Yancheng Ling
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yancheng Ling has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
- Zhenliang Ma
KTH Royal Institute of Technology
7 shared publications - Qi Zhang
School of Transportation and Logistics Engineering, Wuhan University of Technology, Wuhan 430063, China
3 shared publications - Zhenlin Qin
Institute of Physics, Henan Academy of Sciences, Henan, China
3 shared publications - Bangquan Xie
Dongguan Key Laboratory of Intelligent Equipment and Smart Industry, School of Advanced Engineering, Great Bay University, Dongguan, China
2 shared publications - Weng Xiao-Xiong
Department of Traffic Engineering, South China University of Technology, Guangzhou, China, 510640
2 shared publications Xiaoxiong Weng
School of Civil Engineering Transportation, South China University of Technology, Guangzhou, China
2 shared publications
Latest Publications
- Open Access
- 26 October 2026
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
Cross-City Transferability of Urban Green View Index Estimation from Satellite Imagery within Selected European Contexts
Traditional green view index (GVI) estimation depends on street-view imagery, which is geographically uneven, costly, and infrequently updated. This paper proposes SatGreenCNN, a convolutional neural network (CNN) that estimates GVI directly from freely available Sentinel-2A Level-2A imagery using eight interpretable ground indices. We validated the approach on 19 European cities partitioned via principal component analysis and K-means clustering into a training cluster (Group Seen, 8 cities) and a held-out cluster (Group Unseen, 11 cities). The resulting evidence is interpreted as cross-city transferability within selected European contexts. SatGreenCNN reached an R 2 = 0 . 66 on the Group Seen set and a mean R 2 = 0 . 59 (range 0.53–0.65) on the Group Unseen set. A complementary ordinal-grade evaluation classifies unseen-city greenery into low/medium/high categories with an accuracy of 0.83, outperforming random forest, XGBoost, LightGBM, and three alternative variants, on average, while using roughly three orders of magnitude fewer parameters than a ResNet18 baseline.
- 26 October 2026
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
Open AccessCross-City Transferability of Urban Green View Index Estimation from Satellite Imagery within Selected European Contexts
Traditional green view index (GVI) estimation depends on street-view imagery, which is geographically uneven, costly, and infrequently updated. This paper proposes SatGreenCNN, a convolutional neural network (CNN) that estimates GVI directly from freely available Sentinel-2A Level-2A imagery using eight interpretable ground indices. We validated the approach on 19 European cities partitioned via principal component analysis and K-means clustering into a training cluster (Group Seen, 8 cities) and a held-out cluster (Group Unseen, 11 cities). The resulting evidence is interpreted as cross-city transferability within selected European contexts. SatGreenCNN reached an R 2 = 0 . 66 on the Group Seen set and a mean R 2 = 0 . 59 (range 0.53–0.65) on the Group Unseen set. A complementary ordinal-grade evaluation classifies unseen-city greenery into low/medium/high categories with an accuracy of 0.83, outperforming random forest, XGBoost, LightGBM, and three alternative variants, on average, while using roughly three orders of magnitude fewer parameters than a ResNet18 baseline.
- Open Access
- 10 February 2026
- Published by Elsevier in Transportation Research Part C: Emerging Technologies
- Vol. 185
A foundational individual mobility prediction model based on open-source large language models
- 10 February 2026
- Published by Elsevier in Transportation Research Part C: Emerging Technologies
- Vol. 185
Open AccessA foundational individual mobility prediction model based on open-source large language models
- Open Access
- 20 November 2025
- Published by Elsevier in Transportation Research Part C: Emerging Technologies
- Vol. 183
A review of knowledge graph construction using large language models in transportation: Problems, methods, and challenges
- 20 November 2025
- Published by Elsevier in Transportation Research Part C: Emerging Technologies
- Vol. 183
Open AccessA review of knowledge graph construction using large language models in transportation: Problems, methods, and challenges
- Open Access
- 31 January 2025
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2679 (5) , 256-272
Causal Graph Discovery for Urban Bus Operation Delays: A Case Study in Stockholm
Bus delays significantly affect urban public transportation by reducing operational efficiency and incurring high costs. Understanding the causes of these delays is essential for developing targeted mitigation strategies. While traditional research focuses on correlation-based analysis, it often fails to uncover the underlying causal mechanisms. This study examines various causal graph discovery algorithms combined with structural equation models (SEMs) to infer the causal relationships among factors that affect bus delays. These algorithms generate causal graphs for bus delays, revealing the interrelations and impacts of various operational factors. SEM is used to quantify the causal effects. This study evaluates the performance of these algorithms from the perspectives of both the statistical data fitting and the causal relationships generated. A case study is conducted using General Transit Feed Specification (GTFS) data from frequent bus routes in Stockholm, Sweden. The validation results demonstrate the effectiveness of data-driven causal discovery models in identifying causal links, particularly when combined with domain knowledge. The empirical analysis shows the complexity of factors contributing to bus delays, emphasizing the necessity of integrating causality into bus delay analysis. For example, a high correlation between origin delay and bus arrival delay (coefficient = 0.63) does not indicate direct causation, and a strong causation between dwell time and arrival delay does not imply a higher correlation (coefficient = 0.12). Comparing variable importance with linear regression (LR) reveals notable differences; origin delay, which is often overlooked by previous studies, is significant in the causal graph model (standardized coefficient = 0.601) but ranks much lower in LR (standardized coefficient = 0.003). These insights underscore the importance of automated, data-driven causal discovery in enhancing decision-making processes and improving the efficiency and reliability of transit services.
- 31 January 2025
- Published by SAGE Publications in Transportation Research Record: Journal of the Transportation Research Board
- Vol. 2679 (5) , 256-272
Open AccessCausal Graph Discovery for Urban Bus Operation Delays: A Case Study in Stockholm
Bus delays significantly affect urban public transportation by reducing operational efficiency and incurring high costs. Understanding the causes of these delays is essential for developing targeted mitigation strategies. While traditional research focuses on correlation-based analysis, it often fails to uncover the underlying causal mechanisms. This study examines various causal graph discovery algorithms combined with structural equation models (SEMs) to infer the causal relationships among factors that affect bus delays. These algorithms generate causal graphs for bus delays, revealing the interrelations and impacts of various operational factors. SEM is used to quantify the causal effects. This study evaluates the performance of these algorithms from the perspectives of both the statistical data fitting and the causal relationships generated. A case study is conducted using General Transit Feed Specification (GTFS) data from frequent bus routes in Stockholm, Sweden. The validation results demonstrate the effectiveness of data-driven causal discovery models in identifying causal links, particularly when combined with domain knowledge. The empirical analysis shows the complexity of factors contributing to bus delays, emphasizing the necessity of integrating causality into bus delay analysis. For example, a high correlation between origin delay and bus arrival delay (coefficient = 0.63) does not indicate direct causation, and a strong causation between dwell time and arrival delay does not imply a higher correlation (coefficient = 0.12). Comparing variable importance with linear regression (LR) reveals notable differences; origin delay, which is often overlooked by previous studies, is significant in the causal graph model (standardized coefficient = 0.601) but ranks much lower in LR (standardized coefficient = 0.003). These insights underscore the importance of automated, data-driven causal discovery in enhancing decision-making processes and improving the efficiency and reliability of transit services.
- Open Access
- 26 June 2024
- Published by Springer Nature in Transportation
- Vol. 53 (2) , 571-602
Real-time bus arrival delays analysis using seemingly unrelated regression model
To effectively manage and control public transport operations, understanding the various factors that impact bus arrival delays is crucial. However, limited research has focused on a comprehensive analysis of bus delay factors, often relying on single-step delay prediction models that are unable to account for the heterogeneous impacts of spatiotemporal factors along the bus route. To analyze the heterogeneous impact of bus arrival delay factors, the paper proposes a set of regression equations conditional on the bus location. A seemingly unrelated regression equation (SURE) model is developed to estimate the regression coefficients, accounting for potential correlations between regression residuals caused by shared unobserved factors among equations. The model is validated using bus operations data from Stockholm, Sweden. The results highlight the importance of developing stop-specific bus arrival delay models to understand the heterogeneous impact of explanatory variables. The significant factors impacting bus arrival delays are primarily associated with bus operations, such as delays at consecutive upstream stops, dwell time, scheduled travel time, recurrent congestion, and current traffic conditions. Factors like the calendar and weather have significant but marginal impacts on arrival delays. The study suggests that different bus operating management strategies, such as schedule adjustments, route optimization, and real-time monitoring and control, should be tailored to the characteristics of stop sections since the impacts of these factors vary depending on the stop location.
- 26 June 2024
- Published by Springer Nature in Transportation
- Vol. 53 (2) , 571-602
Open AccessReal-time bus arrival delays analysis using seemingly unrelated regression model
To effectively manage and control public transport operations, understanding the various factors that impact bus arrival delays is crucial. However, limited research has focused on a comprehensive analysis of bus delay factors, often relying on single-step delay prediction models that are unable to account for the heterogeneous impacts of spatiotemporal factors along the bus route. To analyze the heterogeneous impact of bus arrival delay factors, the paper proposes a set of regression equations conditional on the bus location. A seemingly unrelated regression equation (SURE) model is developed to estimate the regression coefficients, accounting for potential correlations between regression residuals caused by shared unobserved factors among equations. The model is validated using bus operations data from Stockholm, Sweden. The results highlight the importance of developing stop-specific bus arrival delay models to understand the heterogeneous impact of explanatory variables. The significant factors impacting bus arrival delays are primarily associated with bus operations, such as delays at consecutive upstream stops, dwell time, scheduled travel time, recurrent congestion, and current traffic conditions. Factors like the calendar and weather have significant but marginal impacts on arrival delays. The study suggests that different bus operating management strategies, such as schedule adjustments, route optimization, and real-time monitoring and control, should be tailored to the characteristics of stop sections since the impacts of these factors vary depending on the stop location.
- Open Access
- 20 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (10) , 13277-13290
PedAST-GCN: Fast Pedestrian Crossing Intention Prediction Using Spatial–Temporal Attention Graph Convolution Networks
Accurately and timely predicting pedestrian crossing intentions in real-time is critical for operating intelligent vehicles on roads. Although existing models achieve promising accuracy using complex models and video image data, they are constrained for real-time practical use given the high model complexity, time-consuming data preprocessing, and low-quality image data in the wild. To address these, the paper proposes a Spatial-Temporal Attention Graph Convolution Network model for fast pedestrian crossing intention prediction (PedAST-GCN). It uses a lightweight GCN model as the backbone network with simple but robust graph representations of pedestrian crossing intention modality features, including pedestrian pose, bounding box, and vehicle speeds. The model is validated by comparing it with state-of-the-art models on two large-scale public datasets (JAAD and PIE). The results highlight the better performance of the PedAST-GCN model for pedestrian crossing intention prediction in terms of accuracy and computation times. The ablation analysis confirms the value of the backbone layer and graph design, the designed modality features, the effectiveness of attention mechanisms in capturing long-term dependencies (spatial-temporal attention) and fusing heterogeneous features (modality attention), and the robust performance across various observation lengths and in the presence of noisy data.
- 20 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (10) , 13277-13290
Open AccessPedAST-GCN: Fast Pedestrian Crossing Intention Prediction Using Spatial–Temporal Attention Graph Convolution Networks
Accurately and timely predicting pedestrian crossing intentions in real-time is critical for operating intelligent vehicles on roads. Although existing models achieve promising accuracy using complex models and video image data, they are constrained for real-time practical use given the high model complexity, time-consuming data preprocessing, and low-quality image data in the wild. To address these, the paper proposes a Spatial-Temporal Attention Graph Convolution Network model for fast pedestrian crossing intention prediction (PedAST-GCN). It uses a lightweight GCN model as the backbone network with simple but robust graph representations of pedestrian crossing intention modality features, including pedestrian pose, bounding box, and vehicle speeds. The model is validated by comparing it with state-of-the-art models on two large-scale public datasets (JAAD and PIE). The results highlight the better performance of the PedAST-GCN model for pedestrian crossing intention prediction in terms of accuracy and computation times. The ablation analysis confirms the value of the backbone layer and graph design, the designed modality features, the effectiveness of attention mechanisms in capturing long-term dependencies (spatial-temporal attention) and fusing heterogeneous features (modality attention), and the robust performance across various observation lengths and in the presence of noisy data.
- Open Access
- 11 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (2) , 2089-2100
SA-BiGCN: Bi-Stream Graph Convolution Networks With Spatial Attentions for the Eye Contact Detection in the Wild
Eye contact is essential in transmitting information and intention in the wild environment (e.g., urban streets or parking lots) with mixed vehicles and pedestrians. Compared with the vision image data, the human skeleton data are deemed to be robust to unconstrained surroundings and illumination. However, the skeleton graph-based approaches are mainly used for the action recognition. It is challenging to directly apply them to the eye detection task, which is momentary and dynamic given the complex wild environment. This paper proposes a Bi-stream Spatial Attention Graph Convolution Network (SA-BiGCN) for eye contact detection in the wild. We design a directed, nose-centric skeleton graph to capture relevant and hierarchical information and their interactions. We also propose a Bi-stream graph convolution network model with spatial attention to dynamically extract and fuse skeleton joints and bones information. The model was validated by comparing with state-of-art models on three large-scale public datasets, including JAAD, PIE, and LOOK. The results highlight the accuracy and generalization performance of the proposed SA-BiGCN model in detecting the eye contact in the wild environment. The ablation analysis validates the importance of the skeleton graph design, the spatial attention mechanism in the feature fusion process, as well as the model robustness against noisy skeleton data in terms of part occlusions, block occlusions, random occlusions, and random deviations.
- 11 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (2) , 2089-2100
Open AccessSA-BiGCN: Bi-Stream Graph Convolution Networks With Spatial Attentions for the Eye Contact Detection in the Wild
Eye contact is essential in transmitting information and intention in the wild environment (e.g., urban streets or parking lots) with mixed vehicles and pedestrians. Compared with the vision image data, the human skeleton data are deemed to be robust to unconstrained surroundings and illumination. However, the skeleton graph-based approaches are mainly used for the action recognition. It is challenging to directly apply them to the eye detection task, which is momentary and dynamic given the complex wild environment. This paper proposes a Bi-stream Spatial Attention Graph Convolution Network (SA-BiGCN) for eye contact detection in the wild. We design a directed, nose-centric skeleton graph to capture relevant and hierarchical information and their interactions. We also propose a Bi-stream graph convolution network model with spatial attention to dynamically extract and fuse skeleton joints and bones information. The model was validated by comparing with state-of-art models on three large-scale public datasets, including JAAD, PIE, and LOOK. The results highlight the accuracy and generalization performance of the proposed SA-BiGCN model in detecting the eye contact in the wild environment. The ablation analysis validates the importance of the skeleton graph design, the spatial attention mechanism in the feature fusion process, as well as the model robustness against noisy skeleton data in terms of part occlusions, block occlusions, random occlusions, and random deviations.
- Open Access
- 31 August 2023
- Published by Faculty of Transport and Traffic Sciences in Promet - Traffic&Transportation
- Vol. 35 (4) , 567-582
Efficient and Robust Driver Fatigue Detection Framework Based on the Visual Analysis of Eye States
Fatigue detection based on vision is widely employed in vehicles due to its real-time and reliable detection results. With the coronavirus disease (COVID-19) outbreak, many proposed detection systems based on facial characteristics would be unreliable due to the face covering with the mask. In this paper, we propose a robust visual-based fatigue detection system for monitoring drivers, which is robust regarding the coverings of masks, changing illumination and head movement of drivers. Our system has three main modules: face key point alignment, fatigue feature extraction and fatigue measurement based on fused features. The innovative core techniques are described as follows: (1) a robust key point alignment algorithm by fusing global face information and regional eye information, (2) dynamic threshold methods to extract fatigue characteristics and (3) a stable fatigue measurement based on fusing percentage of eyelid closure (PERCLOS) and proportion of long closure duration blink (PLCDB). The excellent performance of our proposed algorithm and methods are verified in experiments. The experimental results show that our key point alignment algorithm is robust to different scenes, and the performance of our proposed fatigue measurement is more reliable due to the fusion of PERCLOS and PLCDB.
- 31 August 2023
- Published by Faculty of Transport and Traffic Sciences in Promet - Traffic&Transportation
- Vol. 35 (4) , 567-582
Open AccessEfficient and Robust Driver Fatigue Detection Framework Based on the Visual Analysis of Eye States
Fatigue detection based on vision is widely employed in vehicles due to its real-time and reliable detection results. With the coronavirus disease (COVID-19) outbreak, many proposed detection systems based on facial characteristics would be unreliable due to the face covering with the mask. In this paper, we propose a robust visual-based fatigue detection system for monitoring drivers, which is robust regarding the coverings of masks, changing illumination and head movement of drivers. Our system has three main modules: face key point alignment, fatigue feature extraction and fatigue measurement based on fused features. The innovative core techniques are described as follows: (1) a robust key point alignment algorithm by fusing global face information and regional eye information, (2) dynamic threshold methods to extract fatigue characteristics and (3) a stable fatigue measurement based on fusing percentage of eyelid closure (PERCLOS) and proportion of long closure duration blink (PLCDB). The excellent performance of our proposed algorithm and methods are verified in experiments. The experimental results show that our key point alignment algorithm is robust to different scenes, and the performance of our proposed fatigue measurement is more reliable due to the fusion of PERCLOS and PLCDB.
- Open Access
- 28 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 67219-67231
Driver Eye Location and State Estimation Based on a Robust Model and Data Augmentation
Eye state evaluation is crucial for vision-based driver fatigue detection. With the outbreak of COVID-19, many proposed models for eye location and state evaluation based on facial landmarks are unreliable due to mask coverings. In this paper, we proposed a robust facial landmark location model for eye location and state evaluation. First, we develop an existing lightweight face alignment model for eye key point locations that is robust in large poses. Then, to develop the performance of our model in a complex driving environment such as an environment with mask coverings, changing illumination, etc., we design a method to augment the training data set based on the original landmark data set without any extra cost. Finally, some facial landmarks around the eyes are extracted, and the eye aspect ratio (EAR) is introduced to evaluate the eye state based on eye key points. The experiment shows that our model achieves significantly improved landmark location performance on a driving simulation data set due to data augmentation. We tested our model on the BioID data set to measure the eye state evaluation performance, and the results showed that our model obtained satisfactory performance with an accuracy of approximately 97.7%. Further testing on the driving simulation data set shows that our model is robust in different driving scenarios with an average accuracy of approximately 93.9%.
- 28 April 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 9, 67219-67231
Open AccessDriver Eye Location and State Estimation Based on a Robust Model and Data Augmentation
Eye state evaluation is crucial for vision-based driver fatigue detection. With the outbreak of COVID-19, many proposed models for eye location and state evaluation based on facial landmarks are unreliable due to mask coverings. In this paper, we proposed a robust facial landmark location model for eye location and state evaluation. First, we develop an existing lightweight face alignment model for eye key point locations that is robust in large poses. Then, to develop the performance of our model in a complex driving environment such as an environment with mask coverings, changing illumination, etc., we design a method to augment the training data set based on the original landmark data set without any extra cost. Finally, some facial landmarks around the eyes are extracted, and the eye aspect ratio (EAR) is introduced to evaluate the eye state based on eye key points. The experiment shows that our model achieves significantly improved landmark location performance on a driving simulation data set due to data augmentation. We tested our model on the BioID data set to measure the eye state evaluation performance, and the results showed that our model obtained satisfactory performance with an accuracy of approximately 97.7%. Further testing on the driving simulation data set shows that our model is robust in different driving scenarios with an average accuracy of approximately 93.9%.