The likelihood function in the Bayesian inference framework describes the concrete wall loss in a specific pipe. Using the Bayes rule and evidence data, the parameters of this function are updated from their prior (i.e., prior knowledge) to their posterior distributions. The application of this framework is demonstrated through a case study that includes processed LiDAR data collected from 26 different RCSSPs (manhole-to-manhole) as evidence and processed laser profiling data collected from other RCSSPs as prior knowledge. The study is conducted on each pipe separately with two potential likelihood functions. Results indicate that both gamma and Weibull distributions effectively capture the concrete wall loss, primarily due to their shape parameters, which control the skewness of the data distribution. The likelihood function, with its posterior parameters, describes corrosion behavior and can be used for proactive assessment of the remaining service life of targeted pipes. Furthermore, the posterior distributions can be integrated as priors in future applications of the Bayesian framework to the same or other pipes, reducing the need for extensive data collection and enabling more efficient resource allocation.]]></description>
      <pubDate>Wed, 24 Jun 2026 13:22:05 GMT</pubDate>
      <guid>https://trid.trb.org/View/2606238</guid>
    </item>
    <item>
      <title>Characteristics and prediction of traffic noise induced by road manhole covers</title>
      <link>https://trid.trb.org/View/2689434</link>
      <description><![CDATA[Traffic noise from manhole covers under vehicular loading represents a significant component of urban traffic noise pollution. Yet, its generation mechanism and influencing factors remain unclear. This study integrated field noise measurements and finite element simulations to uncover the generation mechanism of tire-pavement-manhole cover interaction noise (TPCIN) and the influence patterns of key factors, including vehicle speed and manhole settlement, thereby establishing a TPCIN prediction model. The results demonstrate that the frequency of road manhole traffic noise is concentrated in the 0.4 to 3 kHz range, originating from the coupling of structural vibration, transient impact, cavity resonance, and air-pumping, with its maximum sound pressure level exhibiting a significant logarithmic dependence on both vehicle speed and manhole settlement. The established TPCIN prediction model achieves high accuracy, with prediction errors within ±3 dB(A) and 86.4% within ±2 dB(A). These findings provide a scientific basis for road manhole traffic noise alleviation.]]></description>
      <pubDate>Mon, 13 Apr 2026 09:37:43 GMT</pubDate>
      <guid>https://trid.trb.org/View/2689434</guid>
    </item>
    <item>
      <title>North Airfield Drainage Improvement at Chicago-O'Hare International Airport: Soil Stabilization Using Jet Grouting</title>
      <link>https://trid.trb.org/View/2200447</link>
      <description><![CDATA[Drainage improvements at O'Hare International Airport, Chicago, IL included the installation of a drainage and storm water system to control the overflow from nearby Willow Higgins Creek. This involved construction of a weir structure at the creek and the channeling of water from the weir to a newly constructed reservoir via three, 3.7-m (12-ft) diameter, underground storm sewer lines. The sewer lines passed beneath an existing, 2.3-m (90-inch) diameter high-pressure water main, which was to remain in service throughout sewer line installation. The soil composition around the water main ranged from medium stiff clays to silty sands and sandy silts. Triple-fluid jet grouting was used to stabilize the variable soil profile beneath the water main in preparation for tunneling and installation of the sewer lines. The varying soil strata presented a challenge to the project team to establish a constant set of jet grouting parameters throughout the stabilized zone while keeping within the specified unconfined compressive strength range for the stabilized zone of 690 to 1380 kPa (100 to 200 psi) at 28 days.]]></description>
      <pubDate>Fri, 06 Feb 2026 13:53:33 GMT</pubDate>
      <guid>https://trid.trb.org/View/2200447</guid>
    </item>
    <item>
      <title>Prediction of manhole-cover skid-resistance performance using decision trees and regression analysis</title>
      <link>https://trid.trb.org/View/2598411</link>
      <description><![CDATA[This study conducts skid-resistance coefficient tests on 3,600 manhole covers and collects data on various influencing factors, such as surface conditions, pattern types, starting zones, turning zones, usage duration, wheel track locations, and surrounding traffic volume. A preliminary classification analysis of the data is performed using decision trees to identify key factors affecting manhole-cover skid resistance. Subsequently, a multivariate nonlinear regression model is applied to examine the impact of these factors on the target variables. By integrating decision trees and nonlinear regression in a hybrid modeling approach, the study aims to reveal complex patterns in the data and generate reliable predictive results. The decision-tree analysis using non-continuous variables reveals that manhole covers located at turns and starting zones, and those with worn surfaces are more likely to fail to meet safety standards for skid resistance. The multivariate nonlinear regression model, which compares the predicted values against the actual British Pendulum Number (BPN) test results, achieves an accuracy of 82.9%.]]></description>
      <pubDate>Mon, 24 Nov 2025 15:30:52 GMT</pubDate>
      <guid>https://trid.trb.org/View/2598411</guid>
    </item>
    <item>
      <title>Structural dynamics analysis of asphalt pavement around manhole covers under vehicle loading</title>
      <link>https://trid.trb.org/View/2578641</link>
      <description><![CDATA[Recurring asphalt pavement deterioration around manholes significantly compromises the service quality and service life of the road. This study investigated the structural performance deterioration patterns of the asphalt pavement around manholes and the underlying mechanisms. Specifically, the dynamic load characteristics with vehicles driving over the pavement around the manhole were revealed using a vehicle vibration model, and the modulus field of the pavement was derived from the loading frequency and modulus reduction coefficients. Based on these results, a simulation model of the manhole-surrounding asphalt pavement structure was constructed to analyze the dynamic response of the pavement under moving vehicle loads. Furthermore, the effects of road structural materials, manhole materials, and settlement on the performance and service life of the surrounding pavement were evaluated. The results show that with pavement settlement around the manhole, the dynamic impact coefficient increases to 1.90 as vehicles drive over. The modulus of the asphalt pavement shows an initially slow and then sharp downward trend as the distance to the edge of the manhole cover decreases. Compared to the asphalt pavement elsewhere, the maximum deflection of the manhole-pavement structure increases by approximately 1.39 times, and the peak longitudinal tensile strain beneath the surface course increases by 28.10 times. With reduced modulus of the manhole ring or increased settlement, the fatigue life of the surface course is reduced by 98.9 % and 98.3 %, respectively. This study provides valuable guidance for the structural design of manholes and their surrounding pavement.]]></description>
      <pubDate>Mon, 08 Sep 2025 14:54:03 GMT</pubDate>
      <guid>https://trid.trb.org/View/2578641</guid>
    </item>
    <item>
      <title>C2F-RMD: Automated Road Manhole Detection and Condition Assessment</title>
      <link>https://trid.trb.org/View/2556648</link>
      <description><![CDATA[Road manhole detection and damage assessment are crucial for ensuring the safety and efficiency of transportation systems. Traditional methods, reliant on costly and inaccessible three-dimensional (3D) cameras, pose challenges, especially in resource-limited settings. This study introduces C2F-RMD, a groundbreaking deep learning (DL)–based algorithm that revolutionizes road manhole detection and damage assessment using only two-dimensional (2D) images. C2F-RMD adopts a two-stage approach. In the first stage, the coarse-to-fine (C2F) detection technique, coupled with scale-adaptive region-based convolutional neural network (R-CNN), accurately detects and classifies road manholes. Achieving an impressive F1 score of 0.96 and an intersection over union of 0.95 across eight classes, the C2F model provides robust results. The second stage, road manhole damage (RMD) index, employs self-crack segmentation and an elevation prediction model. The self-crack segmentation, trained without labeled data, attains remarkable accuracy rates: 0.821 for precision, 0.805 for recall, and 0.813 for F1 score. The innovative elevation prediction model forecasts manhole surroundings’ elevation maps using solely 2D image input, with a regression score (R²) of 0.77 and a mean absolute error (MAE) of 4.35 mm. Notably, this method was successfully applied to an 802-km road network in Seoul City, encompassing various road types, including urban, principal, and supplementary roads, as well as expressways. It accurately detected and classified eight types of manholes with an accuracy rate of 0.98. Additionally, the method achieved accuracy rates of 0.80 for crack segmentation, 0.88 for crack segmentation grading, and 0.83 for elevation difference grading in manhole condition evaluation, demonstrating its adaptability in detecting, classifying, and evaluating manholes across diverse road types. This promising approach has the potential to replace traditional manual visual assessments of road manholes.]]></description>
      <pubDate>Fri, 20 Jun 2025 17:03:41 GMT</pubDate>
      <guid>https://trid.trb.org/View/2556648</guid>
    </item>
    <item>
      <title>Deep learning-based predictive models for pavement patching and manholes evaluation</title>
      <link>https://trid.trb.org/View/2487575</link>
      <description><![CDATA[Deep Learning (DL) techniques have been applied to the processing and evaluation of pavement conditions. DL-based predictive models are rapid and have higher accuracy compared to traditional methods. In this research, some models were developed to appraise pavement patching by considering the severity and the manhole cover utilising images collected by the Road Surface Profiler (RSP) from Bus Rapid Transit (BRT) lines in Tehran, Iran. Two cases (scenarios) were regarded in this research: 1) Patching without considering the severity, and 2) Patching by considering the severity. YOLOv5, YOLOv6, and YOLOv7 as one-stage object detection algorithms, and Faster R-CNN (ResNet-50) and Faster R-CNN (MobileNet_v3) as two-stage object detection algorithms were evaluated to choose the desired algorithm. The results exhibited that the trained model on YOLOv5 has better performance and is faster than the rest of the algorithms. In addition, the influence of the classes’ combination on model performance was scrutinised. Based on the results of the other classes, the outcomes of the study suggested that the model can detect and train better when the model is trained by a dataset with all classes. Furthermore, the performance of the model was also investigated by examining the effect of image size and YOLOv5's architectures. Based on the obtained results, YOLOv5 and an image size of 640*640 outperformed the rest. In Case 1, the precision, recall, F1-score, and mAP were 75.8%, 67.4%, 71.35%, and 77.4%, respectively. Moreover, the values of these metrics in Case 2, were 70%, 61.2%, 65.3%, and 64.4%, respectively. Hence, the results demonstrated that the developed model can be used with high performance in the detection and evaluation of pavement patching by severity, and manholes in Pavement Management Systems (PMS) and maintenance and rehabilitation plans.]]></description>
      <pubDate>Tue, 28 Jan 2025 09:19:55 GMT</pubDate>
      <guid>https://trid.trb.org/View/2487575</guid>
    </item>
    <item>
      <title>Prediction of Geyser Occurrences in Covered Manholes of Urban Stormwater Systems</title>
      <link>https://trid.trb.org/View/2446955</link>
      <description><![CDATA[Storm geysers are air–water eruptions from stormwater manholes during intense rainfalls, which raise public safety concerns. They arise from entrapped air release in stormwater tunnels as flow transitions from open to pressurized. This paper experimentally investigated air–water flow characteristics during air pocket release in covered manholes. Four geyser regimes were identified: no geyser, single air-release geyser, single rapid-filling geyser, and multigeysers. The air-release geyser and rapid-filling geyser refer to ejections of water column and air–water mixture induced by distinct mechanisms. Impacts of manhole diameter, cover ventilation area, system pressure head, and initial air pocket volume on geyser regimes were analyzed. As manhole diameter increases or cover ventilation area decreases, maximum geyser heights decrease, subsequently decreasing the likelihood of geyser occurrence. Higher system pressure head and larger initial air pocket volume increase the maximum geyser heights. Equations were derived to predict the maximum geyser heights, and prediction accuracies exceeded 87%, providing safe predictions of geyser occurrences relative to the manhole height. The findings can help provide advance warning of geysers when combined with information of stormwater system structures and real-time monitoring of operating conditions.]]></description>
      <pubDate>Wed, 08 Jan 2025 09:41:16 GMT</pubDate>
      <guid>https://trid.trb.org/View/2446955</guid>
    </item>
    <item>
      <title>Energy Dissipation in Drop Manhole Cascades</title>
      <link>https://trid.trb.org/View/2399861</link>
      <description><![CDATA[Drop manholes are hydraulic devices typically employed to dissipate flow energy in high slope, urban drainage systems. The use of several drops of small height placed in series (drop manhole cascade) is a more economical and effective method for controlling velocity and kinetic energy than a single drop of large height. An experimental campaign was specifically designed to investigate hydraulics of drop manhole cascades at the LIA Laboratory, University of Cassino and Southern Lazio. Results were summarized in equations that quantify overall dissipation performance and describe the hydraulics of the downstream flow in terms of outflow depth and energy head. The performance of single drops and a three-drop cascade are compared. The data show how the energy of the stream passing through the dropshaft cascade quickly achieves a new equilibrium condition downstream from the second manhole. Finally, the energy dissipation of a drop manhole cascade was compared with that of smooth and roughened chutes to highlight similarities and differences between different dissipation structures.]]></description>
      <pubDate>Wed, 17 Jul 2024 10:12:35 GMT</pubDate>
      <guid>https://trid.trb.org/View/2399861</guid>
    </item>
    <item>
      <title>Pavement Design Around Utilities – Best Practice</title>
      <link>https://trid.trb.org/View/2394485</link>
      <description><![CDATA[Asphalt concrete pavement surrounding utility structure covers is prone to settlement, cracking, breaking up over time, and potholing, and these distresses are particularly common in wet-freeze climates (e.g., Minnesota). Several factors contribute to their formation, including design requirements, collar material type and cut shape, construction practices, frost heave, and backfill settlement. If not properly maintained, the distressed pavement can lead to ride quality issues and hazards for vehicles and snowplows. Differences in design details and construction practices can result in different performance; however, differences in pavement performance around utility structure covers are not well documented. The main goal of this Minnesota Local Road Research Board (LRRB) project is to fill this knowledge gap by documenting regional agency best practices for adjusting utility covers and patching the surrounding pavement. Information is gathered through a review of existing information, a review of standard details from agencies in and around Minnesota, an agency survey, and follow-up discussions with agencies that are generally satisfied with their practices. No single best practice is identified for design details;however, common themes among agencies include the importance of both inspecting and testing during construction and achieving adequate compaction of all pavement patch layers. This document is developed to assist local transportation agency personnel and engineering consultants in improving design and maintenance of asphalt concrete pavement around utility covers. It highlights successful and unsuccessful regional practices and trends, factors contributing to pavement damage around utility covers, timing of inspections and maintenance, and a framework for evaluating and modifying practice.]]></description>
      <pubDate>Fri, 28 Jun 2024 14:01:21 GMT</pubDate>
      <guid>https://trid.trb.org/View/2394485</guid>
    </item>
    <item>
      <title>Numerical and experimental investigation on synthetic macrofiber-reinforced concrete manhole exposed to railway loads</title>
      <link>https://trid.trb.org/View/2362373</link>
      <description><![CDATA[The refurbishment of railway lines and the installation of new tracks necessitate the construction of numerous concrete manholes; therefore, the optimization of manholes should be investigated. To this end, the use of innovative materials in addition to advanced design methods with realistic modeling is required. In the case of conservatively designed structures, there exists the possibility of redesigning the structure utilizing suitable fiber-reinforced concrete (FRC) only. The main advantage of synthetic macrofibers over steel is their complete corrosion resistance, which is essential in corrosive environments. Other advantages include their low carbon footprint, reduced construction time, and cost-effectiveness. This paper outlines the optimization process for a conventional cast-in-situ concrete manhole. The imperative for a monolithic construction system stems from the diverse geometries and distinct designs of individual pipe culverts, compounded by the often-challenging accessibility of installation sites. In the optimization phase, synthetic macrofiber reinforcement replaced conventional reinforcing steel bars, using advanced finite element analysis (FEA). The design was not conducted on an equivalent basis, resulting in potential variations in the load-carrying capacity between reinforced concrete (RC) and FRC manholes. Nevertheless, both are deemed suitable for the specified loads. The conventional design method used for RC and the advanced finite element design method used for FRC were scrutinized, taking into account the existing standard environment. Subsequently, a real-scale test was conducted to validate the calculations. Carbon footprint analyses were performed for both the original and proposed solutions, and the results were compared. The solution obtained in this study is unique and pioneering in terms of both the calculation method and the structural design, and the CO2 calculations validate its necessity.]]></description>
      <pubDate>Tue, 30 Apr 2024 15:17:58 GMT</pubDate>
      <guid>https://trid.trb.org/View/2362373</guid>
    </item>
    <item>
      <title>Renewal of Rectangular Culverts Using the RPC Method</title>
      <link>https://trid.trb.org/View/2200977</link>
      <description><![CDATA[The total length of sewers in Tokyo's 23 wards has reached about 15,000 km, of which 2000 km (about 13%) is older than 50 years, which is the designed lifespan of culverts. Many of the aging sewer culverts are located in Tokyo's urban center, which increasingly suffers inundation, road subsidence and odious smells attributable to insufficient capacity or aging of storm sewer culverts. As the amount of sewage increases because of people's changing lifestyles and the amount of stormwater increases under the influence of urbanization, there is an urgent need to renew sewer systems to increase sewer capacity and enhance the earthquake resistance of sewer systems. Today, it is difficult to replace sewers by the open-cut method because of adverse effects on road traffic, underground installations and the living environments of local residents. Non-open-cut methods are being used, therefore, to renew sewers. Renewal methods include the reversed piping method, the formed piping method, the sleeve piping method and the spiral piping method. The RPC (Renewal Precast/Parallel Culvert) method is a non-open-cut sleeve piping method by which existing rectangular culverts of medium- and large-cross sections can be lined efficiently with self-supporting culverts.]]></description>
      <pubDate>Wed, 25 Oct 2023 10:14:29 GMT</pubDate>
      <guid>https://trid.trb.org/View/2200977</guid>
    </item>
    <item>
      <title>Cement and fly ash-treated recycled aggregate blends for backfilling trenches in trafficable areas</title>
      <link>https://trid.trb.org/View/2233704</link>
      <description><![CDATA[The shortage of natural aggregates available for filling excavated pipeline trenches in trafficable areas has prompted the exploration of alternative resources. This study investigates the feasibility of using cement and fly ash-treated recycled aggregates as trench backfill materials subjected to traffic loadings. Blends of recycled glass (RG), plastic (RP), and tire (RT) were treated with different proportions of cement and fly ash, resulting in a total of 8 treated blends. Geotechnical tests including compaction and California Bearing Ratio (CBR) were conducted to evaluate the mixtures according to backfill specifications. Specialized pavement testing, such as repeated load triaxial testing (RLT) and quick shear, simulated real-life stress levels at trafficable areas. Scanning electron microscope (SEM) images were taken to investigate the microstructural characteristics of the cement and fly ash-treated samples. The results showed that the CBR and resilient modulus of the treated blends improved with higher cement, fly ash, and RG contents, while they decreased with increased RT content. Cement-treated blends demonstrated significant improvements in peak shear strength with increased cement and RG contents and decreased RT content. Fly ash-treated blends showed minor improvement in peak shear strength when the fly ash, RG, and RT contents varied. Only cement-treated blends exhibited properties comparable to Class 4 (CL4) crushed rock, which was the control material. Under the same stress levels, cement-treated blends demonstrated up to 1.17 and 2.62 times greater stiffness than CL4 and clay subgrades, respectively. The SEM analyses confirmed that the inclusion of cement in the recycled blends resulted in the formation of greater bonds between particles compared to fly ash, which led to higher strength. These findings highlight the potential of sustainable materials in backfilling pipeline trenches under traffic loadings, reducing the reliance on natural aggregates for this application.]]></description>
      <pubDate>Fri, 22 Sep 2023 08:53:19 GMT</pubDate>
      <guid>https://trid.trb.org/View/2233704</guid>
    </item>
    <item>
      <title>Recycled aggregate blends for backfilling deep trenches in trafficable areas</title>
      <link>https://trid.trb.org/View/2231435</link>
      <description><![CDATA[Limited supplies of natural aggregates for backfilling pipeline trenches in trafficable areas have led to considering secondary resources. Utilizing blends entirely made of recycled aggregates to backfill excavated trenches in trafficable areas is rare if any, where traditionally, natural crushed rock has been used. In this study, 18 blends of various proportions of recycled glass (RG), plastic (RP), tire (RT), and concrete aggregate (RCA) were proposed and studied as alternatives. First, typical pavement tests such as compaction and CBR were carried out to select 7 suitable mix designs based on available backfill specifications. The shortlisted blends were further investigated through a specialized testing program, in particular, repeated load triaxial to simulate site stress levels. The results showed that CBR and resilient modulus characteristics improved by the increase in RCA content and the reduction in RT content. Three recycled material blends were selected as the optimum mix designs, and were next compared with the surrounding natural clay subgrades of the study area. The resilient modulus response of optimum mix designs exhibited up to 1.6 times greater stiffness compared to the clay subgrades under the same stress levels. The outcomes of this study reduce the demand for natural aggregates and promote the use of sustainable materials for backfilling pipeline trenches located under traffic loadings.]]></description>
      <pubDate>Mon, 18 Sep 2023 17:10:39 GMT</pubDate>
      <guid>https://trid.trb.org/View/2231435</guid>
    </item>
    <item>
      <title>Automated Detection of Pavement Manhole on Asphalt Pavements with an Improved YOLOX</title>
      <link>https://trid.trb.org/View/2221043</link>
      <description><![CDATA[Accurate recognition and location of pavement manholes are of great significance for pavement maintenance. This paper proposes an improved You only look once X (YOLOX) for automated detection of manholes on asphalt pavements. The proposed model improves the performance of the YOLOX model in two respects. First, the channel attention mechanism is introduced to enhance the model’s adaptive feature refinement; second, a microscale detection layer is deployed in the YOLOX model to extract more essential and distinct features. The experimental results are impressive, with the improved YOLOX achieving an F1 score and overall intersection-over-union of 98.14% and 91.61%, respectively, on 250 testing images, surpassing other state-of-the-art models such as YOLOv4, Faster R-CNN, EfficientDet, and the original YOLOX. To demonstrate robustness of the proposed model, the improved YOLOX is further applied to process manhole images taken randomly by a smartphone, which differ significantly from those acquired by a laser imaging system. It is found that the improved YOLOX can also yield similar detection efficiency in different scenes, which indicates the proposed model has a strong generalization ability. Particularly, the average frame per second (FPS) of the improved YOLOX is approximately 50.74 FPS using a modern graphic processing unit (GPU) device, implying the promising potential of the proposed model in supporting real-time automated detection of pavement manholes.]]></description>
      <pubDate>Fri, 01 Sep 2023 15:26:11 GMT</pubDate>
      <guid>https://trid.trb.org/View/2221043</guid>
    </item>
  </channel>
</rss>