Applied Research最新文献

{"title":"Fabrication, Characterization, and Antimicrobial Activity of PLA/Chitosan Composites Reinforced With Sugarcane Bagasse–Derived Fillers","authors":"Akbarningrum Fatmawati, Athalia Beatrice Molianto, Theofilus Pranata, Emma Savitri, The Jaya Suteja","doi":"10.1002/appl.70161","DOIUrl":"https://doi.org/10.1002/appl.70161","url":null,"abstract":"<p>This study developed polylactic acid (PLA) composites reinforced with chitosan (CS) and sugarcane bagasse-derived fillers, including untreated sugarcane bagasse (SCB), ultrasound-assisted-alkaline-treated sugarcane bagasse (UASCB), and ultrasound-assisted-alkaline-acid-treated (UAASCB). The objective was to improve the structural and functional properties of PLA composites through fiber treatment and chitosan incorporation. Fiber treatments significantly increased the cellulose content and crystallinity of SCB, with the best result reaching 71.7% cellulose and 91.36% crystallinity after 30 min of 4 wt% NaOH treatment followed by 45 min of 2 N sulfuric acid treatment. Composite characterization was carried out using FTIR, XRD, SEM, and tensile testing. In this study, pure PLA showed a tensile strength of <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 \u0000 <mrow>\u0000 <mn>6.69</mn>\u0000 \u0000 <mo>±</mo>\u0000 \u0000 <mn>2.8</mn>\u0000 </mrow>\u0000 </mrow>\u0000 <annotation> $6.69pm 2.8$</annotation>\u0000 </semantics></math> MPa and a tensile modulus of <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 \u0000 <mrow>\u0000 <mn>313.26</mn>\u0000 \u0000 <mo>±</mo>\u0000 \u0000 <mn>34</mn>\u0000 </mrow>\u0000 </mrow>\u0000 <annotation> $313.26pm 34$</annotation>\u0000 </semantics></math> MPa. The PLA–CS–UASCB composite showed a comparable tensile strength (<span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 \u0000 <mrow>\u0000 <mn>6.31</mn>\u0000 \u0000 <mo>±</mo>\u0000 \u0000 <mn>4.4</mn>\u0000 </mrow>\u0000 </mrow>\u0000 <annotation> $6.31pm 4.4$</annotation>\u0000 </semantics></math>) MPa with a slight increase in tensile modulus (<span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 \u0000 <mrow>\u0000 <mn>360</mn>\u0000 \u0000 <mo>±</mo>\u0000 \u0000 <mn>84</mn>\u0000 </mrow>\u0000 </mrow>\u0000 <annotation> $360pm 84$</annotation>\u0000 </semantics></math>) MPa. In contrast, the PLA–CS–UAASCB composite displayed lower tensile strength (<span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 \u0000 <mrow>\u0000 <mn>2.92</mn>\u0000 \u0000 <mo>±</mo>\u0000 ","PeriodicalId":100109,"journal":{"name":"Applied Research","volume":"5 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/appl.70161","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148783802","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evolution and Influence Analysis of Skid Resistance Performance in Steel Slag Ultra-Thin Wearing Course Based on XGBoost-SHAP Framework","authors":"Wang Wang,&nbsp;Xinquan Xu","doi":"10.1002/appl.70147","DOIUrl":"https://doi.org/10.1002/appl.70147","url":null,"abstract":"<p>The integration of steel slag as a substitute for natural aggregate in ultra-thin wearing courses (UTWC) presents a sustainable approach to pavement maintenance. However, the evolution of skid resistance under multiple loading factors remains insufficiently quantified. This study introduces an interpretable machine learning framework that combines Extreme Gradient Boosting (XGBoost) with SHapley Additive exPlanations (SHAP) to predict and explain the skid resistance decay of steel slag asphalt mixtures subjected to MLS11 accelerated loading. A dataset comprising 168 observations from 24 specimens—featuring three steel slag replacement ratios across seven loading stages—was constructed, incorporating variables related to material composition, aggregate meso-morphology, pavement surface texture, and loading history. Four predictive models—multiple linear regression (MLR), support vector regression (SVR), random forest (RF), and XGBoost—were evaluated using a strict group-aware train-test split to prevent data leakage. Among the evaluated models, RF achieved the highest raw predictive accuracy, while XGBoost showed comparable performance with an <i>R</i>² of 0.979, RMSE of 1.202, and MAPE of 1.64%. Considering its near-optimal predictive accuracy, built-in regularization, and strong compatibility with the SHAP framework, XGBoost was selected as the principal model for subsequent interpretable analysis. Dependence plots revealed significant non-linear saturation effects, with BPN decay accelerating sharply during the first 20 × 10⁴ cycles before stabilizing. Steel slag content showed a threshold effect, beyond which additional anti-skid improvements diminished. Local SHAP waterfall decomposition illustrated that high <i>R</i>z and low <i>N</i> values contributed to elevated BPN predictions, highlighting the “buffer mechanism” of steel slag against polishing-induced texture loss. This framework provides a data-driven tool for optimizing steel slag dosage and predicting residual skid resistance in pavement maintenance.</p>","PeriodicalId":100109,"journal":{"name":"Applied Research","volume":"5 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/appl.70147","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148753528","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Machine Learning Prediction of Soil-Water Characteristic Curve and Its Application in Lateral Earth Pressure Calculation of Retaining Walls","authors":"Cheng Chang,&nbsp;Xiaobin Mu,&nbsp;Libin Han,&nbsp;Shuangjian Niu","doi":"10.1002/appl.70152","DOIUrl":"https://doi.org/10.1002/appl.70152","url":null,"abstract":"<div>\u0000 \u0000 <p>The soil-water characteristic curve (SWCC) is a fundamental parameter that governs the hydro-mechanical behavior of unsaturated soils. Conventional laboratory measurement of SWCC is time-consuming and labor-intensive, while traditional lateral earth pressure design for retaining walls frequently relies on the saturated soil assumption, neglecting the effects of SWCC and resulting in significant systematic deviations in calculations. This study develops a statistically rigorous machine learning (ML) framework for efficient SWCC prediction and its application to lateral earth pressure calculations for pile-supported box counterfort retaining walls. Four ML algorithms with distinct methodological frameworks were employed: extreme learning machine (ELM), least squares support vector machine (LSSVM), projection pursuit regression (PPR), and Bayesian ridge regression (BRR). These algorithms were utilized to construct SWCC prediction models using the cleaned UNSODA database. Model performance was assessed through multi-metric evaluation, paired <i>t</i>-tests for statistical significance, and robustness analysis involving 30 independent runs, with validation conducted on measured silty clay data across 12 suction levels. Results indicate that the ELM model achieves the highest prediction accuracy, demonstrating statistically significant superiority over LSSVM, PPR, and BRR and excellent robustness. Independent validation reveals an average relative error of only 2.25% for ELM-predicted SWCC. The SWCC-based earth pressure calculation rectifies the bidirectional deviations of the traditional saturated method and identifies a neutral point at a depth of 17.5 m for a 25 m-high retaining wall. This study offers a reliable technical approach for rapid SWCC acquisition and refined lateral earth pressure design for retaining structures.</p></div>","PeriodicalId":100109,"journal":{"name":"Applied Research","volume":"5 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/appl.70152","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148752443","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Research on a Lightweight Portable Dynamic Pavement Roughness Measurement System","authors":"Hong Zhang,&nbsp;Yuanshuai Dong,&nbsp;Yun Hou,&nbsp;Chang Liu,&nbsp;Zhenyu Qian","doi":"10.1002/appl.70148","DOIUrl":"https://doi.org/10.1002/appl.70148","url":null,"abstract":"<div>\u0000 \u0000 <p>Pavement roughness measurement constitutes a fundamental basis for evaluating road serviceability, ride comfort, and maintenance decision-making. Existing vehicle-mounted roughness measurement systems generally rely on dedicated survey vehicles, presenting limitations in mounting adaptability, installation complexity, and versatility, which hinder flexible deployment in grassroots maintenance scenarios, particularly for low-volume and rural roads with growing maintenance demands. To address these issues, this study develops a lightweight pavement roughness measurement system suitable for ordinary vehicles, based on the combined sensing principle of laser displacement, acceleration, and rotary encoder. A portable mounting scheme is proposed, along with a data processing framework encompassing multi-sensor synchronized acquisition, signal denoising, vibration displacement compensation, longitudinal profile reconstruction, resampling, and International Roughness Index (IRI) computation. The IRI is computed from the reconstructed longitudinal profile using the standardized quarter-car simulation. Under the tested pavement conditions, experimental results indicate that both the 0.1 and 0.25 m sampling intervals adequately characterize the macroscopic longitudinal profile, while the finer 0.1 m interval is more favorable for capturing localized surface details; shorter computation intervals (10–20 m) exhibit higher sensitivity to localized anomalies than longer ones. Within the speed range of 30–50 km/h, the coefficient of variation of segmental IRI values remains below 5%, demonstrating satisfactory repeatability. Compared with a commercial comprehensive inspection vehicle, the CICS Road Condition Rapid Detection System (CICS), operated at the same speeds, the system yields generally consistent results, with optimal overall performance at 40 km/h, where the correlation coefficient reaches <i>R</i> = 0.989, and the root-mean-square error is 0.098 m/km.</p></div>","PeriodicalId":100109,"journal":{"name":"Applied Research","volume":"5 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/appl.70148","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148752442","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Entropy-Weighted Multi-Factor Grey-Markov Model for Pavement Performance Prediction in Subtropical Hilly Regions","authors":"Chaowei Yu,&nbsp;Wen Li,&nbsp;Weiwen Quan,&nbsp;Kang Jiang,&nbsp;Kefei Liu","doi":"10.1002/appl.70158","DOIUrl":"https://doi.org/10.1002/appl.70158","url":null,"abstract":"<div>\u0000 \u0000 <p>Asphalt pavements in subtropical hilly regions are subjected to the long-term coupled effects of high temperature, high humidity, intense rainfall, and traffic loading, while their performance datasets are often short and fluctuate considerably. To address the limitations of traditional Grey-Markov models, including insufficient multi-factor representation, subjective residual-state partitioning, and inadequate use of multi-step transition information, this study develops an entropy-weighted multi-factor Grey-Markov model for pavement condition index (PCI) prediction. First, a GM(1,1) model is established using the PCI series from 2010 to 2018 to extract the overall deterioration trend of pavement performance. Second, five climate–traffic loading indicators are integrated through the entropy weight method to construct a composite pressure index <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 \u0000 <mrow>\u0000 <mi>F</mi>\u0000 \u0000 <mo>(</mo>\u0000 \u0000 <mi>k</mi>\u0000 \u0000 <mo>)</mo>\u0000 </mrow>\u0000 </mrow>\u0000 <annotation> $F(k)$</annotation>\u0000 </semantics></math> for systematic residual correction. Finally, the remaining stochastic residuals are softly classified using the central-point triangular whitenization weight function and corrected using weighted multi-step Markov transition probabilities. The memory decay coefficient is further optimized by minimizing the mean absolute percentage error (MAPE) of rolling one-step validation from 2017 to 2019. Using PCI data from a road section in Hunan Province from 2010 to 2019, the results show that the proposed model achieves the highest accuracy when the memory decay coefficient <span></span><math>\u0000 <semantics>\u0000 <mrow>\u0000 \u0000 <mrow>\u0000 <mi>ρ</mi>\u0000 </mrow>\u0000 </mrow>\u0000 <annotation> $rho $</annotation>\u0000 </semantics></math> is 0.3. The corresponding one-step, two-step, and three-step transition weights are 0.7194, 0.2158, and 0.0647, respectively. Under this setting, the relative error of the 2019 PCI prediction is 0.4875%, and the rolling-validation MAPE is 0.5839%. Compared with the traditional Grey-Markov model, the relative error and MAPE are reduced by 24.92% and 61.48%, respectively; compared with the equal-weight multi-step scheme, these two metrics are reduced by 18.43% and 18.20%, respectively. The results indicate that combining systematic residual correction with multi-step Markov correction improves the accuracy and rolling-validation stability of small-sample pavement performance prediction.</p></div>","PeriodicalId":100109,"journal":{"name":"Applied Research","volume":"5 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-08-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/appl.70158","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148752436","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Synergistic Effects of Silica and Calcium-Based Dual Admixtures on the Flowability, Microstructure, and Strength Development in Low-Carbon Concrete","authors":"Md. Soybur Rahman,&nbsp;Zakaria Hossain","doi":"10.1002/appl.70157","DOIUrl":"https://doi.org/10.1002/appl.70157","url":null,"abstract":"<p>Eco-friendly construction practices have emerged as a cornerstone of modern development. Due to the substantial carbon emissions from cement production, there is a growing demand for supplementary cementitious materials. This study investigated the synergistic effects of two different wastes, silica-rich rice husk ash (RHA) and calcium-rich seashell powder (SSP), on the flowability, microstructure, strength development, and reduction in carbon emissions. RHA (0%–12.5%) and SSP (0%–12.5%) were incorporated in equal proportions, replacing cement at levels ranging from 0% to 25%. The fresh and hardened properties of the concrete were evaluated through slump, compressive strength, and split tensile strength tests. Microstructural characteristics were analyzed using scanning electron microscopy (SEM) and energy-dispersive X-ray spectroscopy (EDS). In addition, the environmental performance was assessed based on global warming potential (GWP) and eco-strength efficiency (ESE). The results revealed that incorporating RHA and SSP reduced both the workability and the unit weight of the hardened concrete. Replacing 5%–10% of the cement with these ash powders initially had a negative impact on mechanical performance. However, beyond 10% replacement, gradual improvements were observed. At 25% replacement, the modified concrete achieved superior compressive strength, along with enhanced split tensile strength, modulus of elasticity, and cement efficiency. Moreover, the blended concrete demonstrated notable environmental benefits by reducing carbon emissions. Overall, the combined use of RHA and SSP offers a promising approach for sustainable waste management and eco-efficient construction practices.</p>","PeriodicalId":100109,"journal":{"name":"Applied Research","volume":"5 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/appl.70157","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148849344","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Analysis of Trolley Wire Wear and Temperature on an Electric Train Catenary System Based on Image Processing","authors":"Ahmad Sugiana,&nbsp;Heru Syah Putra,&nbsp;Kharisma Bani Adam,&nbsp;Muhammad Kodrat,&nbsp;Dermawan Setiananda","doi":"10.1002/appl.70143","DOIUrl":"https://doi.org/10.1002/appl.70143","url":null,"abstract":"<div>\u0000 \u0000 <p>The catenary system particularly the trolley wire in electric trains, is vulnerable to temperature changes and wear, which affects both safety and operating effectiveness. This study provides an image processing-based approach to replace ineffective human inspections. We collected trolley wire data using a thermal Flir One Gen 3 camera and a Z Cam E2 camera. Convolutional neural network (CNN) techniques were then used to identify wear and arcing (electrical sparks). According to test findings, the thermal Flir One Gen 3 camera measures temperature with an average accuracy of 99.80%. By merging three transfer learning models VGG16, ResNet50, and VGG16-Alt—our trained CNN model was able to identify the suggested arcing with an accuracy of 94.83%, according to experimental data. Additionally, two CNN architectures were tested for trolley wire wear analysis: with specialized tuning for trolley wire contour identification, EfficientNetB0 attained an accuracy of 87.2%, whereas ensemble ResNet50 obtained 87.48% using 5-Fold Cross-Validation.</p></div>","PeriodicalId":100109,"journal":{"name":"Applied Research","volume":"5 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/appl.70143","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148617318","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"From Seedling to Maturity: AI-Based Mustard Crop Growth Tracking Using UAV Time-Series Images","authors":"Harish Kumar Pamnani","doi":"10.1002/appl.70156","DOIUrl":"https://doi.org/10.1002/appl.70156","url":null,"abstract":"<div>\u0000 \u0000 <p>Accurate monitoring of plant phenology is essential for agricultural decision-making, as deciding the right time of fertilizer application, the maturity of the plant, and climate variation. Traditional manual monitoring often fails to capture the temporal variations across large fields. UAV-based imaging combined with deep learning can provide a solution for automated phenological assessment. In this study, we propose an AI-driven framework on UAV-captured Indian mustard plants to predict the phenological progress. We applied deep learning methods, EfficientNet-B0, and a proposed hybrid model of a Vision Transformer + LSTM to predict phenological growth from UVV-captured images of the Indian mustard plant. Both models were trained under a supervised regression setup with extensive augmentation and optimization strategies. The results of the study show that the ViT + LSTM outperforms the EfficientNet-B0 model in terms of prediction accuracy for mustard plant phenology. The <i>R</i>² = 0.9974, minimal errors MSE = 0.0111, and RMSE = 0.149 indicate that the ViT + LSTM provides more accurate phenological predictions on temporal and spatial dependencies. Correlation Analysis confirmed a strong linear and monotonic relationship with the ground truth, with <i>r</i> = 0.9989 and <i>ρ</i> = 0.9946. Gram-CAM visualization showed that the ViT + LSTM captures the meaning area of the plant. These results were further validated using statistical measures such as Pearson's <i>r</i> and Spearman's <i>ρ</i>, which confirmed the reliability and consistency of the model's predictions.</p></div>","PeriodicalId":100109,"journal":{"name":"Applied Research","volume":"5 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/appl.70156","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148849345","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Liquid Holdup Prediction in Wet Natural Gas Pipelines Using an IASO-Optimized BP Neural Network","authors":"Dong Wang,&nbsp;Gao Zhang,&nbsp;Yajuan Wang,&nbsp;Zhongze Sun","doi":"10.1002/appl.70154","DOIUrl":"https://doi.org/10.1002/appl.70154","url":null,"abstract":"<div>\u0000 \u0000 <p>Liquid holdup is a key parameter for assessing liquid accumulation in undulating pipelines and is essential for ensuring their safe and reliable operation. To improve prediction accuracy, this study proposes an enhanced back propagation neural network (BPNN) optimized by an improved atomic search optimization (IASO) algorithm. The proposed IASO-BP model was compared with several established approaches, including mechanistic models (MBII, BB, and BBE) and neural network models (BP, SSA-BP, and ASO-BP). Results show that the IASO-BP model achieves superior performance, with a root mean squared error (RMSE) of 0.04478 and a mean absolute percentage error (MAPE) of 6.38%, outperforming both traditional mechanistic models and other neural network-based benchmarks. Furthermore, a graphical user interface (GUI) was developed to integrate the model into user-friendly prediction software, providing a practical tool for field applications.</p></div>","PeriodicalId":100109,"journal":{"name":"Applied Research","volume":"5 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/appl.70154","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148849346","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An Improved YOLOv8 Model for Small Traffic Sign Detection Using P2 Feature Fusion and CBAM","authors":"Yuan Wang","doi":"10.1002/appl.70155","DOIUrl":"https://doi.org/10.1002/appl.70155","url":null,"abstract":"<div>\u0000 \u0000 <p>Small traffic sign detection in complex road scenes is still challenging. Traffic signs in the image are usually small in size, weak in visual features, and susceptible to interference from complex backgrounds. Although the YOLO-based model performs well in small target detection tasks, its repeated downsampling process may weaken some fine-grained spatial information, thus affecting the detection accuracy. In order to improve this problem, this study proposes P2C-YOLOv8. The model introduces a P2 high-resolution feature layer to retain more location information related to small traffic signs. At the same time, a CBAM module is embedded before prediction to refine multi-scale features and suppress irrelevant background responses. The experiment was carried out on the TT100K2016 subset. Compared with Faster R-CNN, RetinaNet, YOLOv8 and P2-YOLOv8, P2C-YOLOv8 achieved the highest or joint-highest Precision across the three traffic sign categories, reaching 0.899, 0.839, and 0.901 for indication, prohibitory, and warning signs, respectively. Its F1 score is also higher than YOLOv8. The confusion matrix and visualization results show that the model can reduce the false detection caused by background interference and improve the positioning effect of small traffic signs, so as to achieve more stable recognition performance in a complex road environment.</p></div>","PeriodicalId":100109,"journal":{"name":"Applied Research","volume":"5 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2026-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1002/appl.70155","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148617319","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}