Due to the challenges faced by current deep learning models in training, such as incomplete data coverage and difficulty in fully reflecting all actual scenarios, this study explores innovative approaches to data collection and annotation strategies. The aim is to fundamentally solve the problem of performance degradation of models in unknown scenarios by improving data diversity and quality. A refined data collection framework has been designed, combined with feature extraction and representation methods in dynamic scenes, effectively enhancing the adaptability and robustness of the model. In order to further verify the effectiveness of the strategy, this study introduces the Visual Auxiliary Graph Neural Network (VA-GNN) and constructs an innovative model for collaborative control of intelligent transportation systems. The experimental results show that with the increase of training iterations, the VA-GNN model and collaborative control strategy have achieved significant results in reducing the average waiting time of vehicles and the number of vehicles in the same lane queue, which is a qualitative leap compared to traditional methods.
{"title":"Vision-assisted graph neural network for collaborative control in intelligent transportation systems","authors":"Shanqian Lin , Xincheng Wu , Jing Zhao , Xiaohong Zhuang","doi":"10.1016/j.aej.2025.12.006","DOIUrl":"10.1016/j.aej.2025.12.006","url":null,"abstract":"<div><div>Due to the challenges faced by current deep learning models in training, such as incomplete data coverage and difficulty in fully reflecting all actual scenarios, this study explores innovative approaches to data collection and annotation strategies. The aim is to fundamentally solve the problem of performance degradation of models in unknown scenarios by improving data diversity and quality. A refined data collection framework has been designed, combined with feature extraction and representation methods in dynamic scenes, effectively enhancing the adaptability and robustness of the model. In order to further verify the effectiveness of the strategy, this study introduces the Visual Auxiliary Graph Neural Network (VA-GNN) and constructs an innovative model for collaborative control of intelligent transportation systems. The experimental results show that with the increase of training iterations, the VA-GNN model and collaborative control strategy have achieved significant results in reducing the average waiting time of vehicles and the number of vehicles in the same lane queue, which is a qualitative leap compared to traditional methods.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"134 ","pages":"Pages 253-262"},"PeriodicalIF":6.8,"publicationDate":"2025-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145735272","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-12DOI: 10.1016/j.aej.2025.12.001
Peifan Li , Jinluan Ren
Audio-visual deepfake detection presents significant computational challenges in achieving precise temporal boundary localization beyond traditional binary classification approaches. This study presents CCFormer, a cascaded optimization framework that integrates ConvNeXt-V2 visual forgery detection with CrossFormer cross-modal localization for precise temporal forgery localization. The framework employs a two-stage strategy where ConvNeXt-V2 performs efficient suspicious segment screening through multi-scale spatiotemporal feature extraction, while CrossFormer achieves frame-level precision through multi-head cross-modal attention mechanisms for optimal audio-visual feature alignment. Experiments on the LAV-DF dataset demonstrate that CCFormer achieving 96.30 % [email protected] and 84.96 % [email protected] The framework achieves inference time of 23.4 ms per video, representing 58.1 % improvement over conventional end-to-end architectures. Ablation studies reveal that the CrossFormer module increases detection performance in high-precision IoU intervals by 153.4 % compared to the baseline methods. The optimization framework successfully transforms coarse-grained binary classification into precise temporal boundary localization,
{"title":"CCFormer: A cascaded transformer framework for precise temporal audio-visual deepfake localization","authors":"Peifan Li , Jinluan Ren","doi":"10.1016/j.aej.2025.12.001","DOIUrl":"10.1016/j.aej.2025.12.001","url":null,"abstract":"<div><div>Audio-visual deepfake detection presents significant computational challenges in achieving precise temporal boundary localization beyond traditional binary classification approaches. This study presents CCFormer, a cascaded optimization framework that integrates ConvNeXt-V2 visual forgery detection with CrossFormer cross-modal localization for precise temporal forgery localization. The framework employs a two-stage strategy where ConvNeXt-V2 performs efficient suspicious segment screening through multi-scale spatiotemporal feature extraction, while CrossFormer achieves frame-level precision through multi-head cross-modal attention mechanisms for optimal audio-visual feature alignment. Experiments on the LAV-DF dataset demonstrate that CCFormer achieving 96.30 % [email protected] and 84.96 % [email protected] The framework achieves inference time of 23.4 ms per video, representing 58.1 % improvement over conventional end-to-end architectures. Ablation studies reveal that the CrossFormer module increases detection performance in high-precision IoU intervals by 153.4 % compared to the baseline methods. The optimization framework successfully transforms coarse-grained binary classification into precise temporal boundary localization,</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"134 ","pages":"Pages 224-237"},"PeriodicalIF":6.8,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145735286","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-12DOI: 10.1016/j.aej.2025.12.014
Ola Hassan , Ahmed Yasir , Aly Hassan , Magdy Shaheen
City centres are urban locations with a high concentration of economic, administrative, and cultural activities. As urbanisation continues to expand, traffic disruptions rise, adversely impacting the quality of life and the environment. In some developing countries, city centres have become overcrowded and lost their image. The paper proposes a new mobility concept for these city centres, as a primary step towards achieving smart and sustainable mobility. The new mobility aims to meet travel needs, ensure accessibility, ensure safety for all road users, and improve environmental quality. Smart mobility can then lead to an advanced transportation system with innovative technologies. Ultimately, sustainable mobility embodies a long-term vision for future generations of a cleaner, fairer, and more resilient world, without compromising the ability to meet their needs. The new mobility philosophy highlights the pivotal role of urban planners in its successful implementation. The proposed concept is applied to the Alexandria city centre as a case study, only to demonstrate its practicality as a tool for smart mobility. Therefore, two planning scenarios containing different measures are formulated and evaluated using micro-simulation and analysed with sustainable indicators. The application demonstrated that new mobility could be introduced in the traditional Alexandria city centre by 2028.
{"title":"New mobility in Central Areas of Smart Cities, Alexandria as an applied example","authors":"Ola Hassan , Ahmed Yasir , Aly Hassan , Magdy Shaheen","doi":"10.1016/j.aej.2025.12.014","DOIUrl":"10.1016/j.aej.2025.12.014","url":null,"abstract":"<div><div>City centres are urban locations with a high concentration of economic, administrative, and cultural activities. As urbanisation continues to expand, traffic disruptions rise, adversely impacting the quality of life and the environment. In some developing countries, city centres have become overcrowded and lost their image. The paper proposes a new mobility concept for these city centres, as a primary step towards achieving smart and sustainable mobility. The new mobility aims to meet travel needs, ensure accessibility, ensure safety for all road users, and improve environmental quality. Smart mobility can then lead to an advanced transportation system with innovative technologies. Ultimately, sustainable mobility embodies a long-term vision for future generations of a cleaner, fairer, and more resilient world, without compromising the ability to meet their needs. The new mobility philosophy highlights the pivotal role of urban planners in its successful implementation. The proposed concept is applied to the Alexandria city centre as a case study, only to demonstrate its practicality as a tool for smart mobility. Therefore, two planning scenarios containing different measures are formulated and evaluated using micro-simulation and analysed with sustainable indicators. The application demonstrated that new mobility could be introduced in the traditional Alexandria city centre by 2028.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"134 ","pages":"Pages 238-252"},"PeriodicalIF":6.8,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145735287","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-12DOI: 10.1016/j.aej.2025.11.034
Xiao Gao
In the object detection of e-commerce logistics images, traditional models generally face a core contradiction: ”pursuing detection accuracy will sacrifice inference speed, while ensuring real-time performance will compromise accuracy”. To address this issue, this paper proposes a lightweight object detection model LDMamTrack based on the Mamba architecture, which achieves the coordinated optimization of speed and accuracy through three core modules: The LDMambaBlock adopts Hilbert Scan linearization and Linear Deformable Convolution (LDConv), enabling long-range feature modeling with linear complexity while adapting to deformed objects; The Simple Stem replaces the traditional large-kernel convolution with stacked small-kernel convolutions, realizing lightweight extraction of initial features of small objects; The Vision Clue Merge (VCM) module reduces redundant computation and optimizes feature transmission efficiency through dimension splitting and normalization removal design. Experimental results show that LDMamTrack achieves an mAP50 of 78.3% and an mAP50-95 of 52.0% on the LOCO dataset, and an mAP50 of 88.9% and an mAP50-95 of 56.0% on the SKU-110K dataset. Meanwhile, the model’s inference speed reaches 45 FPS. LDMamTrack can meet the needs of real-time detection in e-commerce logistics sorting lines and accurate warehouse inventory, providing technical support for the intelligent upgrading of logistics automation systems.
{"title":"LDMamTrack: A lightweight object detection model based on Mamba architecture for e-commerce logistics images","authors":"Xiao Gao","doi":"10.1016/j.aej.2025.11.034","DOIUrl":"10.1016/j.aej.2025.11.034","url":null,"abstract":"<div><div>In the object detection of e-commerce logistics images, traditional models generally face a core contradiction: ”pursuing detection accuracy will sacrifice inference speed, while ensuring real-time performance will compromise accuracy”. To address this issue, this paper proposes a lightweight object detection model LDMamTrack based on the Mamba architecture, which achieves the coordinated optimization of speed and accuracy through three core modules: The LDMambaBlock adopts Hilbert Scan linearization and Linear Deformable Convolution (LDConv), enabling long-range feature modeling with linear complexity while adapting to deformed objects; The Simple Stem replaces the traditional large-kernel convolution with stacked small-kernel convolutions, realizing lightweight extraction of initial features of small objects; The Vision Clue Merge (VCM) module reduces redundant computation and optimizes feature transmission efficiency through dimension splitting and normalization removal design. Experimental results show that LDMamTrack achieves an mAP50 of 78.3% and an mAP50-95 of 52.0% on the LOCO dataset, and an mAP50 of 88.9% and an mAP50-95 of 56.0% on the SKU-110K dataset. Meanwhile, the model’s inference speed reaches 45 FPS. LDMamTrack can meet the needs of real-time detection in e-commerce logistics sorting lines and accurate warehouse inventory, providing technical support for the intelligent upgrading of logistics automation systems.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"134 ","pages":"Pages 210-223"},"PeriodicalIF":6.8,"publicationDate":"2025-12-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145735285","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-10DOI: 10.1016/j.aej.2025.12.010
Khalid Zaman , Rafiullah Khan , Gan Zengkang , Sajjad Ullah Khan , Farman Ali , Tariq Hussain
This research endeavours to enhance road safety by developing an accurate driver emotion recognition system. A novel model is introduced, incorporating transfer learning principles alongside NasNet-Large CNN and Faster R-CNN, specifically designed for Driver Facial Expression (DFE) analysis. The primary objective is to bolster the recognition accuracy of Driver Facial Expression Recognition (DFER). A noteworthy improvement in the accuracy and efficiency of facial detection is attained by customizing the Faster R-CNN learning module with the Inception V3 model. The capability to accurately detect emotions is of paramount importance, as it facilitates timely interventions to avert potential accidents. To address the challenges associated with DFER accuracy in low-resolution images, this research deploys a myriad of deep learning methodologies. Through a meticulous analysis, the study identifies and implements feasible and superior solutions to enhance DFER accuracy. Additionally, the inherent constraints of low-resolution images are mitigated through the strategic application of data augmentation techniques. The evaluation of this research showcases impressive accuracy levels across diverse datasets, including JAFFE, CK+ , FER-2013, and DFERCD. These findings bear substantial implications for enhancing Advanced Driver Assistance Systems (ADAS) and contribute substantially to the overarching realm of road safety.
{"title":"Accurately recognizing driver emotions through using CNN fused features and NasNet-large model","authors":"Khalid Zaman , Rafiullah Khan , Gan Zengkang , Sajjad Ullah Khan , Farman Ali , Tariq Hussain","doi":"10.1016/j.aej.2025.12.010","DOIUrl":"10.1016/j.aej.2025.12.010","url":null,"abstract":"<div><div>This research endeavours to enhance road safety by developing an accurate driver emotion recognition system. A novel model is introduced, incorporating transfer learning principles alongside NasNet-Large CNN and Faster R-CNN, specifically designed for Driver Facial Expression (DFE) analysis. The primary objective is to bolster the recognition accuracy of Driver Facial Expression Recognition (DFER). A noteworthy improvement in the accuracy and efficiency of facial detection is attained by customizing the Faster R-CNN learning module with the Inception V3 model. The capability to accurately detect emotions is of paramount importance, as it facilitates timely interventions to avert potential accidents. To address the challenges associated with DFER accuracy in low-resolution images, this research deploys a myriad of deep learning methodologies. Through a meticulous analysis, the study identifies and implements feasible and superior solutions to enhance DFER accuracy. Additionally, the inherent constraints of low-resolution images are mitigated through the strategic application of data augmentation techniques. The evaluation of this research showcases impressive accuracy levels across diverse datasets, including JAFFE, CK+ , FER-2013, and DFERCD. These findings bear substantial implications for enhancing Advanced Driver Assistance Systems (ADAS) and contribute substantially to the overarching realm of road safety.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"134 ","pages":"Pages 177-196"},"PeriodicalIF":6.8,"publicationDate":"2025-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145735352","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-10DOI: 10.1016/j.aej.2025.12.003
Gaosong Li , Suai Zhang , Yanqing Lai , Zhenya Wang
Process parameters and active sulphur jointly govern the thermal history and elemental composition of laser cladding coatings, yet their interplay with element mixing in sulphur-containing substrates remains unclear. To solve this problem, a 3D laser cladding model incorporating sulphur was developed using COMSOL Multiphysics. This model simulates the dynamic evolution of elements during laser cladding on the surface of sulphur-containing 45# steel matrix. It predicts the concentration distributions of S, Ni and B, as well as the geometry of the molten pool, under varying laser power, scanning speed and specific energy input conditions.Validation was performed by comparing predicted and measured geometric dimensions, Fe and Ni concentrations. The influence of scanning speed and laser power on element mixing was analyzed through convective mixing time, Peclet number, and flow patterns. Results indicate that when the scanning speed was fixed at 10 mm/s, increasing the laser power from 800 W to 1100 W caused the sulphur concentration to rise from 137 μmol/m³ to 192 μmol/m³ , whilst the concentrations of nickel, boron and chromium decreased from 72 mmol/m³ , 24 mmol/m³ and 133 mmol/m³ to 62 mmol/m³ , 21 mmol/m³ and 114 mmol/m³ , respectively. At constant laser power, sulphur concentration exhibited a non-monotonic variation with scanning speed. Conversely, at a constant laser power, the sulphur concentration first increases and then decreases as the scanning speed increases. At a constant power-speed ratio of 90 J/mm², minimum sulphur concentration and cladding width increased by 45 % and 27 %, respectively, with higher scanning speed also promoting more uniform sulphur distribution. These findings offer quantitative insights for tailoring composition and homogeneity in sulphur-containing laser-cladding layers.
{"title":"Influence of laser power, scanning speed, and power-to-scanning speed ratio on elemental mixing during laser cladding of sulphur-containing matrix","authors":"Gaosong Li , Suai Zhang , Yanqing Lai , Zhenya Wang","doi":"10.1016/j.aej.2025.12.003","DOIUrl":"10.1016/j.aej.2025.12.003","url":null,"abstract":"<div><div>Process parameters and active sulphur jointly govern the thermal history and elemental composition of laser cladding coatings, yet their interplay with element mixing in sulphur-containing substrates remains unclear. To solve this problem, a 3D laser cladding model incorporating sulphur was developed using COMSOL Multiphysics. This model simulates the dynamic evolution of elements during laser cladding on the surface of sulphur-containing 45# steel matrix. It predicts the concentration distributions of S, Ni and B, as well as the geometry of the molten pool, under varying laser power, scanning speed and specific energy input conditions.Validation was performed by comparing predicted and measured geometric dimensions, Fe and Ni concentrations. The influence of scanning speed and laser power on element mixing was analyzed through convective mixing time, Peclet number, and flow patterns. Results indicate that when the scanning speed was fixed at 10 mm/s, increasing the laser power from 800 W to 1100 W caused the sulphur concentration to rise from 137 μmol/m³ to 192 μmol/m³ , whilst the concentrations of nickel, boron and chromium decreased from 72 mmol/m³ , 24 mmol/m³ and 133 mmol/m³ to 62 mmol/m³ , 21 mmol/m³ and 114 mmol/m³ , respectively. At constant laser power, sulphur concentration exhibited a non-monotonic variation with scanning speed. Conversely, at a constant laser power, the sulphur concentration first increases and then decreases as the scanning speed increases. At a constant power-speed ratio of 90 J/mm², minimum sulphur concentration and cladding width increased by 45 % and 27 %, respectively, with higher scanning speed also promoting more uniform sulphur distribution. These findings offer quantitative insights for tailoring composition and homogeneity in sulphur-containing laser-cladding layers.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"134 ","pages":"Pages 163-176"},"PeriodicalIF":6.8,"publicationDate":"2025-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145735283","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-10DOI: 10.1016/j.aej.2025.11.043
Zhongshi Xu
Emotion generation plays a key role in multimodal affective computing, such as in intelligent customer service and virtual assistants. However, most existing methods rely on a single modality or simple modal fusion, failing to fully capture the complex relationships between multimodal information. This results in emotional responses that lack consistency and diversity. To address these issues, we propose an interactive emotion-guided generation method (MMEG) based on multimodal data fusion. MMEG combines graph convolutional networks (GCN) and cross-modal attention mechanisms to capture complex modality dependencies. It also employs generative adversarial networks (GANs) to enhance the quality of generated responses. Experimental results on the IEMOCAP and MELD datasets show that MMEG outperforms existing methods in emotion recognition accuracy, generation quality, and response diversity. The model achieves superior performance in key metrics such as ROUGE-L, BLEU, and F1-score, while also improving emotional consistency. This method offers an effective solution for multimodal emotion generation with broad applications in affective computing and intelligent interaction.
{"title":"Interactive emotion-guided generation with cross-modal attention and graph convolutional networks","authors":"Zhongshi Xu","doi":"10.1016/j.aej.2025.11.043","DOIUrl":"10.1016/j.aej.2025.11.043","url":null,"abstract":"<div><div>Emotion generation plays a key role in multimodal affective computing, such as in intelligent customer service and virtual assistants. However, most existing methods rely on a single modality or simple modal fusion, failing to fully capture the complex relationships between multimodal information. This results in emotional responses that lack consistency and diversity. To address these issues, we propose an interactive emotion-guided generation method (MMEG) based on multimodal data fusion. MMEG combines graph convolutional networks (GCN) and cross-modal attention mechanisms to capture complex modality dependencies. It also employs generative adversarial networks (GANs) to enhance the quality of generated responses. Experimental results on the IEMOCAP and MELD datasets show that MMEG outperforms existing methods in emotion recognition accuracy, generation quality, and response diversity. The model achieves superior performance in key metrics such as ROUGE-L, BLEU, and F1-score, while also improving emotional consistency. This method offers an effective solution for multimodal emotion generation with broad applications in affective computing and intelligent interaction.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"134 ","pages":"Pages 197-209"},"PeriodicalIF":6.8,"publicationDate":"2025-12-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145735282","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-09DOI: 10.1016/j.aej.2025.11.052
Xuefeng Zhao, Haodong Chen, Meng Zhang, Dechun Lu, Zhe Sun
Monitoring of intrusion at construction sites is crucial to ensure personnel safety. However, current systems struggle to automatically determine the spatial extents of hoisting areas and reliably assess worker movements in dynamic construction environments. This study proposes a novel three-dimensional (3D) automatic intrusion detection method that uniquely integrates Building Information Modeling (BIM) data with Real-Time Kinematic (RTK)-enhanced Global Positioning System (GPS) data. The proposed methodology automatically extracts BIM parameters to compute dynamic spatial boundaries of hoisting areas and converts geographic coordinates into a unified 3D virtual environment. The study’s key novelty lies in its rule-based approach that considers both worker location and movement direction to minimize false alarms, addressing a critical limitation in existing position-only detection systems. A dual alert mechanism is implemented, facilitating real-time warnings through intelligent safety helmets for field workers and a comprehensive web-based management interface for supervisors. Validation tests demonstrate substantial improvement in detection accuracy. Proposed rule-based algorithms, which incorporate both spatial position and movement direction analysis, achieved a mean error rate of 8.9 % compared to 42.8 % for traditional position-only methods tested under identical conditions. This represents a 79.2 % reduction in false alarms compared to traditional position-based methods. This scalable solution offers significant potential for enhancing personnel safety management across diverse construction sites and can be extended to monitor multiple workers simultaneously. The system’s integration capabilities make it suitable for widespread adoption in construction safety practices. However, the current implementation is limited to outdoor environments and single-worker scenarios, with future research needed to address indoor applications and multi-worker detection scenarios.
{"title":"Automatic intrusion detection and warning method in the hoisting scenario integrated BIM and GPS","authors":"Xuefeng Zhao, Haodong Chen, Meng Zhang, Dechun Lu, Zhe Sun","doi":"10.1016/j.aej.2025.11.052","DOIUrl":"10.1016/j.aej.2025.11.052","url":null,"abstract":"<div><div>Monitoring of intrusion at construction sites is crucial to ensure personnel safety. However, current systems struggle to automatically determine the spatial extents of hoisting areas and reliably assess worker movements in dynamic construction environments. This study proposes a novel three-dimensional (3D) automatic intrusion detection method that uniquely integrates Building Information Modeling (BIM) data with Real-Time Kinematic (RTK)-enhanced Global Positioning System (GPS) data. The proposed methodology automatically extracts BIM parameters to compute dynamic spatial boundaries of hoisting areas and converts geographic coordinates into a unified 3D virtual environment. The study’s key novelty lies in its rule-based approach that considers both worker location and movement direction to minimize false alarms, addressing a critical limitation in existing position-only detection systems. A dual alert mechanism is implemented, facilitating real-time warnings through intelligent safety helmets for field workers and a comprehensive web-based management interface for supervisors. Validation tests demonstrate substantial improvement in detection accuracy. Proposed rule-based algorithms, which incorporate both spatial position and movement direction analysis, achieved a mean error rate of 8.9 % compared to 42.8 % for traditional position-only methods tested under identical conditions. This represents a 79.2 % reduction in false alarms compared to traditional position-based methods. This scalable solution offers significant potential for enhancing personnel safety management across diverse construction sites and can be extended to monitor multiple workers simultaneously. The system’s integration capabilities make it suitable for widespread adoption in construction safety practices. However, the current implementation is limited to outdoor environments and single-worker scenarios, with future research needed to address indoor applications and multi-worker detection scenarios.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"134 ","pages":"Pages 81-97"},"PeriodicalIF":6.8,"publicationDate":"2025-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145735321","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-12-09DOI: 10.1016/j.aej.2025.11.054
Şenol Bayraktar, Cem Alparslan, Gülşah Pehli̇van
In this study, structural, mechanical, and machinability properties of the Zn-40Al-2Cu-2Si material obtained by gravity die casting were revealed in the casted and heat treated (HTed) states. Heat treatment (HT) was performed using 24 h at 375 ˚C for solutionizing, quenching, and 2 h at 150 ˚C for aging. The microstructure of the casted material was determined to comprise α-Al, η, α+ η, ε (CuZn4) phases and Si particles. After HT, it was observed that the phases precipitated in the inner part of the grains forming the α-Al phase and in the boundary regions. However, no dimensional and relative difference was observed in the Si particles. It was stated that the hardness and tensile strength of the material increased while elongation to fracture decreased with the HT. Machinability tests were performed with PVD-ZrN+TiAlN coated carbide using various cutting speeds (Vs), feed rates (f), and a fixed depth of cut (DoC) in turning. It was revealed that the solutionizing-aging process developed the machining characteristics of the material. While V was inversely proportional to cutting force (F), surface roughness (Ra), and built-up edge (BUE) at constant f, these variables changed directly proportional to the f at constant V. The lowest F, Ra, and built up edge (BUE) were observed at V of 250 m/min and f of 0.04 mm/rev. The highest of these results were observed at V of 150 m/min and f of 0.12 mm/rev. Feed marks, smeared layers, and cavities were seen on the cut surfaces. It was observed that the chip creation turned into a shorter and more brittle structure compared to the as-cast alloy due to solutionizing-aging during machining.
{"title":"Machinability performance evaluation in turning of Zn-40Al-2Cu-2Si alloy: The effect of solutionizing-artificial aging","authors":"Şenol Bayraktar, Cem Alparslan, Gülşah Pehli̇van","doi":"10.1016/j.aej.2025.11.054","DOIUrl":"10.1016/j.aej.2025.11.054","url":null,"abstract":"<div><div>In this study, structural, mechanical, and machinability properties of the Zn-40Al-2Cu-2Si material obtained by gravity die casting were revealed in the casted and heat treated (HTed) states. Heat treatment (HT) was performed using 24 h at 375 ˚C for solutionizing, quenching, and 2 h at 150 ˚C for aging. The microstructure of the casted material was determined to comprise α-Al, η, α+ η, ε (CuZn<sub>4</sub>) phases and Si particles. After HT, it was observed that the phases precipitated in the inner part of the grains forming the α-Al phase and in the boundary regions. However, no dimensional and relative difference was observed in the Si particles. It was stated that the hardness and tensile strength of the material increased while elongation to fracture decreased with the HT. Machinability tests were performed with PVD-ZrN+TiAlN coated carbide using various cutting speeds (<em>Vs</em>), feed rates (<em>f</em>), and a fixed depth of cut (DoC) in turning. It was revealed that the solutionizing-aging process developed the machining characteristics of the material. While <em>V</em> was inversely proportional to cutting force (<em>F</em>), surface roughness (<em>Ra</em>), and built-up edge (BUE) at constant <em>f</em>, these variables changed directly proportional to the <em>f</em> at constant <em>V</em>. The lowest <em>F</em>, <em>Ra,</em> and built up edge (BUE) were observed at <em>V</em> of 250 m/min and <em>f</em> of 0.04 mm/rev. The highest of these results were observed at <em>V</em> of 150 m/min and <em>f</em> of 0.12 mm/rev. Feed marks, smeared layers, and cavities were seen on the cut surfaces. It was observed that the chip creation turned into a shorter and more brittle structure compared to the as-cast alloy due to solutionizing-aging during machining.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"134 ","pages":"Pages 152-162"},"PeriodicalIF":6.8,"publicationDate":"2025-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145735284","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Event detection (ED) seeks to identify and categorize event triggers in unstructured text. A major challenge in Chinese ED is word boundary ambiguity due to the lack of explicit delimiters. Although word granularity provides clearer boundaries and shorter token sequences, it is underexplored due to challenges in granularity alignment and long-tail issues at the syntactic and semantic levels. To address these challenges, we propose Anticipatory prototype and Syntactic-structure Enhanced Event Detection (AntED), the first ED framework at word granularity. AntED incorporates Contrastive Learning-based Out-of-Vocab word representation (CLOV) module, which can generate high-quality embeddings for OOV words in a plug-and-play manner, achieving unified word granularity. We further design a Tail-aware Heterogeneous Graph ATtention Network (THGAT), ensuring equal representation of low-frequency syntactic relations. Moreover, prompt-based Anticipatory Prototype (AnP) learning is used to model event category prototypes and to enhance performance in semantic-scarcity settings. Extensive experiments on three datasets demonstrate that AntED achieves state-of-the-art performance. Especially in the trigger identification subtask, AntED outperforms other methods by over 2% F1 on DuEE and FewFC, and by more than 6% on ACE2005 compared to LLaMA3. These findings highlight the effectiveness of word-granularity ED and encourage further research into its advantages.
{"title":"Towards a word-granularity paradigm for Chinese event detection: Targeting long-tail challenges in syntax and semantics","authors":"Yuewei Zhou , Zhijie Qu , Yongquan Liang , Yifeng Zhang , Jinquan Zhang , Lina Ni","doi":"10.1016/j.aej.2025.11.053","DOIUrl":"10.1016/j.aej.2025.11.053","url":null,"abstract":"<div><div>Event detection (ED) seeks to identify and categorize event triggers in unstructured text. A major challenge in Chinese ED is word boundary ambiguity due to the lack of explicit delimiters. Although word granularity provides clearer boundaries and shorter token sequences, it is underexplored due to challenges in granularity alignment and long-tail issues at the syntactic and semantic levels. To address these challenges, we propose <strong>Ant</strong>icipatory prototype and Syntactic-structure Enhanced <strong>E</strong>vent <strong>D</strong>etection (AntED), the first ED framework at word granularity. AntED incorporates <strong>C</strong>ontrastive <strong>L</strong>earning-based <strong>O</strong>ut-of-<strong>V</strong>ocab word representation (CLOV) module, which can generate high-quality embeddings for OOV words in a plug-and-play manner, achieving unified word granularity. We further design a <strong>T</strong>ail-aware <strong>H</strong>eterogeneous <strong>G</strong>raph <strong>AT</strong>tention Network (THGAT), ensuring equal representation of low-frequency syntactic relations. Moreover, prompt-based <strong>An</strong>ticipatory <strong>P</strong>rototype (AnP) learning is used to model event category prototypes and to enhance performance in semantic-scarcity settings. Extensive experiments on three datasets demonstrate that AntED achieves state-of-the-art performance. Especially in the trigger identification subtask, AntED outperforms other methods by over 2% F1 on DuEE and FewFC, and by more than 6% on ACE2005 compared to LLaMA3. These findings highlight the effectiveness of word-granularity ED and encourage further research into its advantages.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"134 ","pages":"Pages 135-151"},"PeriodicalIF":6.8,"publicationDate":"2025-12-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145735271","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号