Heying Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Heying Li has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Jianchen Zhang
College of Geography and Environmental Science, Henan University, Kaifeng 475004, China
5 shared publications - Huiling Ma
Key Laboratory of Urban Land Resources Monitoring and Simulation, Ministry of Natural Resources, Shenzhen 518034, China
3 shared publications - Guangxia Wang
College of Geography and Environmental Science, Henan University, Kaifeng 475004, China
2 shared publications - Haohua Zheng
College of Geography and Environmental Science, Henan University, Kaifeng 475004, China
2 shared publications - Xuefei Hong
Key Laboratory of Urban Land Resources Monitoring and Simulation, Ministry of Natural Resources, Shenzhen 518040, China
2 shared publications - Xueye Chen
Key Laboratory of Urban Land Resources Monitoring and Simulation, Ministry of Natural Resources, Shenzhen 518000, China
2 shared publications
Latest Publications
- Open Access
- 25 August 2024
- Published by MDPI AG in ISPRS International Journal of Geo-Information
- Vol. 13 (9) , 300
Road Network Intelligent Selection Method Based on Heterogeneous Graph Attention Neural Network
Selecting road networks in cartographic generalization has consistently posed formidable challenges, driving research toward the application of intelligent models. Despite previous efforts, the accuracy and connectivity preservation in these studies, particularly when dealing with road types of similar sample sizes, still warrant improvement. To address these shortcomings, we introduce a Heterogeneous Graph Attention Network (HAN) for road selection, where the feature masking method is initially utilized to assess the significance of road features. Concentrating on the most relevant features, two meta-paths are introduced within the HAN framework: one for aggregating features of the same road type within the first-order neighborhood, emphasizing local connectivity, and another for extending this aggregation to the second-order neighborhood, capturing a broader spatial context. For a comprehensive evaluation, we use a set of metrics considering both quantitative and qualitative aspects of the road network. On road types with similar sample sizes, the HAN model outperforms other models in both transductive and inductive tasks. Its accuracy (ACC) is higher by 1.62% and 0.67%, and its F1-score is higher by 1.43% and 0.81%, respectively. Additionally, it enhances the overall connectivity of the selected network. In summary, our HAN-based method provides an advanced solution for road network selection, surpassing previous approaches in terms of accuracy and connectivity preservation.
- 25 August 2024
- Published by MDPI AG in ISPRS International Journal of Geo-Information
- Vol. 13 (9) , 300
Open AccessRoad Network Intelligent Selection Method Based on Heterogeneous Graph Attention Neural Network
Selecting road networks in cartographic generalization has consistently posed formidable challenges, driving research toward the application of intelligent models. Despite previous efforts, the accuracy and connectivity preservation in these studies, particularly when dealing with road types of similar sample sizes, still warrant improvement. To address these shortcomings, we introduce a Heterogeneous Graph Attention Network (HAN) for road selection, where the feature masking method is initially utilized to assess the significance of road features. Concentrating on the most relevant features, two meta-paths are introduced within the HAN framework: one for aggregating features of the same road type within the first-order neighborhood, emphasizing local connectivity, and another for extending this aggregation to the second-order neighborhood, capturing a broader spatial context. For a comprehensive evaluation, we use a set of metrics considering both quantitative and qualitative aspects of the road network. On road types with similar sample sizes, the HAN model outperforms other models in both transductive and inductive tasks. Its accuracy (ACC) is higher by 1.62% and 0.67%, and its F1-score is higher by 1.43% and 0.81%, respectively. Additionally, it enhances the overall connectivity of the selected network. In summary, our HAN-based method provides an advanced solution for road network selection, surpassing previous approaches in terms of accuracy and connectivity preservation.
- Open Access
Study on Spatial Distribution Dispersion Evaluation and Driving Forces of Rural Settlements in the Yellow River Basin
The spatial distribution pattern of rural settlements in the Yellow River Basin is scattered and numerous. It is of great significance to study the discrete distribution of rural settlements for achieving high-quality development and promoting rural revitalization strategy. In this paper, we propose an enhanced evaluation model for assessing the spatial distribution dispersion of rural settlements, incorporating the weight of road grade (the road grade refers to the ranking of traffic capacity and importance of a particular type of road, indicating varying levels of time accessibility). We investigate the dispersion characteristics of rural settlements in the Yellow River Basin in 2020, focusing on both county and city scales. Furthermore, we conduct a comprehensive analysis of the spatial differentiation and scale effects of dispersion evaluation outcomes and their driving forces. Our findings reveal the following insights: (1) The road grade significantly influences the dispersion evaluation. When considering road grade in the dispersion calculation, the results align more closely with the actual situation. (2) The dispersion of rural settlements in the Yellow River Basin exhibits a decreasing trend from west to east. Specifically, the dispersion is higher in the upper reaches compared to the middle and lower reaches. Both city and county scales show spatial autocorrelation in dispersion, with a positive spatial correlation observed. High dispersion values cluster in the west, while low values concentrate in the east. Notably, the agglomeration degree is more pronounced at the county scale than at the city scale, highlighting more localized patterns of agglomeration and dispersion. (3) The multiscale geographically weighted regression model emerges as the optimal model for analyzing the driving forces of dispersion. At the city scale, factors such as river density, road density, and rural economy negatively impact dispersion. However, at the county scale, average elevation and rural economy positively affect dispersion, whereas river density, road density, and rural population density have a negative influence. By incorporating the weight of road grade into our evaluation model, we provide a more nuanced understanding of the spatial distribution dispersion of rural settlements in the Yellow River Basin. Our findings offer valuable insights for policymakers and planners seeking to optimize rural settlement patterns and promote sustainable rural development.
Open AccessStudy on Spatial Distribution Dispersion Evaluation and Driving Forces of Rural Settlements in the Yellow River Basin
The spatial distribution pattern of rural settlements in the Yellow River Basin is scattered and numerous. It is of great significance to study the discrete distribution of rural settlements for achieving high-quality development and promoting rural revitalization strategy. In this paper, we propose an enhanced evaluation model for assessing the spatial distribution dispersion of rural settlements, incorporating the weight of road grade (the road grade refers to the ranking of traffic capacity and importance of a particular type of road, indicating varying levels of time accessibility). We investigate the dispersion characteristics of rural settlements in the Yellow River Basin in 2020, focusing on both county and city scales. Furthermore, we conduct a comprehensive analysis of the spatial differentiation and scale effects of dispersion evaluation outcomes and their driving forces. Our findings reveal the following insights: (1) The road grade significantly influences the dispersion evaluation. When considering road grade in the dispersion calculation, the results align more closely with the actual situation. (2) The dispersion of rural settlements in the Yellow River Basin exhibits a decreasing trend from west to east. Specifically, the dispersion is higher in the upper reaches compared to the middle and lower reaches. Both city and county scales show spatial autocorrelation in dispersion, with a positive spatial correlation observed. High dispersion values cluster in the west, while low values concentrate in the east. Notably, the agglomeration degree is more pronounced at the county scale than at the city scale, highlighting more localized patterns of agglomeration and dispersion. (3) The multiscale geographically weighted regression model emerges as the optimal model for analyzing the driving forces of dispersion. At the city scale, factors such as river density, road density, and rural economy negatively impact dispersion. However, at the county scale, average elevation and rural economy positively affect dispersion, whereas river density, road density, and rural population density have a negative influence. By incorporating the weight of road grade into our evaluation model, we provide a more nuanced understanding of the spatial distribution dispersion of rural settlements in the Yellow River Basin. Our findings offer valuable insights for policymakers and planners seeking to optimize rural settlement patterns and promote sustainable rural development.
- Open Access
- 17 May 2024
- Published by MDPI AG in Preprints.org
Road Network Intelligent Selection Method Based on Heterogeneous Graph Attention Neural Network
Selecting road networks in cartographic generalization has consistently posed formidable challenges, driving research towards the application of intelligent models. Despite previous efforts, the accuracy and connectivity preservation in these studies, particularly when dealing with road types of similar sample sizes, still warrant improvement. To address these shortcomings, we introduce a Heterogeneous Graph Attention Network (HAN) for road selection, where feature masking method is initially utilized to assess the significance of road features. Concentrating on the most relevant features, two meta-paths are introduced within the HAN framework: one for aggregating features of the same road type within the first-order neighborhood, emphasizing local connectivity, and another extending this aggregation to the second-order neighborhood, capturing a broader spatial context. For a comprehensive evaluation, we introduce a set of metrics considering both quantitative and qualitative aspects of the road network. Compared to traditional neural networks, our HAN model demonstrates superior accuracy and F1-score, particularly for road types with comparable samples. Additionally, it enhances the overall connectivity of the selected network. In summary, our HAN-based method provides an advanced solution for road network selection, surpassing previous approaches in terms of accuracy and connectivity preservation.
- 17 May 2024
- Published by MDPI AG in Preprints.org
Open AccessRoad Network Intelligent Selection Method Based on Heterogeneous Graph Attention Neural Network
Selecting road networks in cartographic generalization has consistently posed formidable challenges, driving research towards the application of intelligent models. Despite previous efforts, the accuracy and connectivity preservation in these studies, particularly when dealing with road types of similar sample sizes, still warrant improvement. To address these shortcomings, we introduce a Heterogeneous Graph Attention Network (HAN) for road selection, where feature masking method is initially utilized to assess the significance of road features. Concentrating on the most relevant features, two meta-paths are introduced within the HAN framework: one for aggregating features of the same road type within the first-order neighborhood, emphasizing local connectivity, and another extending this aggregation to the second-order neighborhood, capturing a broader spatial context. For a comprehensive evaluation, we introduce a set of metrics considering both quantitative and qualitative aspects of the road network. Compared to traditional neural networks, our HAN model demonstrates superior accuracy and F1-score, particularly for road types with comparable samples. Additionally, it enhances the overall connectivity of the selected network. In summary, our HAN-based method provides an advanced solution for road network selection, surpassing previous approaches in terms of accuracy and connectivity preservation.
- Open Access
Surface Water Resource Accessibility Assessment of Rural Settlements in the Yellow River Basin
Analyzing the spatial relationship between humans and water is crucial for regional development and water allocation schemes, particularly in the face of extreme water scarcity in the Yellow River Basin. A quantitative evaluation model of surface water resource accessibility (SWRA) has been developed, with rural settlements serving as the research unit. This model is built upon three key dimensions: topography, distance, and surface water resources within the Yellow River Basin. The results show that: (1) The SWRA range spans from 0.13 to 0.88, with an average value of 0.47 and a standard deviation of 0.05. Higher SWRA values are concentrated in the eastern and western regions, while lower values are predominantly found in the central area. (2) The gradient of SWRA across the 12 catchments, from low to high, is as follows: Sanmenxia station, Lanzhou station, Shizuishan station, Longmen station, Tongguan station, Toudaoguai station, Xiaolangdi station, Huayuankou station, Lijin station, Gaocun station, Ai Shan station, and Tangnaihai station. (3) At the city scale, the SWRA values are generally higher in the eastern areas of 10 cities, with one exception being higher in the west. Conversely, in the western areas of nine cities, the SWRA values are lower. The remaining cities exhibit SWRA values at a medium level. The correlation coefficient between primary industry gross domestic product (GDP) and SWRA is 0.271 (N = 56, Sig = 0.043, in 0.05 level, the correlation is significant), which confirms that SWRA serves as a factor influencing GDP and is appropriately designed for assessing water accessibility. Consequently, managers can utilize SWRA results to make informed decisions regarding regional development and water allocation.
Open AccessSurface Water Resource Accessibility Assessment of Rural Settlements in the Yellow River Basin
Analyzing the spatial relationship between humans and water is crucial for regional development and water allocation schemes, particularly in the face of extreme water scarcity in the Yellow River Basin. A quantitative evaluation model of surface water resource accessibility (SWRA) has been developed, with rural settlements serving as the research unit. This model is built upon three key dimensions: topography, distance, and surface water resources within the Yellow River Basin. The results show that: (1) The SWRA range spans from 0.13 to 0.88, with an average value of 0.47 and a standard deviation of 0.05. Higher SWRA values are concentrated in the eastern and western regions, while lower values are predominantly found in the central area. (2) The gradient of SWRA across the 12 catchments, from low to high, is as follows: Sanmenxia station, Lanzhou station, Shizuishan station, Longmen station, Tongguan station, Toudaoguai station, Xiaolangdi station, Huayuankou station, Lijin station, Gaocun station, Ai Shan station, and Tangnaihai station. (3) At the city scale, the SWRA values are generally higher in the eastern areas of 10 cities, with one exception being higher in the west. Conversely, in the western areas of nine cities, the SWRA values are lower. The remaining cities exhibit SWRA values at a medium level. The correlation coefficient between primary industry gross domestic product (GDP) and SWRA is 0.271 (N = 56, Sig = 0.043, in 0.05 level, the correlation is significant), which confirms that SWRA serves as a factor influencing GDP and is appropriately designed for assessing water accessibility. Consequently, managers can utilize SWRA results to make informed decisions regarding regional development and water allocation.
- Open Access
- 26 January 2024
- Published by MDPI AG in Preprints.org
Surface Water Resources Accessibility Assessment of Rural Settlement in the Yellow River Basin
Under the extreme shortage of water resources in the Yellow River Basin, it is of great signifi-cance to analyze the spatial distribution relationship between rural settlement and water re-sources for the coordinated development of human and water resources in the Yellow River Ba-sin. However, the water allocation scheme depended on statistical data in the previous work, not on rural settlements in space. Furthermore, previous researches seldom focus on the evaluation of surface water resources accessibility according to the spatial distribution characteristics of ru-ral settlements and water resources, especially, in the Yellow River Basin and using rural set-tlement as unit. Therefore, a quantitative evaluation model of surface water resources accessibil-ity is constructed by using the GIS network analysis method, taking rural settlement as the re-search unit, based on three dimensions of topography, distance, and surface water resources, at the Yellow River basin scale. The results show that: (1) The surface runoff of rural settlement in the Yellow River basin is mostly less than 0.05 km/m2, reaching 99.83%. In other words, the Yel-low River Basin is extremely short of water resources. The surface water resources accessibility is affected by distance, topography, and surface water resources. The evaluation results show that the range of surface water resources accessibility is 0.13~0.88, with an average value of 0.47 and a standard deviation of 0.05. The High surface water resources accessibility value in the Yel-low River Basin is in the east and west, and the low value is mainly distributed in the middle. (2) The range from low to high of surface water resources accessibility in 12 catchments is Sanmen-xia station, Lanzhou station, Shizuishan station, Longmen station, Tongguan station, Toudaoguai station, Xiaolangdi station, Huayuankou station, Lijin station, Gaocun station, Ai Shan station and Tangnaihai station. (3) The value of surface water resources accessibility at the city scale is higher in the east of 10 cities and in the west of 1 city. However, it is lower in the west of 9 cities. And the values of the other cities are at a medium level. The correlation coefficient with primary industry GDP and surface water resources accessibility is 0.271 (N=56, Sig=0.043, in 0.05 level, the correlation is significant), which verifies the rationality of surface water resources accessibility model to some extent.
- 26 January 2024
- Published by MDPI AG in Preprints.org
Open AccessSurface Water Resources Accessibility Assessment of Rural Settlement in the Yellow River Basin
Under the extreme shortage of water resources in the Yellow River Basin, it is of great signifi-cance to analyze the spatial distribution relationship between rural settlement and water re-sources for the coordinated development of human and water resources in the Yellow River Ba-sin. However, the water allocation scheme depended on statistical data in the previous work, not on rural settlements in space. Furthermore, previous researches seldom focus on the evaluation of surface water resources accessibility according to the spatial distribution characteristics of ru-ral settlements and water resources, especially, in the Yellow River Basin and using rural set-tlement as unit. Therefore, a quantitative evaluation model of surface water resources accessibil-ity is constructed by using the GIS network analysis method, taking rural settlement as the re-search unit, based on three dimensions of topography, distance, and surface water resources, at the Yellow River basin scale. The results show that: (1) The surface runoff of rural settlement in the Yellow River basin is mostly less than 0.05 km/m2, reaching 99.83%. In other words, the Yel-low River Basin is extremely short of water resources. The surface water resources accessibility is affected by distance, topography, and surface water resources. The evaluation results show that the range of surface water resources accessibility is 0.13~0.88, with an average value of 0.47 and a standard deviation of 0.05. The High surface water resources accessibility value in the Yel-low River Basin is in the east and west, and the low value is mainly distributed in the middle. (2) The range from low to high of surface water resources accessibility in 12 catchments is Sanmen-xia station, Lanzhou station, Shizuishan station, Longmen station, Tongguan station, Toudaoguai station, Xiaolangdi station, Huayuankou station, Lijin station, Gaocun station, Ai Shan station and Tangnaihai station. (3) The value of surface water resources accessibility at the city scale is higher in the east of 10 cities and in the west of 1 city. However, it is lower in the west of 9 cities. And the values of the other cities are at a medium level. The correlation coefficient with primary industry GDP and surface water resources accessibility is 0.271 (N=56, Sig=0.043, in 0.05 level, the correlation is significant), which verifies the rationality of surface water resources accessibility model to some extent.