[go: up one dir, main page]
Next Issue
Volume 12, October
Previous Issue
Volume 12, August
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
6.9
2.8
Journals
IJGI
Volume 12
Issue 9
share announcement

ISPRS Int. J. Geo-Inf., Volume 12, Issue 9 (September 2023) – 39 articles

Cover Story (view full-size image): The intelligent integration of earth observation (EO) data, primarily represented in the form of raster data cubes using knowledge graphs (KGs), plays a crucial role in managing geospatial data heterogeneity and enhancing semantics. This paper introduces a framework that conceptually defines a semantic model for raster data cubes, extending GeoSPARQL ontology. This model combines raster data cube semantics with feature-based geometry and spatial relationships, enabling spatiotemporal queries using SPARQL through ontological concepts. The implementation of this framework involves virtual querying, which refers to dynamically constructing the geospatial knowledge graph during the querying process. This approach eliminates the need to pre-translate all data into a KG, thereby reducing redundancy and cutting storage and processing costs. View this paper
  • Issues are regarded as officially published after their release is announced to the table of contents alert mailing list.
  • You may sign up for e-mail alerts to receive table of contents of newly released issues.
  • PDF is the official format for papers published in both, html and pdf forms. To view the papers in pdf format, click on the "PDF Full-text" link, and use the free Adobe Reader to open them.
Order results
Result details
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
31 pages, 17346 KiB  
Article
Spatiotemporal Analytics of Environmental Sounds and Influencing Factors Based on Urban Sensor Network Data
by Yanjie Zhao, Jin Cheng, Shaohua Wang, Lei Qin and Xueyan Zhang
ISPRS Int. J. Geo-Inf. 2023, 12(9), 386; https://doi.org/10.3390/ijgi12090386 - 21 Sep 2023
Viewed by 1654
Abstract
Urban construction has accelerated the deterioration of the urban sound environment, which has constrained urban development and harmed people’s health. This study aims to explore the spatiotemporal patterns of environmental sound and determine the influencing factors on the spatial differentiation of sound, thus [...] Read more.
Urban construction has accelerated the deterioration of the urban sound environment, which has constrained urban development and harmed people’s health. This study aims to explore the spatiotemporal patterns of environmental sound and determine the influencing factors on the spatial differentiation of sound, thus supporting sustainable urban planning and decision-making. Fine-grained sound data are used in most urban sound-related research, but such data are difficult to obtain. For this problem, this study analyzed sound trends using Array of Things (AoT) sensing data. Additionally, this study explored the influences on the spatial differentiation of sound using GeoDetector (version number: 1.0-4), thus addressing the limitation of previous studies that neglected to explore the influences on spatial heterogeneity. Our experimental results showed that sound levels in different areas of Chicago fluctuated irregularly over time. During the morning peak on weekdays: the four southern areas of Chicago have a high–high sound gathering mode, and the remaining areas are mostly randomly distributed; the sound level of a certain area has a significant negative correlation with population density, park area, and density of bike route; park area and population density are the main factors affecting the spatial heterogeneity of Chicago’s sound; and population density and park area play an essential role in factor interaction. This study has some theoretical significance and practical value. Residents can choose areas with lower noise for leisure activities according to the noise map of this study. While planning urban development, urban planners should pay attention to the single and interactive effects of factors in the city, such as parks, road network structures, and points of interest, on the urban sound environment. Researchers can build on this study to conduct studies on larger time scales. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Study area.</p> Full article ">Figure 2
<p>Spatial distribution for AoT nodes in Chicago.</p> Full article ">Figure 3
<p>The workflow of environmental sounds analysis.</p> Full article ">Figure 4
<p>The establishment process of Thiessen polygon.</p> Full article ">Figure 5
<p>Noise level of (<b>A</b>) node 001e0610bc10; (<b>B</b>) node 001e06117b44; (<b>C</b>) node 001e0611441e; (<b>D</b>) node 001e06118509.</p> Full article ">Figure 6
<p>Partition based on Thiessen polygon.</p> Full article ">Figure 7
<p>Noise level maps at (<b>A</b>) morning peak on weekdays; (<b>B</b>) morning peak on weekends; (<b>C</b>) evening peak on weekdays; and (<b>D</b>) evening peak on weekends (unit: dB).</p> Full article ">Figure 8
<p>Partition based on weighted Thiessen polygon.</p> Full article ">Figure 9
<p>Scatter plot of global Moran’s <span class="html-italic">I</span>: (<b>A</b>) morning peak on weekdays; (<b>B</b>) evening peak on weekdays; (<b>C</b>) morning peak on weekends; (<b>D</b>) evening peak on weekends.</p> Full article ">Figure 10
<p>Clustering diagram of local Moran’s <span class="html-italic">I</span>: (<b>A</b>) morning peak on weekdays; (<b>B</b>) evening peak on weekdays.</p> Full article ">Figure A1
<p>Sound level of node 001e0610bc10.</p> Full article ">Figure A2
<p>Sound level of node 001e06113ace.</p> Full article ">Figure A3
<p>Sound level of node 001e0610f703.</p> Full article ">Figure A4
<p>Sound level of node 001e06117b44.</p> Full article ">Figure A5
<p>Sound level of node 001e061183f5.</p> Full article ">Figure A6
<p>Sound level of node 001e061182a7.</p> Full article ">Figure A7
<p>Sound level of node 001e06113acb.</p> Full article ">Figure A8
<p>Sound level of node 001e0610f732.</p> Full article ">Figure A9
<p>Sound level of node 001e0610f05c.</p> Full article ">Figure A10
<p>Sound level of node 001e0610ee43.</p> Full article ">Figure A11
<p>Sound level of node 001e06118295.</p> Full article ">Figure A12
<p>Sound level of node 001e0610e538.</p> Full article ">Figure A13
<p>Sound level of node 001e061146cb.</p> Full article ">Figure A14
<p>Sound level of node 001e0610ba46.</p> Full article ">Figure A15
<p>Sound level of node 001e0610ba15.</p> Full article ">Figure A16
<p>Sound level of node 001e0610bbe5.</p> Full article ">Figure A17
<p>Sound level of node 001e0610ee36.</p> Full article ">Figure A18
<p>Sound level of node 001e06113ad8.</p> Full article ">Figure A19
<p>Sound level of node 001e0611441e.</p> Full article ">Figure A20
<p>Sound level of node 001e06112e77.</p> Full article ">Figure A21
<p>Sound level of node 001e0610f6db.</p> Full article ">Figure A22
<p>Sound level of node 001e06113107.</p> Full article ">Figure A23
<p>Sound level of node 001e06113a24.</p> Full article ">Figure A24
<p>Sound level of node 001e061130f4.</p> Full article ">Figure A25
<p>Sound level of node 001e06113d20.</p> Full article ">Figure A26
<p>Sound level of node 001e061146ba.</p> Full article ">Figure A27
<p>Sound level of node 001e061144be.</p> Full article ">Figure A28
<p>Sound level of node 001e0610ba13.</p> Full article ">Figure A29
<p>Sound level of node 001e0611536c.</p> Full article ">Figure A30
<p>Sound level of node 001e06118509.</p> Full article ">Figure A31
<p>Sound level of node 001e0611850f.</p> Full article ">Figure A32
<p>Sound level of node 001e061184a3.</p> Full article ">Figure A33
<p>Sound level of node 001e0610ee5d.</p> Full article ">Figure A34
<p>Sound level of node 001e061144cd.</p> Full article ">Figure A35
<p>Sound level of node 001e0611462f.</p> Full article ">
22 pages, 3489 KiB  
Article
Carbon Emission Patterns and Carbon Balance Zoning in Urban Territorial Spaces Based on Multisource Data: A Case Study of Suzhou City, China
by Zhenlong Zhang, Xiaoping Yu, Yanzhen Hou, Tianhao Chen, Yun Lu and Honghu Sun
ISPRS Int. J. Geo-Inf. 2023, 12(9), 385; https://doi.org/10.3390/ijgi12090385 - 20 Sep 2023
Cited by 5 | Viewed by 1779
Abstract
The concept of green and low-carbon development is integrated into territorial spatial planning and district control research. It is one of the systematic policy tools for emission reduction and carbon sequestration, greatly contributing to achieving the double carbon goal. This paper presents a [...] Read more.
The concept of green and low-carbon development is integrated into territorial spatial planning and district control research. It is one of the systematic policy tools for emission reduction and carbon sequestration, greatly contributing to achieving the double carbon goal. This paper presents a method for measuring the carbon emissions of urban territorial spaces using multisource big data, aiming to identify the spatial patterns and levels of carbon emissions at microspatial scales. The spatial patterns of carbon emissions were used to construct a carbon balance zoning method to evaluate the regional differences in the spatial distribution of carbon emissions, taking Suzhou as an example to achieve carbon balance zoning at the micro scale of the city. Based on our research, the following was determined: (1) Suzhou’s total carbon emissions in 2020 was approximately 240.3 million tons, with the industrial sector accounting for 81.32% of these emissions. The total carbon sink was about 0.025 million tons. (2) In Suzhou City, the high-value plots of carbon emissions are mainly located in industrial agglomeration areas. By contrast, low-value plots are primarily located in suburban areas and various carbon sink functional areas, exhibiting a scattered distribution. (3) The territorial space unit was divided into four functional areas of carbon balance, with 36 low-carbon economic zone units accounting for 37.11%, 29 carbon-source control zone units accounting for 29.90%, 14 carbon-sink functional zone units accounting for 14.43%, and 18 high-carbon optimization zone units accounting for 18.56%. As a result of this study, carbon balance zoning was achieved at the grassroots space level, which will assist the city in low-carbon and refined urban governance. Some ideas and references are also provided to formulate policies for low-carbon development at the micro scale of a city. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Location of Suzhou City: (<b>a</b>) location of Suzhou in the Yangtze River Delta; (<b>b</b>) administrative boundary of Suzhou.</p> Full article ">Figure 2
<p>An overview of the methodological flowchart.</p> Full article ">Figure 3
<p>The relationship between territorial space land and carbon emissions.</p> Full article ">Figure 4
<p>The proportion of carbon emissions by departments and land use.</p> Full article ">Figure 5
<p>Carbon emissions of Suzhou City: (<b>a</b>) carbon emissions from land for the territorial space; (<b>b</b>) carbon emissions from a grid for the territorial space.</p> Full article ">Figure 6
<p>Carbon emissions and sink of township units in Suzhou City: (<b>a</b>) carbon emissions; (<b>b</b>) carbon sinks.</p> Full article ">Figure 7
<p>Township territorial space unit carbon compensation rate.</p> Full article ">Figure 8
<p>Carbon balance indexes of township units: (<b>a</b>) economic contribution coefficient; (<b>b</b>) eco-logical support coefficient.</p> Full article ">Figure 9
<p>(<b>a</b>) Township territorial space unit carbon balance zoning; (<b>b</b>) number of carbon balance zoning types; (<b>c</b>) the proportion of carbon balance zoning types.</p> Full article ">
19 pages, 8508 KiB  
Article
Influence of the Built Environment on Pedestrians’ Route Choice in Leisure Walking
by Yifu Ge, Zhongyu He and Kai Shang
ISPRS Int. J. Geo-Inf. 2023, 12(9), 384; https://doi.org/10.3390/ijgi12090384 - 19 Sep 2023
Viewed by 1591
Abstract
Exploring the relationship between leisure walking and the built environment will provide an improvement in human health and well-being. It is, therefore, necessary to explore the most relevant scale for leisure walking and how the association between the built environment and leisure walking [...] Read more.
Exploring the relationship between leisure walking and the built environment will provide an improvement in human health and well-being. It is, therefore, necessary to explore the most relevant scale for leisure walking and how the association between the built environment and leisure walking varies across scales. Three hundred volunteers were recruited to wear GPS loggers, and a total dataset of 268 tracks from 105 individuals was collected. The shortest possible routes between starting and ending points were generated and compared to the actual routes using the paired T-test. An improved grid-based buffer approach was proposed, and statistics for the grid cells intersecting the paths were calculated. Grid cells were calculated for six scales: 50 m, 100 m, 200 m, 500 m, 800 m, and 1600 m. The results showed that the actual paths were on average 24.97% longer than the shortest path. The mean, standard deviation, and minimum and maximum values of the built environment variables were all significantly associated with leisure walking. The most relevant spatial scale was found to be the 100 m scale. Overall, the smaller the scale, the more significant the association. Participants showed a preference for moderately compact urban forms, diverse options for destinations, and greener landscapes in leisure walking route choice. Full article
(This article belongs to the Topic Applications of Spatial Science and Technology in Health Research)
►▼ Show Figures

Figure 1

Figure 1
<p>Methodology of this study.</p> Full article ">Figure 2
<p>Map-matched GPS traces in Nanjing, China.</p> Full article ">Figure 3
<p>(<b>a</b>) A 200 m fishnet buffer based on 400 m grid cells intersecting the selected route (400 m scale) on the left and (<b>b</b>) 800 m grid cells intersecting the selected route on the right. (The gray area represents the generated buffer area, and the light blue squares represent the grid cells intersecting the path).</p> Full article ">Figure 4
<p>(<b>a</b>) Less overlap between actual path and paired shortest path and (<b>b</b>) more overlap between actual path and paired shortest path.</p> Full article ">Figure 5
<p>(<b>a</b>) Distribution of building coverage and (<b>b</b>) distribution of FAR (results of Kernel density analysis).</p> Full article ">Figure 6
<p>Distribution of POIs of various destinations.</p> Full article ">Figure 7
<p>(<b>a</b>) Distribution of land-use mix at 100 m scale and (<b>b</b>) distribution of land-use mix at 500 m scale (results of Kernel density analysis).</p> Full article ">
38 pages, 7226 KiB  
Article
An Ontology-Based Knowledge Representation Method for Typhoon Events from Chinese News Reports
by Danjie Chen, Yan Zheng, Liqun Ma and Fen Qin
ISPRS Int. J. Geo-Inf. 2023, 12(9), 383; https://doi.org/10.3390/ijgi12090383 - 19 Sep 2023
Viewed by 1743
Abstract
Typhoons are destructive weather events. News media reports contain large amounts of typhoon information. Transforming this information into useful knowledge to provide a basis for mining typhoon knowledge and supporting disaster prevention and relief is urgently required to solve this problem. Knowledge representation [...] Read more.
Typhoons are destructive weather events. News media reports contain large amounts of typhoon information. Transforming this information into useful knowledge to provide a basis for mining typhoon knowledge and supporting disaster prevention and relief is urgently required to solve this problem. Knowledge representation can be used to address this problem, although it presents several challenges. These challenges lie in expressing the static and dynamic characteristics of typhoons and formalizing the knowledge representation method and making it suitable for machine processing. Moreover, the general Chinese time and space representation method is overly cumbersome for use in ontologies. The present study proposes an ontology-based typhoon event representation method that solves the representation problems of the typhoon static concept and dynamic features. Furthermore, it summarizes the fixed patterns of time and space in Chinese news and designs a time and space model suitable for typhoon event ontologies. From the ontology population, typhoon event ontology instances are created, and the typhoon event ontology model is applied to the analysis of typhoon processes, verifying the effectiveness of the typhoon event ontology model. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Flow chart of the ontology model of typhoon events.</p> Full article ">Figure 2
<p>Category tree of typhoon events.</p> Full article ">Figure 3
<p>Relationships of typhoon events.</p> Full article ">Figure 4
<p>Diagram of uninterrupted time. t<sub>1</sub>, t<sub>2,</sub> and t<sub>3</sub> are three consecutive periods or intervals.</p> Full article ">Figure 5
<p>Diagram of discrete times. t<sub>1</sub>, t<sub>2,</sub> and t<sub>3</sub> are three disconnected periods or intervals.</p> Full article ">Figure 6
<p>Example of mode M6.</p> Full article ">Figure 7
<p>Diagram depicting the static object ontology model of typhoon events.</p> Full article ">Figure 8
<p>Diagram of the typhoon event ontology model.</p> Full article ">Figure 9
<p>Diagram depicting the process behind the typhoon ontology population method.</p> Full article ">Figure 10
<p>Diagram of the structure of lattice BiLSTM-CRF model.</p> Full article ">Figure 11
<p>Diagram of category and number of named entities identified from typhoon news.</p> Full article ">Figure 12
<p>Diagram of typhoon event ontology model in Protégé.</p> Full article ">Figure 13
<p>The typhoon event ontology and its properties and relationships in Protégé.</p> Full article ">Figure 14
<p>Definition of typhoon development event in OWL.</p> Full article ">Figure 15
<p>Case-study diagram of typhoon event ontology model.</p> Full article ">Figure 16
<p>Diagram showing quantity of different categories of events.</p> Full article ">Figure 17
<p>Diagram showing temporal distribution of different categories of events.</p> Full article ">Figure 18
<p>Diagram of spatial distribution of typhoon events.</p> Full article ">Figure 19
<p>Diagram displaying the proportion of different types of events in each province (city).</p> Full article ">
25 pages, 16002 KiB  
Article
Cross-Attention-Guided Feature Alignment Network for Road Crack Detection
by Chuan Xu, Qi Zhang, Liye Mei, Xiufeng Chang, Zhaoyi Ye, Junjian Wang, Lang Ye and Wei Yang
ISPRS Int. J. Geo-Inf. 2023, 12(9), 382; https://doi.org/10.3390/ijgi12090382 - 19 Sep 2023
Cited by 4 | Viewed by 1486
Abstract
Road crack detection is one of the important issues in the field of traffic safety and urban planning. Currently, road damage varies in type and scale, and often has different sizes and depths, making the detection task more challenging. To address this problem, [...] Read more.
Road crack detection is one of the important issues in the field of traffic safety and urban planning. Currently, road damage varies in type and scale, and often has different sizes and depths, making the detection task more challenging. To address this problem, we propose a Cross-Attention-guided Feature Alignment Network (CAFANet) for extracting and integrating multi-scale features of road damage. Firstly, we use a dual-branch visual encoder model with the same structure but different patch sizes (one large patch and one small patch) to extract multi-level damage features. We utilize a Cross-Layer Interaction (CLI) module to establish interaction between the corresponding layers of the two branches, combining their unique feature extraction capability and contextual understanding. Secondly, we employ a Feature Alignment Block (FAB) to align the features from different levels or branches in terms of semantics and spatial aspects, which significantly improves the CAFANet’s perception of the damage regions, reduces background interference, and achieves more precise detection and segmentation of damage. Finally, we adopt multi-layer convolutional segmentation heads to obtain high-resolution feature maps. To validate the effectiveness of our approach, we conduct experiments on the public CRACK500 dataset and compare it with other mainstream methods. Experimental results demonstrate that CAFANet achieves excellent performance in road crack detection tasks, which exhibits significant improvements in terms of F1 score and accuracy, with an F1 score of 73.22% and an accuracy of 96.78%. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Overview of the proposed CAFANet.</p> Full article ">Figure 2
<p>Architecture of the multi-scale path embedding and multi-head convolutional self-attention.</p> Full article ">Figure 3
<p>Architecture of the cross-layer interaction.</p> Full article ">Figure 4
<p>Architecture of the feature alignment block.</p> Full article ">Figure 5
<p>CRACK500 data sample.</p> Full article ">Figure 6
<p>A few examples of the proposed method’s segmentation results.</p> Full article ">Figure 7
<p>Visualization of attention maps in CRACK500 dataset.</p> Full article ">Figure 8
<p>Segmentation result examples obtained by various methods for transverse cracks.</p> Full article ">Figure 9
<p>Segmentation result examples obtained by various methods for transverse cracks.</p> Full article ">Figure 10
<p>Segmentation result examples obtained by various methods for transverse cracks.</p> Full article ">Figure 11
<p>Segmentation result examples obtained by various methods for longitudinal cracks.</p> Full article ">Figure 12
<p>Segmentation result examples obtained by various methods for longitudinal cracks.</p> Full article ">Figure 13
<p>Segmentation result examples obtained by various methods for longitudinal cracks.</p> Full article ">Figure 14
<p>Segmentation result examples obtained by various methods for mesh cracks.</p> Full article ">Figure 15
<p>Segmentation result examples obtained by various methods for mesh cracks.</p> Full article ">Figure 16
<p>Segmentation result examples obtained by various methods for mesh cracks.</p> Full article ">Figure 17
<p>The quantitative comparison of <span class="html-italic">Precision</span>, <span class="html-italic">IOU_1</span>, <span class="html-italic">F1 score</span>, and <span class="html-italic">mIOU</span> metrics.</p> Full article ">Figure 18
<p>CPRID data sample.</p> Full article ">Figure 19
<p>CPRID segmentation result examples.</p> Full article ">
19 pages, 8551 KiB  
Article
A Comprehensive Evaluation of Machine Learning and Classical Approaches for Spaceborne Active-Passive Fusion Bathymetry of Coral Reefs
by Jian Cheng, Liang Cheng, Sensen Chu, Jizhe Li, Qixin Hu, Li Ye, Zhiyong Wang and Hui Chen
ISPRS Int. J. Geo-Inf. 2023, 12(9), 381; https://doi.org/10.3390/ijgi12090381 - 19 Sep 2023
Cited by 2 | Viewed by 1738
Abstract
Satellite-derived bathymetry (SDB) techniques are increasingly valuable for deriving high-quality bathymetric maps of coral reefs. Investigating the performance of the related SDB algorithms in purely spaceborne active–passive fusion bathymetry contributes to formulating reliable bathymetric strategies, particularly for areas such as the Spratly Islands, [...] Read more.
Satellite-derived bathymetry (SDB) techniques are increasingly valuable for deriving high-quality bathymetric maps of coral reefs. Investigating the performance of the related SDB algorithms in purely spaceborne active–passive fusion bathymetry contributes to formulating reliable bathymetric strategies, particularly for areas such as the Spratly Islands, where in situ observations are exceptionally scarce. In this study, we took Anda Reef as a case study and evaluated the performance of eight common SDB approaches by integrating Sentinel-2 images with Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2). The bathymetric maps were generated using two classical and six machine-learning algorithms, which were then validated with measured sonar data. The results illustrated that all models accurately estimated the depth of coral reefs in the 0–20 m range. The classical algorithms (Lyzenga and Stumpf) exhibited a mean absolute error (MAE), root mean square error (RMSE), and mean absolute percentage error (MAPE) of less than 0.990 m, 1.386 m, and 11.173%, respectively. The machine learning algorithms generally outperformed the classical algorithms in accuracy and bathymetric detail, with a coefficient of determination (R2) ranging from 0.94 to 0.96 and an RMSE ranging from 1.034 m to 1.202 m. The multilayer perceptron (MLP) achieved the highest accuracy and consistency with an RMSE of as low as 1.034 m, followed by the k-nearest neighbor (KNN) (1.070 m). Our results provide a practical reference for selecting SDB algorithms to accurately obtain shallow water bathymetry in subsequent studies. Full article
(This article belongs to the Topic Advances in Earth Observation and Geosciences)
►▼ Show Figures

Figure 1

Figure 1
<p>The map of the study area. (<b>a</b>) Location of Anda Reef in the SCS region. (<b>b</b>) Location of Anda Reef in Zhenghe Reefs. (<b>c</b>) Different colored lines of ICESat-2 trajectories represent the data obtained on different dates.</p> Full article ">Figure 2
<p>The 12 high-quality Sentinel-2 images of Anda Reef, four images per year from 2019 to 2021. Most images are close to the best conditions for visual judgment (e.g., almost no clouds, low turbidity, and low white caps).</p> Full article ">Figure 3
<p>The spatial distribution of in situ observations of Anda Reef.</p> Full article ">Figure 4
<p>The multialgorithm evaluation framework of classical and machine learning methods for SDB by integrating Sentinel-2 images and ICESat-2 ATL03.</p> Full article ">Figure 5
<p>The ATL03 GT3L data of Anda Reef on 30 March 2022. (<b>a</b>) Spatial distribution of the track. The red line indicates the whole trajectory and the yellow line depicts the seafloor signal photon. (<b>b</b>) The classification results of the sea surface and seafloor photons.</p> Full article ">Figure 6
<p>(<b>a</b>) The optimal composite Sentinel-2 image. (<b>b</b>) The distribution characteristics of ICESat-2 bathymetric points.</p> Full article ">Figure 7
<p>Accuracy analysis of eight models based on the measured data: (<b>a</b>) MLP, (<b>b</b>) KNN, (<b>c</b>) SVM, (<b>d</b>) RF, (<b>e</b>) XGBoost, (<b>f</b>) LGBM, (<b>g</b>) Stumpf, and (<b>h</b>) Lyzenga.</p> Full article ">Figure 8
<p>Bathymetric maps of Anda Reef developed from (<b>a</b>) MLP, (<b>b</b>) KNN, (<b>c</b>) SVM, (<b>d</b>) RF, (<b>e</b>) XGBoost, (<b>f</b>) LGBM, (<b>g</b>) Stumpf, and (<b>h</b>) Lyzenga. The purple box line compares the local detail of various SDB bathymetry maps.</p> Full article ">Figure 9
<p>Depth profiles of bathymetric maps using eight SDB methods for Anda Reef. (<b>a</b>–<b>d</b>) represent profiles along transect P1, P2, P3, and P4, respectively, where P1 is the depth segments of the measured sonar data. The black arrows indicate the direction of the profile line. The red box lines highlight areas with significant differences in the various bathymetric maps.</p> Full article ">
26 pages, 18181 KiB  
Article
Assessing the Hazard Degree of Wadi Malham Basin in Saudi Arabia and Its Impact on North Train Railway Infrastructure
by Fatmah Nassir Alqreai and Hamad Ahmed Altuwaijri
ISPRS Int. J. Geo-Inf. 2023, 12(9), 380; https://doi.org/10.3390/ijgi12090380 - 17 Sep 2023
Cited by 1 | Viewed by 1778
Abstract
The North Train Railway in the Kingdom of Saudi Arabia (KSA) extends over vast areas, crossing various terrains, including valleys, sand veins, plateaus, and hills. Therefore, the railway was designed and implemented to suit this environmental diversity under the highest safety standards. However, [...] Read more.
The North Train Railway in the Kingdom of Saudi Arabia (KSA) extends over vast areas, crossing various terrains, including valleys, sand veins, plateaus, and hills. Therefore, the railway was designed and implemented to suit this environmental diversity under the highest safety standards. However, the railway may be subject to hazards for various reasons. In general, the possibility of direct surface runoff disasters increases if there are residential areas and facilities within the boundaries of drainage basins. Therefore, these areas should be studied, and the degree of hazard in drainage basins should be accurately determined. Hence, this study analyzed the degree of risk of 14 drainage basins affecting the North Train Railway within the Wadi Malham drainage basin. The risk degree model was used with eight parameters that have hydrological indications to give an idea of the behavior of direct surface runoff and alter the risk of direct surface runoff. We found that 28.57% of the total basins in the study area have overall score values indicating they are high-risk basins, namely basins 6, 7, 13, and 14. It is recommended to estimate the rainfall depth during different return periods, analyze soil permeability and land use classification in the study area, and apply hydrological modeling of drainage basins, which contributes to estimating the volume and peak of direct surface runoff in such arid and semi-arid environments that do not contain hydrometric stations to monitor the runoff. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Study Area.</p> Full article ">Figure 2
<p>Topographic maps, scale: 1:50,000. Source: Ministry of Petroleum and Mineral Resources, 1982.</p> Full article ">Figure 3
<p>Topographic maps, scale: 1:25,000. Source: General Commission for Survey, 2015.</p> Full article ">Figure 4
<p>Study Methodology.</p> Full article ">Figure 5
<p>DEM ALOS PALSAR RTC for the study area. Source: [<a href="#B40-ijgi-12-00380" class="html-bibr">40</a>].</p> Full article ">Figure 6
<p>Drainage basins and networks affecting the North Train Railway within the basin of Wadi Malham.</p> Full article ">Figure 7
<p>Basin area.</p> Full article ">Figure 8
<p>Drainage density in basins.</p> Full article ">Figure 9
<p>Stream frequency in basins.</p> Full article ">Figure 10
<p>Basin relief ratio.</p> Full article ">Figure 11
<p>Ruggedness number of basins.</p> Full article ">Figure 12
<p>Circularity ratio of basins.</p> Full article ">Figure 13
<p>Length of overland flow in basins.</p> Full article ">Figure 14
<p>Weighted mean bifurcation ratio in drainage basins.</p> Full article ">Figure 15
<p>Hazard degree of direct surface runoff in drainage basins in the study area.</p> Full article ">
25 pages, 4280 KiB  
Article
A Wandering Detection Method Based on Processing GPS Trajectories Using the Wavelet Packet Decomposition Transform for People with Cognitive Impairment
by Naghmeh Jafarpournaser, Mahmoud Reza Delavar and Maryam Noroozian
ISPRS Int. J. Geo-Inf. 2023, 12(9), 379; https://doi.org/10.3390/ijgi12090379 - 17 Sep 2023
Cited by 1 | Viewed by 1577
Abstract
The increasing prevalence of cognitive disorders among the elderly is a significant consequence of the global aging phenomenon. Wandering stands out as the most prominent and challenging symptom in these patients, with potential irreversible consequences such as loss or even death. Thus, harnessing [...] Read more.
The increasing prevalence of cognitive disorders among the elderly is a significant consequence of the global aging phenomenon. Wandering stands out as the most prominent and challenging symptom in these patients, with potential irreversible consequences such as loss or even death. Thus, harnessing technological advancements to mitigate caregiving burdens and disease-related repercussions becomes paramount. Numerous studies have developed algorithms and smart healthcare and telemedicine systems for wandering detection. Broadly, these algorithms fall into two categories: those estimating path complexity and those relying on historical trajectory data. However, motion signal processing methods are rarely employed in this context. This paper proposes a motion-signal-processing-based algorithm utilizing the wavelet packet transform (WPT) with a fourth-order Coiflet mother wavelet. The algorithm identifies wandering patterns solely based on patients’ positional data on the current traversed path and variations in wavelet coefficients within the frequency–time spectrum of motion signals. The model’s independence from prior motion behavior data enhances its compatibility with the pronounced instability often seen in these patients. A performance assessment of the proposed algorithm using the Geolife open-source dataset achieved accuracy, precision, specificity, recall, and F-score metrics of 83.06%, 92.62%, 83.06%, 83.06%, and 87.58%, respectively. Timely wandering detection not only prevents irreversible consequences but also serves as a potential indicator of progression to severe Alzheimer’s in patients with mild cognitive impairment, enabling timely interventions for preventing disease progression. This underscores the importance of advancing wandering detection algorithms. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Wandering patterns [<a href="#B14-ijgi-12-00379" class="html-bibr">14</a>]: (<b>a</b>) random; (<b>b</b>) lapping; (<b>c</b>) pacing.</p> Full article ">Figure 2
<p>Flowchart of the proposed methodology.</p> Full article ">Figure 3
<p>Signal classification using DWT (<a href="http://ataspinar.com/2018/12/21/a-guide-for-using-the-wavelet-transform-in-machine-learning" target="_blank">http://ataspinar.com/2018/12/21/a-guide-for-using-the-wavelet-transform-in-machine-learning</a> (accessed on 10 September 2023)).</p> Full article ">Figure 4
<p>Scale 4 decomposition procedure by DWT. H represents the low-pass filter and G the high-pass filter [<a href="#B69-ijgi-12-00379" class="html-bibr">69</a>].</p> Full article ">Figure 5
<p>Wavelet decomposition tree [<a href="#B70-ijgi-12-00379" class="html-bibr">70</a>].</p> Full article ">Figure 6
<p>Types of discrete wavelet function [<a href="#B73-ijgi-12-00379" class="html-bibr">73</a>].</p> Full article ">Figure 7
<p>Cleaned trajectory by removing outliers [<a href="#B51-ijgi-12-00379" class="html-bibr">51</a>].</p> Full article ">Figure 8
<p>The result of applying WPD to the real motion signal: (<b>a</b>) real motion signal with pacing pattern; (<b>b</b>) changes in wavelet coefficients in Xt; (<b>c</b>) changes in wavelet coefficients in Yt (red circle: onset of wandering).</p> Full article ">Figure 9
<p>The result of applying WPD to the real motion signal: (<b>a</b>) real motion signal with lapping pattern; (<b>b</b>) changes in wavelet coefficients in Xt; (<b>c</b>) changes in wavelet coefficients in Yt (red circle: onset of wandering).</p> Full article ">Figure 10
<p>The result of applying WPD to the real motion signal: (<b>a</b>) an example of a real motion signal; (<b>b</b>) changes in wavelet coefficients in Xt; (<b>c</b>) changes in wavelet coefficients in Yt (red circle: onset of wandering).</p> Full article ">
14 pages, 692 KiB  
Article
A Latent-Factor-Model-Based Approach for Traffic Data Imputation with Road Network Information
by Xing Su, Wenjie Sun, Chenting Song, Zhi Cai and Limin Guo
ISPRS Int. J. Geo-Inf. 2023, 12(9), 378; https://doi.org/10.3390/ijgi12090378 - 15 Sep 2023
Viewed by 1561
Abstract
With the rapid development of the economy, car ownership has grown rapidly, which causes many traffic problems. In recent years, intelligent transportation systems have been used to solve various traffic problems. To achieve effective and efficient traffic management, intelligent transportation systems need a [...] Read more.
With the rapid development of the economy, car ownership has grown rapidly, which causes many traffic problems. In recent years, intelligent transportation systems have been used to solve various traffic problems. To achieve effective and efficient traffic management, intelligent transportation systems need a large amount of complete traffic data. However, the current traffic data collection methods result in different forms of missing data. In the last twenty years, although many approaches have been proposed to impute missing data based on different mechanisms, these all have their limitations, which leads to low imputation accuracy, especially when the collected traffic data have a large amount of missing values. To this end, this paper proposes a latent-factor-model-based approach to impute the missing traffic data. In the proposed approach, the spatial information of the road network is first combined with the spatiotemporal matrix of the original traffic data. Then, the latent-factor-model-based algorithm is employed to impute the missing data in the combined matrix of the traffic data. Based on the real traffic data from METR-LA, we found that the imputation accuracy of the proposed approach was better than that of most of the current traffic-data-imputation approaches, especially when the original traffic data are limited. Full article
(This article belongs to the Topic Artificial Intelligence in Navigation)
►▼ Show Figures

Figure 1

Figure 1
<p>An example of a spatiotemporal matrix of original traffic data <math display="inline"><semantics> <mrow> <mi>V</mi> <mo>∈</mo> <msup> <mi mathvariant="double-struck">R</mi> <mrow> <mi>I</mi> <mo>·</mo> <mi>J</mi> </mrow> </msup> </mrow> </semantics></math>.</p> Full article ">Figure 2
<p>Three kinds of traffic data missing patterns.</p> Full article ">Figure 3
<p>The process of LFM based factorization of <span class="html-italic">X</span>.</p> Full article ">Figure 4
<p>An example of road network.</p> Full article ">Figure 5
<p>The original and normalized adjacent matrix of road network.</p> Full article ">Figure 6
<p>The combined matrix <math display="inline"><semantics> <mi mathvariant="italic">VD</mi> </semantics></math>.</p> Full article ">
16 pages, 1629 KiB  
Article
Spatio-Temporal Information Extraction and Geoparsing for Public Chinese Resumes
by Xiaolong Li, Wu Zhang, Yanjie Wang, Yongbin Tan and Jing Xia
ISPRS Int. J. Geo-Inf. 2023, 12(9), 377; https://doi.org/10.3390/ijgi12090377 - 13 Sep 2023
Cited by 4 | Viewed by 1585
Abstract
As an important carrier of individual information, the resume is an important data source for studying the spatio-temporal evolutionary characteristics of individual and group behaviors. This study focuses on spatio-temporal information extraction and geoparsing from resumes to provide basic technical support for spatio-temporal [...] Read more.
As an important carrier of individual information, the resume is an important data source for studying the spatio-temporal evolutionary characteristics of individual and group behaviors. This study focuses on spatio-temporal information extraction and geoparsing from resumes to provide basic technical support for spatio-temporal research based on resume text. Most current studies on resume text information extraction are oriented toward recruitment work, such as the automated information extraction, classification, and recommendation of resumes. These studies ignore the spatio-temporal information of individual and group behaviors implied in resumes. Therefore, this study takes the public resumes of teachers in key universities in China as the research data, proposes a set of spatio-temporal information extraction solutions for electronic resumes of public figures, and designs a spatial entity geoparsing method, which can effectively extract and spatially locate spatio-temporal information in the resumes. To verify the effectiveness of the proposed method, text information extraction models such as BiLSTM-CRF, BERT-CRF, and BERT-BiLSTM-CRF are selected to conduct comparative experiments, and the spatial entity geoparsing method is verified. The experimental results show that the precision of the selected models on the named entity recognition task is 96.23% and the precision of the designed spatial entity geoparsing method is 97.91%. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Example of crawled resumes of key university teachers. The resume has been shortened for display.</p> Full article ">Figure 2
<p>Framework of spatio-temporal information extraction and geocoding based on public resumes.</p> Full article ">Figure 3
<p>Example of Baidu Baike page content division.</p> Full article ">Figure 4
<p>Method of spatial entity geocoding based on Baidu Baike.</p> Full article ">
21 pages, 6474 KiB  
Article
Redesigning Graphical User Interface of Open-Source Geospatial Software in a Community-Driven Way: A Case Study of GRASS GIS
by Linda Karlovska, Anna Petrasova, Vaclav Petras and Martin Landa
ISPRS Int. J. Geo-Inf. 2023, 12(9), 376; https://doi.org/10.3390/ijgi12090376 - 10 Sep 2023
Viewed by 2260
Abstract
Learning to use geographic information system (GIS) software effectively may be intimidating due to the extensive range of features it offers. The GRASS GIS software, in particular, presents additional challenges for first-time users in terms of its complex startup procedure and unique terminology [...] Read more.
Learning to use geographic information system (GIS) software effectively may be intimidating due to the extensive range of features it offers. The GRASS GIS software, in particular, presents additional challenges for first-time users in terms of its complex startup procedure and unique terminology associated with its data structure. On the other hand, a substantial part of the GRASS user community including us as developers recognized and embraced the advantages of the current approach. Given the controversial nature of the whole issue, we decided to actively involve regular users by conducting several formal surveys and by performing usability testing. Throughout this process, we discovered that resolving specific software issues through pure user-centered design is not always feasible, particularly in the context of open-source scientific software where the boundary between users and developers is very fuzzy. To address this challenge, we adopted the user-centered methodology tailored to the requirements of open-source scientific software development, which we refer to as community-driven design. This paper describes the community-driven redesigning process on the GRASS GIS case study and sets a foundation for applying community-driven design in other open-source scientific projects by providing insights into effective software development practices driven by the needs and input of the project’s community. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Example of a GRASS hierarchical data structure: one directory with two locations (S-JTSK and WGS84 CRS) containing a total of three mapsets with nine vector layers and five raster layers.</p> Full article ">Figure 2
<p>Widgets providing functions for the data hierarchy management in GRASS GIS: (<b>a</b>) historically used independent start-up window, (<b>b</b>) improved Data Catalog which is part of the main software window.</p> Full article ">Figure 3
<p>Methodology of community-driven redesigning process and GUI evaluation.</p> Full article ">Figure 4
<p>Application of the methodology to GRASS GUI redesigning process (not including informal feedback solicitation and evaluation).</p> Full article ">Figure 5
<p>Proposal of first-time user mode: Diagram illustrating the logical flow, with hints displayed in orange color and integrated buttons.</p> Full article ">Figure 6
<p>The prototype scene presented in Q3: the first-time user hint devoted to the GRASS startup and data structure info.</p> Full article ">Figure 7
<p>The full implementation of the first-time user mode: the first hint devoted to the GRASS data hierarchy topic.</p> Full article ">Figure 8
<p>The infobar presented to a user after launching GRASS GIS into the temporary location. The hint explains the unusual start-up situation and suggests the next steps.</p> Full article ">Figure 9
<p>Single-window GUI mockup. The background from wxPython single-window demo project is partly overlayed by screenshots from GRASS GUI. All components were put together in GIMP software 2.10.20 [<a href="#B47-ijgi-12-00376" class="html-bibr">47</a>].</p> Full article ">Figure 10
<p>Performance-based measures of GRASS lab usability testing: Task error rate (<b>a</b>) and task completion time (<b>b</b>).</p> Full article ">Figure A1
<p>Gaze plot selection for the startup screen. Circle size reflects fixation duration. Upper figures exhibit quick problem understanding while lower figures indicate confusion.</p> Full article ">Figure A2
<p>Heat map of the situation after startup combining partial heat maps from all participants (multi-window GUI).</p> Full article ">Figure A3
<p>Heat map of the situation after startup combining partial heat maps from all participants (single-window GUI).</p> Full article ">
30 pages, 32076 KiB  
Article
An Ontology-Based Framework for Geospatial Integration and Querying of Raster Data Cube Using Virtual Knowledge Graphs
by Younes Hamdani, Guohui Xiao, Linfang Ding and Diego Calvanese
ISPRS Int. J. Geo-Inf. 2023, 12(9), 375; https://doi.org/10.3390/ijgi12090375 - 8 Sep 2023
Cited by 5 | Viewed by 2758
Abstract
The integration of the raster data cube alongside another form of geospatial data (e.g., vector data) raises considerable challenges when it comes to managing and representing it using knowledge graphs. Such integration can play an invaluable role in handling the heterogeneity of geospatial [...] Read more.
The integration of the raster data cube alongside another form of geospatial data (e.g., vector data) raises considerable challenges when it comes to managing and representing it using knowledge graphs. Such integration can play an invaluable role in handling the heterogeneity of geospatial data and linking the raster data cube to semantic technology standards. Many recent approaches have been attempted to address this issue, but they often lack robust formal elaboration or solely concentrate on integrating raster data cubes without considering the inclusion of semantic spatial entities along with their spatial relationships. This may constitute a major shortcoming when it comes to performing advanced geospatial queries and semantically enriching geospatial models. In this paper, we propose a framework that can enable such semantic integration and advanced querying of raster data cubes based on the virtual knowledge graph (VKG) paradigm. This framework defines a semantic representation model for raster data cubes that extends the GeoSPARQL ontology. With such a model, we can combine the semantics of raster data cubes with features-based models that involve geometries as well as spatial and topological relationships. This could allow us to formulate spatiotemporal queries using SPARQL in a natural way by using ontological concepts at an appropriate level of abstraction. We propose an implementation of the proposed framework based on a VKG system architecture. In addition, we perform an experimental evaluation to compare our framework with other existing systems in terms of performance and scalability. Finally, we show the potential and the limitations of our implementation and we discuss several possible future works. Full article
(This article belongs to the Topic Geospatial Knowledge Graph)
►▼ Show Figures

Figure 1

Figure 1
<p>An abstract overview of the classes and properties defined in the GeoSPARQL standard (<b>left sub-figure</b>) and the types of spatial relationships where only the topological ones are implemented by the GeoSPARQL standard (<b>right sub-figure</b>).</p> Full article ">Figure 2
<p>GeoSPARQL geometry taxonomy in compliance with OGC standards [<a href="#B34-ijgi-12-00375" class="html-bibr">34</a>].</p> Full article ">Figure 3
<p>Different abstraction of the data cube spatial atom.</p> Full article ">Figure 4
<p>Diagram of the developed raster data cube vocabulary. Solid lines indicate object or data properties, whereas arrows indicate the direction of property relations. Dotted lines indicate subclass relations. Dashed lines without arrowheads indicate the connection between disjoint classes. The green rectangles indicate object properties, while the pink rectangles indicate data properties. In the case of object properties, their functional or inverse-functional nature is specified where applicable. Classes represented in dark blue correspond to the classes of the ontological vocabulary we have developed, while those in lighter blue represent classes reused from another ontology (in this case, the GeoSPARQL ontology).</p> Full article ">Figure 5
<p>Taxonomy used for the classification of geospatial semantic queries.</p> Full article ">Figure 6
<p>System architecture is presented as three layers and six processes. The orange dashed arrows indicate the inputs to each system component. The green arrows indicate the processes involved in the query answer. The numbers indicate the order in which each process must be performed to obtain an answer to a query.</p> Full article ">Figure 7
<p>Database design for representing semantic data cube.</p> Full article ">Figure 8
<p>Study area and dataset.</p> Full article ">Figure 9
<p>Map visualization of query results of Q5 (<b>left</b>) and Q6 (<b>right</b>).</p> Full article ">Figure 10
<p>Map visualization of the query result for Q7.</p> Full article ">Figure 11
<p>Visualization of the query result for Q8. The numbers ranging from 1 to 5 delineate the chronological order of paths followed, beginning from the initial point and concluding at the final destination while tracking the maximum temperature. The directional arrows positioned along these pathways serve to illustrate the specific direction of movement.</p> Full article ">Figure 12
<p>List of query results of Q9.</p> Full article ">Figure 13
<p>Graphs highlighting the evolution of computation time of four queries of the Ontop system depending on the size of temporal window in comparison with Geold and Strabon.</p> Full article ">Figure 14
<p>Graphs highlighting the evolution of computation time of five advanced queries on the Ontop system depending on the size of temporal window.</p> Full article ">
18 pages, 2550 KiB  
Article
Geospatial Analysis in Web Browsers—Comparison Study on WebGIS Process-Based Applications
by Rostislav Netek, Tereza Pohankova, Oldrich Bittner and Daniel Urban
ISPRS Int. J. Geo-Inf. 2023, 12(9), 374; https://doi.org/10.3390/ijgi12090374 - 7 Sep 2023
Viewed by 2402
Abstract
With the rapid development of internet technologies in recent years, the shift from the desktop to the web platform can be seen within geospatial analysis. While analytical tools, such as buffer or clip, are routinely used in desktop environments, WebGIS deals with geographic [...] Read more.
With the rapid development of internet technologies in recent years, the shift from the desktop to the web platform can be seen within geospatial analysis. While analytical tools, such as buffer or clip, are routinely used in desktop environments, WebGIS deals with geographic information, including geospatial analysis, within the online environment. The main aim of this paper is to perform a comparison and evaluation of vector-oriented online geoprocessing tools in a WebGIS environment, supported by the development of a custom solution for geospatial analysis. The application called GeOnline is developed and tested as a case study to demonstrate the availability of spatial analysis tools within the web browser. It implements the specialized geospatial library Turf.js, which allows using non-trivial geospatial analysis, such as intersect, clip or calculate centroids. It handles client-side processes. Both a functionality comparison and performance testing are carried out, while the paper primarily focuses on data-driven (data-based) analysis and not only on visual-driven (visual-based) analysis. The comparative study evaluates five geospatial tools (ArcGIS Online, GISCloud, CARTO, FOURSQUARE, GeOnline) and summarizes the solutions from different aspects, including the number of supported operations. Finally, performance tests on GeOnline separately and among alternative solutions are performed. While ArcGIS Online is considered the most comprehensive solution on the market, GeOnline performs well compared to alternative solutions. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Categorization of web maps by Sack [<a href="#B21-ijgi-12-00374" class="html-bibr">21</a>].</p> Full article ">Figure 2
<p>GeOnline interface.</p> Full article ">Figure 3
<p>Source code ensuring the “Line to polygon” operation.</p> Full article ">Figure 4
<p>Cluster and Outlier Analysis in Foursquare Studio.</p> Full article ">Figure 5
<p>DevTools—an integral tool in web browsers used for performance testing evaluation.</p> Full article ">Figure 6
<p>Creating a buffer in GeOnline and measuring indicator values.</p> Full article ">
21 pages, 6217 KiB  
Article
Novel CNN-Based Approach for Reading Urban Form Data in 2D Images: An Application for Predicting Restaurant Location in Seoul, Korea
by Jeyun Yang and Youngsang Kwon
ISPRS Int. J. Geo-Inf. 2023, 12(9), 373; https://doi.org/10.3390/ijgi12090373 - 7 Sep 2023
Cited by 5 | Viewed by 1968
Abstract
Artificial intelligence (AI) has demonstrated its ability to complete complex tasks in various fields. In urban studies, AI technology has been utilized in some limited domains, such as control of traffic and air quality. This study uses AI to better understand diverse urban [...] Read more.
Artificial intelligence (AI) has demonstrated its ability to complete complex tasks in various fields. In urban studies, AI technology has been utilized in some limited domains, such as control of traffic and air quality. This study uses AI to better understand diverse urban studies data through a novel approach that uses a convolutional neural network (CNN). In this study, a building outline in the form of a two-dimensional image is used with its corresponding metadata to test the applicability of CNN in reading urban data. MobileNet, a high-efficiency CNN model, is trained to predict the location of restaurants in each building in Seoul, Korea. Consequently, using only 2D image data, the model satisfactorily predicts the locations of restaurants (AUC = 0.732); the model with 2D images and their metadata has higher performance but has an overfitting problem. In addition, the model using only 2D image data accurately predicts the regional distribution of restaurants and shows some typical urban forms with restaurants. The proposed model has several technical limitations but shows the potential to provide a further understanding of urban settings. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Representation of research data and methodology.</p> Full article ">Figure 2
<p>Illustration of the structure of MobileNet [<a href="#B81-ijgi-12-00373" class="html-bibr">81</a>].</p> Full article ">Figure 3
<p>Central places in Seoul.</p> Full article ">Figure 4
<p>Predicted and actual distribution of restaurants. (<b>a</b>) Predicted distribution of restaurants by Model 1 (CNN-only model). (<b>b</b>) Predicted distribution of restaurants by Model 2 (CNN-MLP model). (<b>c</b>) Predicted distribution of restaurants by Model 3 (MLP-only model). (<b>d</b>) Actual distribution of restaurants in Seoul.</p> Full article ">Figure 5
<p>Distribution of restaurants by type of commercial areas.</p> Full article ">Figure 6
<p>Examples of predicted restaurant locations by CNN-only model. Red lines indicate neighborhood roads while blue lines indicate arterial roads. (<b>a</b>) Yeokchon-dong (Eunpyeong-gu). (<b>b</b>) Nonhyeon-dong (Gangnam-gu). (<b>c</b>) Dohwa-dong (Mapo-gu).</p> Full article ">
21 pages, 4240 KiB  
Article
Detecting Turning Relationships and Time Restrictions of OSM Road Intersections from Crowdsourced Trajectories
by Xin Chen, Longgang Xiang, Fengwei Jiao and Huayi Wu
ISPRS Int. J. Geo-Inf. 2023, 12(9), 372; https://doi.org/10.3390/ijgi12090372 - 6 Sep 2023
Cited by 1 | Viewed by 1586
Abstract
OpenStreetMap (OSM) road networks provide public digital maps underlying many spatial applications such as routing engines and navigation services. However, turning relationships and time restrictions at OSM intersections are lacking in these maps, posing a threat to the accuracy and reliability of the [...] Read more.
OpenStreetMap (OSM) road networks provide public digital maps underlying many spatial applications such as routing engines and navigation services. However, turning relationships and time restrictions at OSM intersections are lacking in these maps, posing a threat to the accuracy and reliability of the services. In this paper, a new turn information detection method for OSM intersections using the dynamic connection information from crowdsourced trajectory data is proposed to address this problem. In this solution, the OSM intersection structure is extracted and simplified and crowdsourced trajectories are projected onto OSM road segments using an improved Hidden Markov Model (HMM) map matching method that explicitly traces the turning connections in road networks. Optimal path analysis increases the turning support related to short road segments. On this basis, this study transforms complex turning identification scenarios into the simple analyses of traffic connectivity. Furthermore, a voting strategy is used to identify and calculate turning time restrictions. The experimental results, using trajectory data from three cities in China, show that the turning relationships can be detected at a precision of 90.71% with a recall of 96.55% and an F1-value of 93.54% in Shanghai. For Wuhan, the precision is 95.33% and the recall is 95.00%, with an F1-value of 95.16%. The precision and recall when identifying turning time restrictions both reach 90% in Xiamen. These results demonstrate the effectiveness of the proposed turning detection method. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Workflow of the proposed method.</p> Full article ">Figure 2
<p>The process of intersection simplification: (<b>a</b>) the existing CPs; (<b>b</b>) all CPs detected after road interruption; (<b>c</b>) the CPs belong to one intersection; (<b>d</b>) the intersection is simplified as a connection point, and the road direction is derived from OSM road attributes [<a href="#B27-ijgi-12-00372" class="html-bibr">27</a>].</p> Full article ">Figure 3
<p>Identifying reverse driving with the improved map matching method: (<b>a</b>) the route from <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>P</mi> </mrow> <mrow> <mn>2</mn> </mrow> </msub> </mrow> </semantics></math> to <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>P</mi> </mrow> <mrow> <mn>3</mn> </mrow> </msub> </mrow> </semantics></math> calculated by the traditional matching algorithm concludes two U-turns (from <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>e</mi> </mrow> <mrow> <mn>2</mn> </mrow> </msub> </mrow> </semantics></math> to <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>e</mi> </mrow> <mrow> <mn>1</mn> </mrow> </msub> </mrow> </semantics></math> and then back to <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>e</mi> </mrow> <mrow> <mn>2</mn> </mrow> </msub> </mrow> </semantics></math>). Conversely, the reverse-driving matched point <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>P</mi> </mrow> <mrow> <mn>3</mn> </mrow> </msub> </mrow> </semantics></math> is projected on the position of matched point of <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>P</mi> </mrow> <mrow> <mn>2</mn> </mrow> </msub> </mrow> </semantics></math> with the improved method, which makes the route more reasonable; (<b>b</b>) the matched result of a reverse driving trajectory with our method, where the number corresponds to the point index in the trajectory.</p> Full article ">Figure 4
<p>The driving routes and the time gaps of the adjacent matched points: (<b>a</b>) two candidate routes from <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>C</mi> </mrow> <mrow> <mn>1</mn> </mrow> </msub> </mrow> </semantics></math> to <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>C</mi> </mrow> <mrow> <mn>2</mn> </mrow> </msub> </mrow> </semantics></math>, and (<b>b</b>) the real travel time of the subtrajectory on the candidate route equals <math display="inline"><semantics> <mrow> <mo>Δ</mo> <mi>t</mi> </mrow> </semantics></math>(<math display="inline"><semantics> <mrow> <msub> <mrow> <mi>C</mi> </mrow> <mrow> <mi>i</mi> </mrow> </msub> <mo>,</mo> <msub> <mrow> <mi>C</mi> </mrow> <mrow> <mi>i</mi> <mo>+</mo> <mn>3</mn> </mrow> </msub> </mrow> </semantics></math>) − <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>t</mi> </mrow> <mrow> <mi>l</mi> <mi>e</mi> <mi>a</mi> <mi>v</mi> <mi>e</mi> </mrow> </msub> </mrow> </semantics></math> − <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>t</mi> </mrow> <mrow> <mi>e</mi> <mi>n</mi> <mi>t</mi> <mi>e</mi> <mi>r</mi> </mrow> </msub> </mrow> </semantics></math>.</p> Full article ">Figure 5
<p>The voting flow of banned traffic time.</p> Full article ">Figure 6
<p>A detour corresponding to a direct left turn: a turning relationship could be completed by a detour consisting of multiple roads, and the red arrows are the paths of the two turns.</p> Full article ">Figure 7
<p>Test areas and trajectory datasets (the red points in the test area are the checked intersections): (<b>a</b>) Shanghai; (<b>b</b>) Wuhan; (<b>c</b>) Xiamen.</p> Full article ">Figure 8
<p>Distributions of <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>a</mi> <mi>n</mi> <mi>g</mi> <mi>l</mi> <mi>e</mi> </mrow> <mrow> <mn>1</mn> </mrow> </msub> </mrow> </semantics></math>, <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>a</mi> <mi>n</mi> <mi>g</mi> <mi>l</mi> <mi>e</mi> </mrow> <mrow> <mn>2</mn> </mrow> </msub> </mrow> </semantics></math>, and <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>a</mi> <mi>n</mi> <mi>g</mi> <mi>l</mi> <mi>e</mi> </mrow> <mrow> <mn>3</mn> </mrow> </msub> </mrow> </semantics></math>: <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>a</mi> <mi>n</mi> <mi>g</mi> <mi>l</mi> <mi>e</mi> </mrow> <mrow> <mn>1</mn> </mrow> </msub> </mrow> </semantics></math> and <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>a</mi> <mi>n</mi> <mi>g</mi> <mi>l</mi> <mi>e</mi> </mrow> <mrow> <mn>2</mn> </mrow> </msub> </mrow> </semantics></math> are concentrated in the area of nearly 0°, while <math display="inline"><semantics> <mrow> <msub> <mrow> <mi>a</mi> <mi>n</mi> <mi>g</mi> <mi>l</mi> <mi>e</mi> </mrow> <mrow> <mn>3</mn> </mrow> </msub> </mrow> </semantics></math> is mainly distributed in the area with large values.</p> Full article ">Figure 9
<p>Comparison of the experiments with different approaches in different areas: (<b>a</b>) Shanghai and (<b>b</b>) Wuhan.</p> Full article ">Figure 10
<p>Examples of the identified turning relationships at intersections: (<b>a</b>) a crossroad; (<b>b</b>) a complex intersection; (<b>c</b>) the yellow turning curves related to the short road segments in the red box cannot be detected by Efentakis’ method; (<b>d</b>) an intersection with parallel or same-directional road segments; (<b>e</b>) U-turns in the middle of road segment.</p> Full article ">Figure 11
<p>The identified time-restricted turning relationships at different intersections, where the yellow arrow is the time-restricted turning relationships: (<b>a</b>–<b>d</b>) crossroads; (<b>e</b>) intersection with wrong result; (<b>f</b>) a special intersection.</p> Full article ">
22 pages, 4047 KiB  
Article
Nonlinear Hierarchical Effects of Housing Prices and Built Environment Based on Multiscale Life Circle—A Case Study of Chengdu
by Yandi Song, Shaoyao Zhang and Wei Deng
ISPRS Int. J. Geo-Inf. 2023, 12(9), 371; https://doi.org/10.3390/ijgi12090371 - 6 Sep 2023
Cited by 2 | Viewed by 1898
Abstract
Determining the optimal planning scale for urban life circles and analyzing the associated built environment factors are crucial for comprehending and regulating residential differentiation. This study aims to bridge the current research void concerning the nonlinear hierarchical relationships between the built environment and [...] Read more.
Determining the optimal planning scale for urban life circles and analyzing the associated built environment factors are crucial for comprehending and regulating residential differentiation. This study aims to bridge the current research void concerning the nonlinear hierarchical relationships between the built environment and residential differentiation under the multiscale effect. Specifically, six indicators were derived from urban crowdsourcing data: diversity of built environment function (DBEF1), density of built environment function (DBEF2), blue–green environment (BGE), traffic accessibility (TA), population vitality (PV), and shopping vitality (SV). Then, a gradient boosting decision tree (GBDT) was applied to derive the analysis of these indicators. Finally, the interpretability of machine learning was leveraged to quantify the relative importance and nonlinear relationships between built environment indicators and housing prices. The results indicate a hierarchical structure and inflection point effect of the built environment on residential premiums. Notably, the impact trend of the built environment on housing prices within a 15 min life circle remains stable. The effect of crowd behavior, as depicted by PV and SV, on housing prices emerges as the most significant factor. Furthermore, this study also categorizes housing into common and high-end residences, thereby unveiling that distinct residential neighborhoods exhibit varying degrees of dependence on the built environment. The built environment exerts a scale effect on the formation of residential differentiation, with housing prices exhibiting increased sensitivity to the built environment at a smaller life circle scale. Conversely, the effect of the built environment on housing prices is amplified at a larger life circle scale. Under the dual influence of the scale and hierarchical effect, this framework can dynamically adapt to the uncertainty of changes in life circle planning policies and residential markets. This provides strong theoretical support for exploring the optimal life circle scale, alleviating residential differentiation, and promoting group fairness. Full article
(This article belongs to the Special Issue Application of Geographical Information System in Urban Design, Management or Evaluation)
►▼ Show Figures

Figure 1

Figure 1
<p>Research framework and technical route of this paper. (<b>a</b>) calculate the average price of residential units and display it in the residential center. (<b>b</b>) construct multiscale life circle and classify residential neighborhoods. (<b>c</b>) establishes the composition of the built environment indicators and the nonlinear relationship driven by GBDT. In which, figures 1–3 are conceptual diagrams depicting the nonlinear relationship between built environment indicators and housing prices for high-end residences under different scale of life circle, while figures 4–6 represent common residences. The same color lines represent the same scale of life circle.)</p> Full article ">Figure 2
<p>Overview of the study area and housing price in residential neighborhoods.</p> Full article ">Figure 3
<p>The heterogeneity of housing prices in residential neighborhoods.</p> Full article ">Figure 4
<p>Chengdu metropolitan housing market division results.</p> Full article ">Figure 5
<p>Nonlinear relationship between the built environment factors and housing price in different life circles. (<b>a</b>) DBEF1. (<b>b</b>) DBEF2. (<b>c</b>) BGE. (<b>d</b>) TA. (<b>e</b>) PV. (<b>f</b>) SV.</p> Full article ">Figure 6
<p>The nonlinear relationship between the built environment and the price of different grades of housing in the 10 min life circle. (<b>a</b>) DBEF1. (<b>b</b>) DBEF2. (<b>c</b>) BGE. (<b>d</b>) TA. (<b>e</b>) SV. (<b>f</b>) PV.</p> Full article ">Figure 7
<p>The nonlinear relationship between the built environment and the price of different grades of housing in the 15 min life circle. (<b>a</b>) DBEF1. (<b>b</b>) DBEF2. (<b>c</b>) BGE. (<b>d</b>) TA. (<b>e</b>) SV. (<b>f</b>) PV.</p> Full article ">Figure 8
<p>The nonlinear relationship between the built environment and the price of different grades of housing in the 20 min life circle. (<b>a</b>) DBEF1. (<b>b</b>) DBEF2. (<b>c</b>) BGE. (<b>d</b>) TA. (<b>e</b>) SV. (<b>f</b>) PV.</p> Full article ">
25 pages, 6669 KiB  
Article
A Multi-Framework of Google Earth Engine and GEV for Spatial Analysis of Extremes in Non-Stationary Condition in Southeast Queensland, Australia
by Hadis Pakdel, Dev Raj Paudyal, Sreeni Chadalavada, Md Jahangir Alam and Majid Vazifedoust
ISPRS Int. J. Geo-Inf. 2023, 12(9), 370; https://doi.org/10.3390/ijgi12090370 - 6 Sep 2023
Cited by 2 | Viewed by 1801
Abstract
The frequency and severity of extremes, including extreme precipitation events, extreme evapotranspiration and extreme water storage deficit events, are changing. Thus, the necessity for developing a framework that estimates non-stationary conditions is urgent. The aim of this paper is to develop a framework [...] Read more.
The frequency and severity of extremes, including extreme precipitation events, extreme evapotranspiration and extreme water storage deficit events, are changing. Thus, the necessity for developing a framework that estimates non-stationary conditions is urgent. The aim of this paper is to develop a framework using the geeSEBAL platform, Generalised Extreme Value (GEV) models and spatiotemporal analysis techniques that incorporate the physical system in terms of cause and effect. Firstly, the geeSEBAL platform has enabled the estimation of actual evapotranspiration (ETa) with an unprecedented level of spatial-temporal resolution. Following this, the Non-stationary Extreme Value Analysis (NEVA) approach employs the Bayesian method using a Differential Evolution Markov Chain technique to calculate the frequency and magnitude of extreme values across the parameter space. Station and global climate datasets have been used to analyse the spatial and temporal variation of rainfall, reference evapotranspiration (ETo), ETa and water storage (WS) variables in the Lockyer Valley located in Southeast Queensland (SEQ), Australia. Frequency analysis of rainfall, ETa, and water storage deficit for 14 stations were performed using a GEV distribution under stationary and non-stationary assumptions. Comparing the ETa, ETo and ERA5 rainfall with station data showed reasonable agreement as follows: Pearson correlation of 0.59–0.75 for ETa, RMSE of 45.23–58.56 mm for ETa, Pearson correlation of 0.96–0.97 for ETo, RMSE of 73.13–87.73 mm for ETo and Pearson correlation of 0.87–0.92 for rainfall and RMSE of 37.53–57.10 mm for rainfall. The lower and upper uncertainty bounds between stationary and non-stationary conditions for rainfall station data of Gatton varied from 550.98 mm (stationary) to 624.97 mm (non-stationary), and for ERA5 rainfall datasets, 441.30 mm (stationary) to 450.77 mm (non-stationary). The results demonstrate that global climate datasets underestimate the difference between stationary and non-stationary conditions by 9.47 mm compared to results of 73.99 mm derived from station data. Similarly, the results demonstrate less variation between stationary and non-stationary conditions in water storage, followed by a sharp variation in rainfall and moderate variation in evapotranspiration. The findings of this study indicate that neglecting the non-stationary condition in some hydrometeorological variables can lead to underestimating their amounts. This framework can be applied to any geographical area for estimating extreme conditions, providing valuable insights for infrastructure planning and design, risk assessment and disaster management. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Flowchart of geeSEBAL, GEE and GEV model to estimate extremes. (<b>a</b>) geeSEBAL algorithm, (<b>b</b>) accuracy assessment of hydrometeorological variables from station and global climate data, (<b>c</b>) spatial pattern of hydrometeorological variables from station and global climate data, (<b>d</b>) extreme events analysis (Modified [<a href="#B11-ijgi-12-00370" class="html-bibr">11</a>]).</p> Full article ">Figure 2
<p>Geographical location of studied area in Australia (<b>Left</b>). Hydro-meteorological stations considered along the Lockyer Catchment (<b>Right</b>).</p> Full article ">Figure 3
<p>Changes in spatial pattern of hydroclimate parameters (annual cumulative rainfall (column 1), reference evapotranspiration (column 2), actual evapotranspiration (column 3) and water storage deficit (column 4)) from 2000 to 2020 based on geeSEBAL algorithm resulting from station datasets over the Lockyer Catchment.</p> Full article ">Figure 3 Cont.
<p>Changes in spatial pattern of hydroclimate parameters (annual cumulative rainfall (column 1), reference evapotranspiration (column 2), actual evapotranspiration (column 3) and water storage deficit (column 4)) from 2000 to 2020 based on geeSEBAL algorithm resulting from station datasets over the Lockyer Catchment.</p> Full article ">Figure 4
<p>Changes in spatial pattern of hydroclimate parameters (annual cumulative rainfall (column 1), reference evapotranspiration (column 2), actual evapotranspiration (column 3) and water storage deficit (column 4)) from 2000 to 2020 based on geeSEBAL algorithm resulted from ERA5 reanalysis datasets over the Lockyer Catchment.</p> Full article ">Figure 4 Cont.
<p>Changes in spatial pattern of hydroclimate parameters (annual cumulative rainfall (column 1), reference evapotranspiration (column 2), actual evapotranspiration (column 3) and water storage deficit (column 4)) from 2000 to 2020 based on geeSEBAL algorithm resulted from ERA5 reanalysis datasets over the Lockyer Catchment.</p> Full article ">Figure 5
<p>The relationship between monthly cumulative rainfall, reference evapotranspiration and actual evapotranspiration derived from geeSEBAL for the period 1990–2020 with the station datasets (<b>a</b>–<b>u</b>) in the Lockyer Catchment.</p> Full article ">Figure 5 Cont.
<p>The relationship between monthly cumulative rainfall, reference evapotranspiration and actual evapotranspiration derived from geeSEBAL for the period 1990–2020 with the station datasets (<b>a</b>–<b>u</b>) in the Lockyer Catchment.</p> Full article ">Figure 6
<p>The output of NEVA’s non-stationary GEV framework, standard return levels with design exceedance probability for <span class="html-italic">ET</span><sub>a</sub> based on station data (<b>a</b>–<b>e</b>) (Figure generated using MATLAB).</p> Full article ">Figure 6 Cont.
<p>The output of NEVA’s non-stationary GEV framework, standard return levels with design exceedance probability for <span class="html-italic">ET</span><sub>a</sub> based on station data (<b>a</b>–<b>e</b>) (Figure generated using MATLAB).</p> Full article ">Figure 7
<p>The output of NEVA’s non-stationary GEV framework, standard return levels with design exceedance probability for <span class="html-italic">ET</span><sub>a</sub> based on global climate data (<b>a</b>–<b>d</b>) (Figure generated using MATLAB).</p> Full article ">Figure 7 Cont.
<p>The output of NEVA’s non-stationary GEV framework, standard return levels with design exceedance probability for <span class="html-italic">ET</span><sub>a</sub> based on global climate data (<b>a</b>–<b>d</b>) (Figure generated using MATLAB).</p> Full article ">
21 pages, 6051 KiB  
Article
Large-Scale Mobile-Based Analysis for National Travel Demand Modeling
by Bat-hen Nahmias-Biran, Shuki Cohen, Vladimir Simon and Israel Feldman
ISPRS Int. J. Geo-Inf. 2023, 12(9), 369; https://doi.org/10.3390/ijgi12090369 - 5 Sep 2023
Viewed by 1393
Abstract
Mobile phones have achieved a high rate of penetration and gained great interest in the field of travel behavior studies. However, mobile phone data exploitation for national travel models has only been sporadically studied thus far. This work focuses on one of the [...] Read more.
Mobile phones have achieved a high rate of penetration and gained great interest in the field of travel behavior studies. However, mobile phone data exploitation for national travel models has only been sporadically studied thus far. This work focuses on one of the most extensive cellular surveys of its kind carried out thus far in the world, which was performed for two years between 2018 and 2019 with the participation of the two largest cellular providers in Israel, as well as leading GPS companies. The large-scale cell phone survey covered half the population using cellphones aged 8+ in Israel and uncovered local and national trip patterns, revealing the structure of nationwide travel demand. The methodology consists of the following steps: (1) plausibility and quality checks for the data of the mobile operators and the GPS data providers; (2) algorithm development for trip detection, home/work location detection, location and time accuracy, and expansion factors; (3) accuracy test of origin–destination matrices at different resolutions, revisions of algorithms, and reproduction of data; and (4) validation of results by comparison to reliable external data sources. The results are characterized by high accuracy and representativeness of demand and indicate a strong correlation between the cellular survey and other reliable sources. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Trip detection mechanism.</p> Full article ">Figure 2
<p>Population per locale: comparison of Central Bureau of Statistics data and cellular data.</p> Full article ">Figure 3
<p>O-D blank point cell illustration as a result of privacy limitations.</p> Full article ">Figure 4
<p>Trip distribution by (<b>a</b>) day type and time of day and (<b>b</b>) travel length and time of day.</p> Full article ">Figure 5
<p>Ridership by half-hour distribution in (<b>a</b>) Tel Aviv and (<b>b</b>) Jerusalem.</p> Full article ">Figure 6
<p>Vehicle kilometers traveled (VKT) by range of travel distance.</p> Full article ">Figure 7
<p>Average daily ridership on workdays by origin and destination (in thousands of km) for (<b>a</b>) tourists and (<b>b</b>) the Israeli population.</p> Full article ">Figure 8
<p>Ridership for central zones on weekdays at morning peak (6 am–9 am): (<b>a</b>) incoming; and (<b>b</b>) outgoing.</p> Full article ">Figure 9
<p>Light, medium, and heavy commuters in Israel.</p> Full article ">Figure 10
<p>Trips to cities: correlation between the cellular data and travel habit survey for (<b>a</b>) all day and (<b>b</b>) morning peak hours.</p> Full article ">Figure 11
<p>Trips to major metropolitan areas by time of day: a comparison between cellular data and travel habit survey at the (<b>a</b>) state level; (<b>b</b>) Tel Aviv; (<b>c</b>) Jerusalem; (<b>d</b>) Beer-Sheva; and (<b>e</b>) Haifa metropolitan level.</p> Full article ">Figure 12
<p>Trips crossing the Tel Aviv metropolitan area’s cordons: A comparison between the CD, cordon survey and THS in (<b>a</b>) absolute numbers and (<b>b</b>) percentages. In (<b>c</b>), areas A–D are shown.</p> Full article ">
21 pages, 7309 KiB  
Article
A Spatial Information Extraction Method Based on Multi-Modal Social Media Data: A Case Study on Urban Inundation
by Yilong Wu, Yingjie Chen, Rongyu Zhang, Zhenfei Cui, Xinyi Liu, Jiayi Zhang, Meizhen Wang and Yong Wu
ISPRS Int. J. Geo-Inf. 2023, 12(9), 368; https://doi.org/10.3390/ijgi12090368 - 5 Sep 2023
Cited by 1 | Viewed by 2243
Abstract
With the proliferation and development of social media platforms, social media data have become an important source for acquiring spatiotemporal information on various urban events. Providing accurate spatiotemporal information for events contributes to enhancing the capabilities of urban management and emergency responses. However, [...] Read more.
With the proliferation and development of social media platforms, social media data have become an important source for acquiring spatiotemporal information on various urban events. Providing accurate spatiotemporal information for events contributes to enhancing the capabilities of urban management and emergency responses. However, existing research regarding mining spatiotemporal information of events often solely focuses on textual content and neglects data from other modalities such as images and videos. Therefore, this study proposes an innovative spatiotemporal information extraction method, which extracts the spatiotemporal information of events from multimodal data on Weibo at coarse- and fine-grained hierarchical levels and serves as a beneficial supplement to existing urban event monitoring methods. This paper utilizes the “20 July 2021 Zhengzhou Heavy Rainfall” incident as an example to evaluate and analyze the effectiveness of the proposed method. Results indicate that in coarse-grained spatial information extraction using only textual data, our method achieved a spatial precision of 87.54% within a 60 m range and reached 100% spatial precision for ranges beyond 200 m. For fine-grained spatial information extraction, the introduction of other modal data, such as images and videos, resulted in a significant improvement in spatial error. These results demonstrate the ability of the MIST-SMMD (Method of Identifying Spatiotemporal Information of Social Media Multimodal Data) to extract spatiotemporal information from urban events at both coarse and fine levels and confirm the significant advantages of multimodal data in enhancing the precision of spatial information extraction. Full article
(This article belongs to the Topic Urban Sensing Technologies)
►▼ Show Figures

Figure 1

Figure 1
<p>The Overall Structure of the MIST-SMMD Process.</p> Full article ">Figure 2
<p>Flowchart of the Spatiotemporal Standardization.</p> Full article ">Figure 3
<p>Three Common Examples of Standardization (Weibo posts have been translated).</p> Full article ">Figure 4
<p>Typical High-Quality (Positive) and Low-Quality (Negative) Images.</p> Full article ">Figure 5
<p>Effect of Each Level of the Model on the Matching Results.</p> Full article ">Figure 6
<p>(<b>a</b>) Spatial Distribution of Inundation Points from Coarse-Grained Extraction in China from 18 to 20 July 2021; (<b>b</b>) Official Iundation Points and Area in Zhengzhou City from 18 to 20 July 2021.</p> Full article ">Figure 7
<p>Spatial Precision of Coarse-grained Spatial Information Extraction within Different Buffer Ranges.</p> Full article ">Figure 8
<p>Comparison of Coarse and Fine-grained Extraction.</p> Full article ">Figure 9
<p>Ablation Experiments for Fine-grained Extraction.</p> Full article ">Figure 10
<p>Matching Results of the LSGL with High-Recognizability Images (<b>a</b>) and Low-Recognizability Images (<b>b</b>).</p> Full article ">Figure 11
<p>The Effect of Noise Points on LSGL Matching Results.</p> Full article ">Figure 12
<p>LSGL Matching Results in Complex Scenarios.</p> Full article ">
22 pages, 28169 KiB  
Article
Evolution of the Urban Network in the Upper Yellow River Region of China: Enterprise Flow, Network Connections, and Influence Mechanisms—A Case Study of the Ningxia Urban Agglomeration along the Yellow River
by Jiagang Zhai, Mingji Li, Mengjiao Ming, Marbiya Yimit and Jinlu Bi
ISPRS Int. J. Geo-Inf. 2023, 12(9), 367; https://doi.org/10.3390/ijgi12090367 - 5 Sep 2023
Viewed by 1300
Abstract
Given the significant role of the Ningxia Urban Agglomeration along the Yellow River in reshaping the urban network and promoting coordinated development in the upper Yellow River region of China, this paper takes enterprise flow as the explicit manifestation of the regional urban [...] Read more.
Given the significant role of the Ningxia Urban Agglomeration along the Yellow River in reshaping the urban network and promoting coordinated development in the upper Yellow River region of China, this paper takes enterprise flow as the explicit manifestation of the regional urban network and interprets the evolution of the regional urban network structure and its influencing mechanisms through the different types of enterprise flow. The results indicate the following: (1) The external network is primarily focused on outflow investments towards North China, East China, and Northwest China. The overall inflow sources form a multi-origin structure dominated by North China and East China. Jinfeng and Xingqing serve as core hubs for enterprise exports in the external network and destinations for incoming enterprises. However, in terms of productive manufacturing connections, there is a spatial organizational pattern driven by multiple cities. (2) In the internal network, there is a concentric connection structure centered around Jinfeng and Xingqing. The productive service connections are relatively active, while the productive manufacturing connections are relatively concentrated between Jinfeng, Xingqing, Ningdong, and Lingwu. (3) In the external network, the main feature is the absorption of external elements to foster development momentum. In the internal network, Jinfeng and Xingqing serve as the contact and radiation sources, influencing various nodes. However, the driving capacity is weak. (4) The market demand and coordinated development both demonstrate significant promoting effects on the connections within the external and internal networks. The sluggish adjustment and transformation of the regional industrial structure resulted in a temporary negative inhibitory effect on the development of transformation. The negative impact of urban investment activities and the positive impact of government management are reflected within the internal network. (5) Improvements in urban management and service functions as well as external borrowing can promote connection in different networks. However, borrowing economic activity can have a negative impact in different networks. (6) Industrial agglomeration can promote enterprise connections in different networks and generate spatial spillover effects. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>The evolution from central place theory to network externality.</p> Full article ">Figure 2
<p>Location of the Ningxia Urban Agglomeration along the Yellow River.</p> Full article ">Figure 3
<p>The evolution of enterprise overall outflow in the external network. Note: Because of the small amount of data from Taiwan, Hong Kong and Macau, these regions are not included, and all the following are the same. In addition, meanings of abbreviations for various regions in China: Northeast China (NEC), North China (NC), East China (EC), Central China (CC), South China (SC), Southwest China (SWC), Northwest China (NWC). And each color represents a city. These are the same for <a href="#ijgi-12-00367-f004" class="html-fig">Figure 4</a> and <a href="#ijgi-12-00367-f005" class="html-fig">Figure 5</a>.</p> Full article ">Figure 4
<p>The evolution of the enterprise outflow of the production service industry in the external network.</p> Full article ">Figure 5
<p>The evolution of enterprise outflow of the productive manufacturing industry in the external network.</p> Full article ">Figure 6
<p>The evolution of the overall enterprise inflow in the external network. Note: And each color represents a region. These are the same for <a href="#ijgi-12-00367-f007" class="html-fig">Figure 7</a> and <a href="#ijgi-12-00367-f008" class="html-fig">Figure 8</a>.</p> Full article ">Figure 7
<p>The evolution of the enterprise inflow of the production service industry in the external network.</p> Full article ">Figure 8
<p>The evolution of the enterprise inflow of the production manufacturing industry in the external network.</p> Full article ">Figure 9
<p>The spatial pattern of overall enterprise flow in the internal network.</p> Full article ">Figure 10
<p>The spatial pattern of the overall enterprise flow of the production service industry in the internal network.</p> Full article ">Figure 11
<p>The spatial pattern of the overall enterprise flow of the production manufacturing industry in the internal network.</p> Full article ">Figure 12
<p>The dominant connection direction and urban connectivity of the network nodes.</p> Full article ">Figure 13
<p>The connection characteristics and influence mechanisms in the urban network of the Ningxia Urban Agglomeration along the Yellow River.</p> Full article ">
22 pages, 9732 KiB  
Article
Gated Recurrent Unit Embedded with Dual Spatial Convolution for Long-Term Traffic Flow Prediction
by Qingyong Zhang, Lingfeng Zhou, Yixin Su, Huiwen Xia and Bingrong Xu
ISPRS Int. J. Geo-Inf. 2023, 12(9), 366; https://doi.org/10.3390/ijgi12090366 - 5 Sep 2023
Cited by 3 | Viewed by 1389
Abstract
Considering the spatial and temporal correlation of traffic flow data is essential to improve the accuracy of traffic flow prediction. This paper proposes a traffic flow prediction model named Dual Spatial Convolution Gated Recurrent Unit (DSC-GRU). In particular, the GRU is embedded with [...] Read more.
Considering the spatial and temporal correlation of traffic flow data is essential to improve the accuracy of traffic flow prediction. This paper proposes a traffic flow prediction model named Dual Spatial Convolution Gated Recurrent Unit (DSC-GRU). In particular, the GRU is embedded with the DSC unit to enable the model to synchronously capture the spatiotemporal dependence. When considering spatial correlation, current prediction models consider only nearest-neighbor spatial features and ignore or simply overlay global spatial features. The DSC unit models the adjacent spatial dependence by the traditional static graph and the global spatial dependence through a novel dependency graph, which is generated by calculating the correlation between nodes based on the correlation coefficient. More than that, the DSC unit quantifies the different contributions of the adjacent and global spatial correlation with a modified gated mechanism. Experimental results based on two real-world datasets show that the DSC-GRU model can effectively capture the spatiotemporal dependence of traffic data. The prediction precision is better than the baseline and state-of-the-art models. Full article
(This article belongs to the Topic Geocomputation and Artificial Intelligence for Mapping)
►▼ Show Figures

Figure 1

Figure 1
<p>Effect of the topological structure. Similar trends exist between nodes.</p> Full article ">Figure 2
<p>Periodicity of traffic flow data: (<b>a</b>) Daily periodicity. (<b>b</b>) Weekly periodicity. ① traffic flow data of a workday. ② traffic flow data of a weekday. ③ traffic flow data of a holiday. ④ traffic flow data of one week.</p> Full article ">Figure 3
<p>Framework of the DSC-GRU model.</p> Full article ">Figure 4
<p>Node distribution: (<b>a</b>) Adjacent space. (<b>b</b>) Global space.</p> Full article ">Figure 5
<p>Integration process: (<b>a</b>) Adjacent space. (<b>b</b>) Global space.</p> Full article ">Figure 6
<p>The overall structure of the DSC unit.</p> Full article ">Figure 7
<p>The overall prediction process of the DSC-GRU model.</p> Full article ">Figure 8
<p>Evaluation metrics at different horizons of the model on PeMS04: (<b>a</b>) MAE. (<b>b</b>) RMSE. (<b>c</b>) MAPE.</p> Full article ">Figure 9
<p>Evaluation metrics at different horizons of the model on PeMS08: (<b>a</b>) MAE. (<b>b</b>) RMSE. (<b>c</b>) MAPE.</p> Full article ">Figure 10
<p>Visualization of prediction results: (<b>a</b>) One-day traffic flow at node 9 on PeMS04. (<b>b</b>) One-day traffic flow at node 9 on PeMS08. (<b>c</b>) One-week traffic flow at node 9 on PeMS04. (<b>d</b>) One-week traffic flow at node 9 on PeMS08.</p> Full article ">Figure 11
<p>Partial node correlation heat map: (<b>a</b>) Heatmap for PeMS04. (<b>b</b>) Heatmap for PeMS08.</p> Full article ">Figure 12
<p>Evaluation metrics under different <math display="inline"><semantics> <msup> <mi>R</mi> <mn>2</mn> </msup> </semantics></math> threshold values: (<b>a</b>) PeMS04. (<b>b</b>) PeMS08.</p> Full article ">Figure 13
<p>Evaluation metrics under different numbers of DSC hidden layer neurons: (<b>a</b>) PeMS04. (<b>b</b>) PeMS08.</p> Full article ">Figure 14
<p>Evaluation metrics under different numbers of GRU hidden layer neurons: (<b>a</b>) PeMS04. (<b>b</b>) PeMS08.</p> Full article ">
25 pages, 7400 KiB  
Article
Assessing Potable Water Access and Its Implications for Households’ Livelihoods: The Case of Sibi in the Nkwanta North District, Ghana
by Kingsley Kanjin, Richard Adade, Julia Quaicoe and Minxuan Lan
ISPRS Int. J. Geo-Inf. 2023, 12(9), 365; https://doi.org/10.3390/ijgi12090365 - 2 Sep 2023
Viewed by 4010
Abstract
Despite water being a basic human need, the residents of Sibi in Ghana’s Nkwanta North District struggle to obtain potable water, which negatively influences their livelihoods. This study aimed to evaluate the impacts on households’ livelihoods due to difficulties in accessing potable water [...] Read more.
Despite water being a basic human need, the residents of Sibi in Ghana’s Nkwanta North District struggle to obtain potable water, which negatively influences their livelihoods. This study aimed to evaluate the impacts on households’ livelihoods due to difficulties in accessing potable water and accordingly give policy recommendations. Data were collected through questionnaire surveys, interviews, geographic information systems (GIS), and remote sensing (RS) techniques. Questionnaire surveys were administered to 314 randomly selected household heads. The results indicated that the water sources available in Sibi were not sufficient; the boreholes and public tabs/standpipes in the communities were not dependable for regular access. As a result, households needed to depend on distant streams and dams for water. The households generally spent more than two hours at the water sources to collect water. Evidently, the Sibi residents did not have sufficient access to potable water, which severely affected their livelihoods. It is recommended that government agencies collaborate with related non-governmental organizations (NGOs) to help expand potable water projects in Sibi, Ghana. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Geographical location of the study area within the Nkwanta North District of Ghana. Source: Authors’ construct, 2023.</p> Full article ">Figure 2
<p>A summary of the research methods.</p> Full article ">Figure 3
<p>Sustainable Livelihood Framework. Source: Adapted from DFID (1999).</p> Full article ">Figure 4
<p>Connections between water access, water management, and livelihoods. Source: Authors’ construct (2023).</p> Full article ">Figure 5
<p>Locations of available water point sources in the study communities.</p> Full article ">Figure 6
<p>Functionality of the water point sources in the three study communities.</p> Full article ">Figure 7
<p>Cost–distance to water point sources in the study communities.</p> Full article ">Figure 8
<p>Average waiting times of respondents at the water point sources.</p> Full article ">Figure 9
<p>Photo showing people collecting water with pans. Source: Authors captured (2021).</p> Full article ">Figure 10
<p>Water storage facilities. Source: Authors captured (2021).</p> Full article ">Figure A1
<p>100 m by 100 m square sampling grids.</p> Full article ">Figure A2
<p>Distribution of the overall livelihood impact index.</p> Full article ">
19 pages, 2880 KiB  
Review
A Critical Review of Smart City Frameworks: New Criteria to Consider When Building Smart City Framework
by Fan Shi and Wenzhong Shi
ISPRS Int. J. Geo-Inf. 2023, 12(9), 364; https://doi.org/10.3390/ijgi12090364 - 1 Sep 2023
Cited by 4 | Viewed by 3440
Abstract
In the face of persistent challenges posed by urbanization and climate change, the contemporary era has witnessed a growing urgency for urban intelligence and sustainable development. Consequently, a plethora of smart city schedules and policies have emerged, with smart city assessment serving as [...] Read more.
In the face of persistent challenges posed by urbanization and climate change, the contemporary era has witnessed a growing urgency for urban intelligence and sustainable development. Consequently, a plethora of smart city schedules and policies have emerged, with smart city assessment serving as a pivotal benchmark for gauging policy effectiveness. However, owing to the inherent ambiguity of the smart city definition and the complexity of application scenarios, designers and decision-makers often struggle to ascertain their desired assessment frameworks swiftly and effectively. In this context, our study undertook a comprehensive analysis and comparative assessment of 33 recently introduced or inferred evaluation frameworks, drawn from a broad spectrum of extensive and longstanding research efforts. The overarching goal was to provide valuable reference points for designers and decision-makers navigating this intricate landscape. The assessment was conducted across seven key dimensions: generalizability, comprehensiveness, availability, flexibility, scientific rigor, transparency, and interpretability. These criteria hold the potential not only to guide the development trajectory and focus of upcoming smart city assessment models but also to serve as invaluable guidelines for stakeholders evaluating the outcomes of such models. Furthermore, they can serve as robust support for designers and decision-makers in their pursuit of targeted frameworks. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Six dimensions of smart city adapted from Cohen’s six-wheel model.</p> Full article ">Figure 2
<p>Checking process framework for the criterion ‘Scientific’.</p> Full article ">Figure 3
<p>Analysis result pertaining to the criteria ‘Generalizability’.</p> Full article ">Figure 4
<p>Analysis result pertaining to the criteria ‘Generalizability’ in the selected frameworks.</p> Full article ">Figure 5
<p>Analysis result pertaining to the criteria ‘Comprehensive’.</p> Full article ">Figure 6
<p>Analysis result pertaining to the criteria ‘Comprehensive’ in the selected frameworks.</p> Full article ">Figure 7
<p>Analysis result pertaining to the criteria ‘Availability’.</p> Full article ">Figure 8
<p>Analysis result pertaining to the criteria ‘Availability’ in the selected frameworks.</p> Full article ">Figure 9
<p>Analysis result pertaining to the criteria ‘Flexibility’.</p> Full article ">Figure 10
<p>Analysis result pertaining to the criteria ‘Scientific’.</p> Full article ">Figure 11
<p>Analysis result pertaining to the criteria ‘Scientific’ of specified weighting methods.</p> Full article ">Figure 12
<p>Analysis result pertaining to the criteria ‘Transparency’.</p> Full article ">Figure 13
<p>Analysis result pertaining to the criteria ‘Interpretability’.</p> Full article ">
19 pages, 24985 KiB  
Article
A Method for Regularizing Buildings through Combining Skeleton Lines and Minkowski Addition
by Guoqing Chen and Haizhong Qian
ISPRS Int. J. Geo-Inf. 2023, 12(9), 363; https://doi.org/10.3390/ijgi12090363 - 1 Sep 2023
Cited by 1 | Viewed by 1261
Abstract
With the increasing availability of remote sensing images, the regularization of jagged building outlines extracted from high-resolution remote sensing images has become a current research hotspot. Based on an existing method proposed earlier by this author for extracting the skeleton lines of buildings [...] Read more.
With the increasing availability of remote sensing images, the regularization of jagged building outlines extracted from high-resolution remote sensing images has become a current research hotspot. Based on an existing method proposed earlier by this author for extracting the skeleton lines of buildings through integrating vector and raster data using jagged building skeleton lines as the input data, a new method is proposed here for regularizing building outlines through combining the skeleton lines with the Minkowski addition algorithm. Since the size and orientation of the structuring elements remain constant in the traditional morphological method, they can easily lead to large changes in the area between the regularized results and area of the original building. In this work, structuring elements are constructed with the adaptive adjustment of size and orientation. The proposed method has an outstanding ability to maintain the area of the original building. The orthogonal characteristics of the building can be better preserved via rotating the structuring elements. Finally, the angular bisector method is used to dissipate conflicts among the redundant vertices in the building outlines. In comparison to the simplification method used in QGIS software, the method proposed in this paper could reduce the variation in the area while maintaining the orthogonal characteristics of the building more significantly. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Minkowski addition [<a href="#B35-ijgi-12-00363" class="html-bibr">35</a>]. (<b>a</b>) Region to be processed. (<b>b</b>) Structuring element. Where the position indicated by the red arrow is the origin of the structuring element (<b>c</b>) Result of Minkowski addition.</p> Full article ">Figure 2
<p>Processing flow of the method proposed in this paper.</p> Full article ">Figure 3
<p>Shape and origin of structuring elements.</p> Full article ">Figure 4
<p>Shape and origin of structuring elements. (<b>a</b>) Building without cross points and (<b>b</b>) building with cross points.</p> Full article ">Figure 5
<p>Definition of angle of rotation.</p> Full article ">Figure 6
<p>Angle of rotation. (<b>a</b>) Position of the first rotation. (<b>b</b>) Position of the second rotation.</p> Full article ">Figure 7
<p>Results of the method in this paper. (<b>a</b>) Results before conflict resolution. (<b>b</b>) Details in the black box of <a href="#ijgi-12-00363-f007" class="html-fig">Figure 7</a>a. The location of the conflict is marked with a black circle.</p> Full article ">Figure 8
<p>Conflict resolution process. (<b>a</b>) Unitized direction vector. (<b>b</b>) Vector of interior angle bisector. (<b>c</b>) Opposite orientation vector of angle bisector. (<b>d</b>) Details of conflict resolution.</p> Full article ">Figure 9
<p>Building nodes before and after conflict resolution. (<b>a</b>) Number of nodes before conflict resolution. (<b>b</b>) Number of nodes after conflict resolution.</p> Full article ">Figure 10
<p>Study area. (<b>a</b>) High-resolution remote sensing images. (<b>b</b>) Extracted vector buildings.</p> Full article ">Figure 11
<p>Regularization results obtained using QGIS software and the proposed method in this study. (<b>a</b>,<b>d</b>,<b>g</b>) High-resolution remote sensing images of the study area. (<b>b</b>,<b>e</b>,<b>h</b>) Regularization results obtained using QGIS software. (<b>c</b>,<b>f</b>,<b>i</b>) Regularization results obtained using the proposed method in this study.</p> Full article ">Figure 12
<p>Results of two methods. (<b>a</b>) Skeleton line before simplification. (<b>b</b>) Skeleton line after simplification. (<b>c</b>) Results of two methods.</p> Full article ">Figure 13
<p>Regularization of a building with a hole. (<b>a</b>) Building to be regularized before filling the hole and (<b>b</b>) after filling the hole. (<b>c</b>) The regularization results of two methods.</p> Full article ">Figure 14
<p>Results of QGIS software and proposed method for regularization of typically distributed buildings. (<b>a1</b>–<b>a5</b>) C-shaped, (<b>b1</b>–<b>b5</b>) F-shaped, (<b>c1</b>–<b>c5</b>) H-shaped, (<b>d1</b>–<b>d5</b>) I-shaped, (<b>e1</b>–<b>e5</b>) L-shaped, (<b>f1</b>–<b>f5</b>) T-shaped, (<b>g1</b>–<b>g5</b>) V-shaped.</p> Full article ">Figure 15
<p>Method for calculating the average of the sum of vertex angles. sNode denotes the number of vertices of the simplified building.</p> Full article ">
16 pages, 5332 KiB  
Article
Exploring Divergent Patterns and Dynamics of Urban and Active Rural Developments—A Case Study of Dezhou City
by Huimin Zhong, Zhengjia Liu and Yihang Huang
ISPRS Int. J. Geo-Inf. 2023, 12(9), 362; https://doi.org/10.3390/ijgi12090362 - 1 Sep 2023
Cited by 2 | Viewed by 1303
Abstract
Clarifying urban-rural spatial explicit structure changes is of great significance for understanding the urban-rural relationship evolution. Previous studies have mostly focused on urban internal spatial structure evolutions and less on the regional scale when it comes to exploring urban and rural evolutions. Nighttime [...] Read more.
Clarifying urban-rural spatial explicit structure changes is of great significance for understanding the urban-rural relationship evolution. Previous studies have mostly focused on urban internal spatial structure evolutions and less on the regional scale when it comes to exploring urban and rural evolutions. Nighttime light can timely reflect the human activities in regions and provides great potential for investigating the evolutions of urban and rural spatial explicit structures. Here, taking Dezhou City, a rapidly urbanizing city in China, as a case study, we employed the local contour tree method and nighttime light data to map urban and active rural extents from 2012 to 2020 and further explored their respective development processes. This study showed that unlike in rural regions, the internally explicit structures of urban regions were more complex, and there were often multiple hotspots inside them. The area of the urban-rural region increased significantly by 39.3% from 2012 to 2020 (p < 0.05). Populations were greatly responsible for the spatial explicit structure changes of urban and active rural regions. The urban and rural region rankings of the identified counties were basically consistent with the urban and rural population rankings. Unlike the perspectives of earlier land use (i.e., built-up land or impervious surface), this study underlined urban and active rural regions in view of the scope of active human activities. These results can likely help policymakers understand current active human activity extents and provide a data-based reference for future public services and infrastructure planning. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Location, digital elevation model (DEM) and land use map in 2020 of the study area.</p> Full article ">Figure 2
<p>A schematic diagram for nighttime light data before (<b>a</b>,<b>A</b>) and after (<b>b</b>,<b>B</b>) conditional function model reconstruction.</p> Full article ">Figure 3
<p>Urban and rural region detection using localization contour tree method, referring to the earlier studies [<a href="#B36-ijgi-12-00362" class="html-bibr">36</a>,<a href="#B42-ijgi-12-00362" class="html-bibr">42</a>]. (<b>a</b>) Nighttime light intensity contour map; (<b>b</b>) Regular contour tree; and (<b>c</b>) Simplified contour tree.</p> Full article ">Figure 4
<p>Flowchart of the methodology adopted in this study.</p> Full article ">Figure 5
<p>Nighttime light contour map of 2020. (<b>a</b>) Dezhou City, (<b>b</b>) Decheng and Lincheng County of Dezhou City and (<b>c</b>) Pingyuan County of Dezhou City.</p> Full article ">Figure 6
<p>(<b>a</b>) Spatial distribution maps of urban-rural regions in 2020. (<b>b</b>) Histogram of urban region and population in each county. (<b>c</b>) Histogram of rural region and population in each county.</p> Full article ">Figure 7
<p>Spatial and temporal evolution characteristics of urban-rural regions during 2012–2020.</p> Full article ">Figure 8
<p>Annual values of area (red line) and nighttime lights (blue line) of urban-rural regions.</p> Full article ">Figure 9
<p>Scatter plots of the urban (blue dots), rural (orange dots) and Decheng (red dots) populations and corresponding areas.</p> Full article ">Figure A1
<p>The area statistics of urban and rural areas in 2020 were obtained by using nighttime light data and local contour tree and urban and rural division method.</p> Full article ">Figure A2
<p>The rural population change in Dezhou City during 2012–2020.</p> Full article ">Figure A3
<p>Satellite remote sensing images of some rural areas in the results of urban-rural structure identification, of which, (<b>a</b>–<b>g</b>) town, (<b>h</b>–<b>j</b>) factory and (<b>k</b>–<b>l</b>) villages.</p> Full article ">
24 pages, 20361 KiB  
Article
Evaluation of Machine Learning Algorithms in the Classification of Multispectral Images from the Sentinel-2A/2B Orbital Sensor for Mapping the Environmental Dynamics of Ria Formosa (Algarve, Portugal)
by Flavo Elano Soares de Souza and José Inácio de Jesus Rodrigues
ISPRS Int. J. Geo-Inf. 2023, 12(9), 361; https://doi.org/10.3390/ijgi12090361 - 1 Sep 2023
Cited by 1 | Viewed by 1681
Abstract
With the growing availability of remote sensing orbital spatial data, the applications of machine learning (ML) algorithms have been leveraging the field of process automation in image classification. The present work aimed to evaluate the precision and accuracy of ML algorithms in the [...] Read more.
With the growing availability of remote sensing orbital spatial data, the applications of machine learning (ML) algorithms have been leveraging the field of process automation in image classification. The present work aimed to evaluate the precision and accuracy of ML algorithms in the classification of Sentinel 2A/2B images from an area of high environmental dynamics, such as Ria Formosa (Algarve, Portugal). The images were submitted to classification by groups of ML algorithms such as the Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbors (KNN), and Decision Tree (DT). The Orfeo Toolbox (OTB) open-source programming package made the algorithms available. Ten samples were collected for each of the 14 land use and cover classes in the Ria Formosa area, totaling 140 samples. Of these, 70% were for training and 30% for validating the classification. The evaluation metrics used were the class discrimination measures: Recall (R), the Global Kappa Index (k), and the General Accuracy Index (OA). The results showed that the KNN and DT algorithms demonstrated a greater discrimination capacity for most classes. SVM and RF significantly improved class discrimination when using larger samples for training. Merging the classified images significantly improved the classification accuracy, ranging from 71% to 81%. This evaluation made it possible to define sets of ML algorithms sensitive to change detection for mapping and monitoring dynamic environments. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Location of the study area. Mosaic Sentinel 2A and 2B base image in false color composite (RGB 12/3/2).</p> Full article ">Figure 2
<p>Image pre-processing software and steps used in this work.</p> Full article ">Figure 3
<p>Polygons of classes selected for training the ML algorithms. Mosaic Sentinel 2A and 2B base image in True Color Rendering (RGB 432).</p> Full article ">Figure 4
<p>Flowchart of the methodology addressed in this work.</p> Full article ">Figure 5
<p>Kappa indexes (<span class="html-italic">Kt</span>) and Overall Accuracy (<span class="html-italic">OAt</span>) were obtained in training (t) using the intersected segments as samples.</p> Full article ">Figure 6
<p>Confusion matrices with Recall (<span class="html-italic">R</span>) values obtained in the training of the algorithms using the intersection samples between the image segments. In (<b>a</b>) SVM; (<b>b</b>) Random Forest; (<b>c</b>) KNN; and (<b>d</b>) Decision Tree.</p> Full article ">Figure 6 Cont.
<p>Confusion matrices with Recall (<span class="html-italic">R</span>) values obtained in the training of the algorithms using the intersection samples between the image segments. In (<b>a</b>) SVM; (<b>b</b>) Random Forest; (<b>c</b>) KNN; and (<b>d</b>) Decision Tree.</p> Full article ">Figure 7
<p>Global Kappa Indexes (<span class="html-italic">Kv</span>) and Overall Accuracy (<span class="html-italic">OAv</span>) were obtained in the validation (v) of classified images that were trained using samples of intersected segments.</p> Full article ">Figure 8
<p>Kappa Indexes (<span class="html-italic">Kt</span>) and Overall Accuracy (<span class="html-italic">OAt</span>) obtained in training (t) using entire segments as samples.</p> Full article ">Figure 9
<p>Confusion matrices with Recall (<span class="html-italic">R</span>) values (discrimination sensitivity) were obtained in the training of the algorithms using samples of entire segments in each image. In (<b>a</b>) SVM; (<b>b</b>) Random Forest; (<b>c</b>) KNN; and (<b>d</b>) Decision Tree.</p> Full article ">Figure 10
<p>Kappa Indexes (<span class="html-italic">Kv</span>) and Overall Accuracy (<span class="html-italic">OAv</span>) obtained in the validation (v) of the classified images with the entire segments.</p> Full article ">Figure 11
<p>Global Kappa Index (<span class="html-italic">Kt</span>) of the training samples using the intersected (Int) and integer or total (Tot) segments in (<b>a</b>); and validation (<span class="html-italic">Kv</span>) of image classifications using intersected segments (Int) and total integers (Tot) in (<b>b</b>); Overall Accuracy Index (<span class="html-italic">OAt</span>) of the training samples using the intersected segments (Int) and total integers (Tot) in (<b>c</b>); and validation (<span class="html-italic">OAv</span>) of image classifications, using the intersected segments (Int) and total integers (Tot) in (<b>d</b>).</p> Full article ">Figure 12
<p>Total pixels sampled in each class for image training using whole (circles) and intersected (triangle) segments. The pixel area corresponds to 10 m × 10 m = 100 m<sup>2</sup> spatial resolution.</p> Full article ">Figure 13
<p>RGB 832 (VISNIR) composites in the false color of Sentinel 2A/2B images and the respective classifications obtained by merging the SVM, Random Forest, KNN, and Decision Tree results algorithms, when trained with samples of intersected segments.</p> Full article ">
23 pages, 10144 KiB  
Article
LBS Tag Cloud: A Centralized Tag Cloud for Visualization of Points of Interest in Location-Based Services
by Xiaoqiang Cheng, Zhongyu Liu, Huayi Wu and Haibo Xiao
ISPRS Int. J. Geo-Inf. 2023, 12(9), 360; https://doi.org/10.3390/ijgi12090360 - 1 Sep 2023
Cited by 1 | Viewed by 1878
Abstract
Taking location-based service (LBS) as the research scenario and aiming at the limitation of visualizing LBS points of interest (POI) in conventional web maps, this article proposes a visualization method of LBS-POI based on tag cloud, which is called “LBS tag cloud”. In [...] Read more.
Taking location-based service (LBS) as the research scenario and aiming at the limitation of visualizing LBS points of interest (POI) in conventional web maps, this article proposes a visualization method of LBS-POI based on tag cloud, which is called “LBS tag cloud”. In this method, the user location is taken as the layout center, and the name of the POI is converted into a text tag and then placed around the center. The tags’ size, color, and placement location are calculated based on other attributes of the POI. The calculation of placement location is at the core of the LBS tag cloud. Firstly, the tag’s initial placement position and layout priority are calculated based on polar coordinates, and the tags are placed in the initial placement position in the order of layout priority. Then, based on the force-directed model, a repulsive force is applied to the tag from the layout center to make it move to a position without overlapping with other tags. During the move, the quadtree partition of the text glyph is used to optimize the detection of overlaps between tags. Taking scenic spots as an example, the experimental results show that the LBS tag cloud can present the attributes and distribution of POIs completely and intuitively and can effectively represent the relationship between the POIs and user location, which is a new visualization form suitable for spatial cognition. Full article
(This article belongs to the Special Issue Application of Geographical Information System in Urban Design, Management or Evaluation)
►▼ Show Figures

Figure 1

Figure 1
<p>The visual variables of text tags.</p> Full article ">Figure 2
<p>Map-based visualization versus LBS tag cloud.</p> Full article ">Figure 3
<p>How to place tags based on polar coordinates.</p> Full article ">Figure 4
<p>The radial displacement of tags.</p> Full article ">Figure 5
<p>How to ensure the correct order of near and far by considering tag adjacency.</p> Full article ">Figure 6
<p>Schematic diagram of approximate glyph outline based on quadtree. The red border in (<b>a</b>) is the rectangular outline of the Text element; (<b>b</b>) the incomplete quadtree corresponding to the character “地” is drawn with a solid red line; and (<b>c</b>) only the leaf nodes covering the edges of the glyph outline are drawn while retaining the rectangular outline.</p> Full article ">Figure 7
<p>LBS tag cloud containing all attractions (the tag position indicates the travel time and the font size indicates the number of comments).</p> Full article ">Figure 8
<p>LBS tag cloud containing all attractions (the position and color indicate the travel time, and the font size indicates the number of comments).</p> Full article ">Figure 9
<p>LBS tag cloud expressing the public transport express index.</p> Full article ">Figure 10
<p>LBS tag cloud for attractions with a travel time of less than 1 h (the tag position indicates the travel time, the font size indicates the number of comments, and the color indicates the rating).</p> Full article ">Figure 11
<p>LBS tag cloud for attractions rated by users between 4.6 and 5.0 (the color indicates the travel time).</p> Full article ">Figure 12
<p>LBS tag cloud for attractions rated by users between 4.6 and 5.0 (the travel time is part of the tag text).</p> Full article ">Figure 13
<p>POI visualization based on Baidu Maps.</p> Full article ">
20 pages, 4003 KiB  
Article
Spatio-Temporal Relevance Classification from Geographic Texts Using Deep Learning
by Miao Tian, Xinxin Hu, Jiakai Huang, Kai Ma, Haiyan Li, Shuai Zheng, Liufeng Tao and Qinjun Qiu
ISPRS Int. J. Geo-Inf. 2023, 12(9), 359; https://doi.org/10.3390/ijgi12090359 - 1 Sep 2023
Viewed by 1793
Abstract
The growing proliferation of geographic information presents a substantial challenge to the traditional framework of a geographic information analysis and service. The dynamic integration and representation of geographic knowledge, such as triples, with spatio-temporal information play a crucial role in constructing a comprehensive [...] Read more.
The growing proliferation of geographic information presents a substantial challenge to the traditional framework of a geographic information analysis and service. The dynamic integration and representation of geographic knowledge, such as triples, with spatio-temporal information play a crucial role in constructing a comprehensive spatio-temporal knowledge graph and facilitating the effective utilization of spatio-temporal big data for knowledge-driven service applications. The existing knowledge graph (or geographic knowledge graph) takes spatio-temporal as the attribute of entity, ignoring the role of spatio-temporal information for accurate retrieval of entity objects and adaptive expression of entity objects. This study approaches the correlation between geographic knowledge and spatio-temporal information as a text classification problem, with the aim of addressing the challenge of establishing meaningful connections among spatio-temporal data using advanced deep learning techniques. Specifically, we leverage Wikipedia as a valuable data source for collecting and filtering geographic texts. The Open Information Extraction (OpenIE) tool is employed to extract triples from each sentence, followed by manual annotation of the sentences’ spatio-temporal relevance. This process leads to the formation of quadruples (time relevance/space relevance) or quintuples (spatio-temporal relevance). Subsequently, a comprehensive spatio-temporal classification dataset is constructed for experiment verification. Ten prominent deep learning text classification models are then utilized to conduct experiments covering various aspects of time, space, and spatio-temporal relationships. The experimental results demonstrate that the Bidirectional Encoder Representations from Transformer-Region-based Convolutional Neural Network (BERT-RCNN) model exhibits the highest performance among the evaluated models. Overall, this study establishes a foundation for future knowledge extraction endeavors. Full article
(This article belongs to the Special Issue Innovative GIS Models and Approaches for Large Environmental and Urban Applications in the Age of AI)
►▼ Show Figures

Figure 1

Figure 1
<p>The overall presented framework, which is divided into three main modules: dataset construction, model training, and experimental validation. The yellow part is the dataset construction, the green part indicates the model used in this experiment, and the pink part indicates the experimental validation phase.</p> Full article ">Figure 2
<p>Some samples of the spatio-temporal correlation text classification dataset.</p> Full article ">Figure 3
<p>Structure of BERT_RCNN text categorization model.</p> Full article ">Figure 4
<p>Structure of circular convolutional layer.</p> Full article ">Figure 5
<p>Loss function graph for temporal text classification.</p> Full article ">Figure 6
<p>Spatial text classification loss function graph.</p> Full article ">Figure 7
<p>Spatio-temporal text classification loss function graph.</p> Full article ">
24 pages, 8082 KiB  
Article
Chinese Modern Architectural Heritage Resources: Perspectives of Spatial Distribution and Influencing Factors
by Yidan Liao, Jeremy Cenci and Jiazhen Zhang
ISPRS Int. J. Geo-Inf. 2023, 12(9), 358; https://doi.org/10.3390/ijgi12090358 - 31 Aug 2023
Cited by 11 | Viewed by 2553
Abstract
Architectural heritage refers to buildings, complexes, and sites with historical, cultural, artistic, technological, and geographical values, including ancient buildings, historical buildings, places of interest, dwellings, and industrial sites. China’s 20th-Century Architectural Heritage List is a state-level list that includes architecture of historical, cultural, [...] Read more.
Architectural heritage refers to buildings, complexes, and sites with historical, cultural, artistic, technological, and geographical values, including ancient buildings, historical buildings, places of interest, dwellings, and industrial sites. China’s 20th-Century Architectural Heritage List is a state-level list that includes architecture of historical, cultural, technological, and artistic value in China in the 20th century. It is the carrier of the past century and the monument to witnessing the change in human knowledge, culture, technology, and even art. This list is from China, a country with a vast land area, a densely populated population, and numerous architectural relics. This study used ArcGIS to analyze 597 cases in 6 batches in China’s 20th-Century Architectural Heritage List. Its spatial structure was studied by calculating the nearest neighbor index, Gini coefficient, imbalance index, and kernel density. The results showed that the distribution of the Chinese modern architectural heritage resources is cohesive and uneven in China. Next, the geographical detector model was used to analyze its influencing factors from the perspective of 12 factors. This study found that the spatial distribution of this type of resource was condensed. The provincial level showed a distribution pattern of seven centers with one core and multiple scattered points. Its distribution in 34 administrative regions is extremely uneven, with 57.29% being located in North and East China. It also focused on analyzing five influencing factors, namely, topography, regional status, culture and education, social and economic development level, and external contact. Exploring its spatial structure and influencing factors will not only enable a comprehensive understanding of the development context and current situation of 20th-century architectural heritage, but also provide a reference for its protection and sustainable use. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Geographical distribution of China’s 20th-century architectural heritage.</p> Full article ">Figure 2
<p>Flowchart of research methodology.</p> Full article ">Figure 3
<p>Lorenz curve of the number of China’s 20th-century architectural heritage.</p> Full article ">Figure 4
<p>China’s 20th-century architectural heritage is distributed in different levels of cities.</p> Full article ">Figure 5
<p>Kernel density distribution of China’s 20th-century architectural heritage.</p> Full article ">Figure 6
<p>Nine periods based on the different development stages of China’s 20th-century architectural heritage.</p> Full article ">Figure 7
<p>Directional distribution of China’s 20th-century architectural heritage within nine time periods.</p> Full article ">Figure 8
<p>Deviation angle of China’s 20th-century architectural heritage spatial direction in different historical stages.</p> Full article ">Figure 9
<p>(<b>a</b>) Statistics on the categories of China’s 20th-century architectural heritage. (<b>b</b>) Statistics on the subcategories of China’s 20th-century architectural heritage.</p> Full article ">Figure 10
<p>Kernel density distribution of different types of China’s 20th-century architectural heritage.</p> Full article ">Figure 11
<p>Tianjin Second Workers’ Cultural Palace.</p> Full article ">Figure 12
<p>Spatial relationships between China’s 20th-century architectural heritage distribution and topography factors.</p> Full article ">Figure 13
<p>The Red Mansion of Peking University.</p> Full article ">Figure 14
<p>Stacking a map of each batch of China’s 20th-century architectural heritage in each province.</p> Full article ">
18 pages, 2914 KiB  
Article
Measuring the Spatial Accessibility of Parks in Wuhan, China, Using a Comprehensive Multimodal 2SFCA Method
by Kainan Mao, Jingzhong Li and Haowen Yan
ISPRS Int. J. Geo-Inf. 2023, 12(9), 357; https://doi.org/10.3390/ijgi12090357 - 31 Aug 2023
Cited by 4 | Viewed by 1820
Abstract
The spatial accessibility of urban parks is an important indicator of the livability level of cities. In this paper, we propose a comprehensive multimodal two-step floating catchment area (CM2SFCA) method which integrates supply capacity, the selection probability of individuals, and variable catchment sizes [...] Read more.
The spatial accessibility of urban parks is an important indicator of the livability level of cities. In this paper, we propose a comprehensive multimodal two-step floating catchment area (CM2SFCA) method which integrates supply capacity, the selection probability of individuals, and variable catchment sizes into the traditional multimodel 2SFCA method. This method is used to measure park accessibility in Wuhan, China. The results show that the spatial distribution of park accessibility under the proposed method is variant. High accessibility areas are clustered near the Third Ring Road with strong supply capacity parks, and low accessibility areas are distributed in the western and southern regions. Compared with the single-model accessibility (bicycling, driving, and public transit) method, we found that the multimodal spatial accessibility, combining the characteristics of three single transportations, can provide a more realistic evaluation. We also explore the spatial relationship between park accessibility and population density by bivariate local Moran’s I statistic and find that the Low Ai-High Pi area is located in the center of the study area, and the Low Ai-Low Pi area is located at the edge of the study area, with a relatively discrete distribution of parks and weak supply capacity. These findings may provide some insights for urban planners to formulate effective policies and strategies to ease the spatial inequity of urban parks. Full article
►▼ Show Figures

Figure 1

Figure 1
<p>Flowchart of the comprehensive multimodal 2SFCA method.</p> Full article ">Figure 2
<p>The spatial distribution of (<b>a</b>) population density and (<b>b</b>) parks in the inner city of Wuhan, China.</p> Full article ">Figure 3
<p>Spatial distribution of (<b>a</b>) park accessibility and (<b>b</b>) hot spot analysis under the CM2SFCA method.</p> Full article ">Figure 4
<p>Spatial distribution of park accessibility under single models: (<b>a</b>) bicycling, (<b>b</b>) driving, and (<b>c</b>) public transit.</p> Full article ">Figure 5
<p>Spatial distribution of park accessibility using: (<b>a</b>) Huff model-based 2SFCA method, (<b>b</b>) V2SFCA method.</p> Full article ">Figure 6
<p>Spatial distribution of the association between park accessibility (Ai) and population density (Pi).</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-39
Previous Issue
Next Issue
Back to TopTop