Lifestyle transformations in the US are creating distinct regional visitation behaviours, particularly in terms of time spent at various drinking Points of Interest (POIs) for alcoholic beverages, an aspect often overlooked in the current literature. This study proposes a framework that integrates human mobility data from Advan, the Centers for Disease Control and Prevention on Unhealthy Drinking Behaviors, and the Census Bureau, offering data into the socio-economic and demographic dimensions of POIs. By examining median dwell times and establishing a probability distribution for POI categories across US counties, the research employs the Jensen-Shannon divergence metric to identify patterns of similarity and difference. The study constructs an undirected weighted network, linking counties with weights reflecting dissimilarities in median dwell time probabilities. Using the community-Louvain method, it clusters these counties into aspiring, moderate, and bustling communities, showcasing their common behaviours. Through the development of six hypotheses concerning drinking, demographics, socioeconomics, environment, and accessibility, significant variances between these communities are quantified. An explanatory model was created to predict median dwell time, measuring the significance of features used in the hypothesis phase. The findings contribute to understanding dwell time to study visitation dynamics in POIs.
Unmanned Aerial Vehicles (UAVs) have become integral tools for data acquisition in various applications, particularly in urban areas where high-resolution spatial information is essential for planning, monitoring, and decision-making. Traditional UAV surveys rely on Ground Control Points (GCPs) to enhance georeferencing accuracy. However, in urban environments characterized by complex topography and limited accessibility, deploying GCPs can be challenging. This paper investigates the accuracy of UAV-based mapping systems in urban areas without the use of GCPs. The study quantifies positional accuracy, point cloud density, and overall mapping precision achieved by the UAV systems without GCPs. Factors influencing accuracy, such as flight altitude, camera specifications, and urban landscape characteristics, are systematically analyzed. Comparative assessments against traditional GCP-dependent UAV surveys are performed to evaluate the reliability of the proposed methodology. Results reveal the feasibility and limitations of GCP-free UAV mapping in urban environments, providing valuable insights for researchers, practitioners, and decision-makers involved in urban planning, infrastructure management, and environmental monitoring. The results show that the difference between the coordinates of checkpoints measured by GPS and Total Station and the coordinates of the same checkpoints determined by drone is quite small. Horizontal accuracy is ± 0.026m and vertical accuracy is ± 0.034m.
In the world of underground utilities where companies use sketch maps to communicate the location of their buried services, asset locator companies collect vast amounts of data in the form of sketches of construction work areas while performing daily operations to locate assets in the field. Leveraging this information can improve their ability to serve clients who require assets to be located during construction work. Typically, construction companies or homeowners must contact a third-party locator company to locate assets such as gas or electrical lines on their property, and the locator company sends a team on-site to perform the task. In this paper, we propose an end-to-end machine learning based pipeline that extracts the knowledge contained in sketches from previous locator jobs, then uses this knowledge to automatically determine the location of assets in the field. Specifically, we infer the GPS coordinates of the assets using both the sketches and GPS coordinates of the work area. Our pipeline comprises two main components: the first component implements several object detection models to extract all relevant data (streets, buildings, asset, street names, building numbers, arrows,...) from the sketches, while the second component implements a graph neural network based model to infer the asset location as a GPS coordinates by parsing the relationship between the extracted data which is fed to the model as a graph. Our approach has the potential to significantly improve the efficiency and accuracy of asset location, benefiting both asset locator companies and their clients.
Using the high-resolution image from UAV is a major challenge for crop classification. The lighting conditions play an essential role in altering spectral reflectance, affecting the classification accuracy. With this barrier, it is difficult to use a single training data set in machine learning and apply to the images taken on different dates and times. This study aims to demonstrate the application of vegetation indices combined with the red, green, or blue dataset to enhance the image classification by using the Unsupervised classification method with the K-mean technic. This study shows the experiments of using three same image layers in three different windows of mean shift filter and presents the experiments of using three different image layers in three different windows. The optimal dataset shows Thailand's best performance of eight traditional cassava fields. As a result, three of the same BI and Blue layers presented the optimal dataset, Kappa coefficient = 0.864938 and 0.729271 for soil and plant classification, respectively. Alternatively, using a different dataset as the input to three windows of the mean shift filters was considered in this study to compare the results with the single-type input. The criterion of one primary dataset and two indices showed the optimal dataset among those experimental results (Kappa coefficient > 0.948681) in soil classification, as well as the optimal dataset for plant classification (Kappa coefficient > 0.773652). It can be concluded that combining complementary spectral features captured by different indices makes the classification model more robust and capable of capturing a broader range of soil characteristics, resulting in higher accuracy and reliability in delineating soil boundaries and plants in the cassava fields.
The urban population is growing rapidly. It has unpredictable nature in terms of its spatial location sometimes it grows faster near to highways and sometimes near to central business district. But it is unpredictable in nature. In this study we will investigate that which type of growth will take place in particular scenario. The scenario means here a trend in growth whether it is compact growth within the city limits or uncoordinated and unplanned urban sprawl or a well-coordinated and planned development in peri urban areas. This is totally dependent on the policy and planning frameworks of city's administrators and its stack-holders. The scope of work is to investigate the type of growth taking place in coming years and suggest the suitable land for development accordingly along with the influencing factors of the growth. The task also includes proper justification for taking appropriate influencing factor those which are affecting the growth for urban sprawl modeling. Cellular Automata modelling has been used for Urban Land-use modeling and prediction of land –use whereas System Dynamics Approach has been taken into consideration for creation of multi-scenario and calculation of land-use demand for scenarios. The result gives a framework for multi-scenario simulation of the land-use.
Understanding crowd dynamics is critical for public safety, transportation optimization, and disease control. Federated learning, a privacy-preserving machine-learning approach, suggests to train models on the devices where data is generated. This maintains server independence for latency-sensitive applications and ensures data privacy as data never leaves the device. Additionally, federated learning facilitates load balancing among participants. Unlike traditional machine learning solutions, federated learning aggregates model parameters of all participating entity. Thus, aggregation in federated learning is challenging. In this paper, we propose Tidal Crowds, a federated crowd flow prediction algorithm, leveraging the benefits of federated learning. Tidal Crowds comes in three parts, data acquisition and processing, machine learning algorithm, and aggregation algorithm. The machine learning algorithm is based on DeepSTN+ that is capable of handling long-range spatial dependencies. Aggregation algorithms include FedAvg and FedProxy. Tidal Crowds aims to examine and enhance prediction accuracy while providing server independence and preserving privacy using federated learning. In our experiments, Tidal Crowds outperforms the original DeepSTN+ in various settings.
In response to the challenges of data scarcity, the dynamic nature of farm geography, and fluctuating production volumes, a top-down methodology was developed in this study, to address the lack of comprehensive data when analyzing the perishable food supply chain. High-density farmlands were delineated into Zones, and macro-level, such as city-specific farm characteristics, annual production volume figures, and complexities from the Jordanian Ministry of Agriculture and the Department of Statistics, were utilized as preset parameters. Farm locations were randomly placed in a 2-D linear manner, emulating real-life complexities. These locations were then augmented by the translation of linear location data into geographical coordinates. This translation was achieved through computations to place markers at specified intervals, utilizing JavaScript-HTML with Google API for precise spatial analysis and to reflect the real-world infrastructure curvature. Additionally, production volumes were allocated to each of these farms using macro-level aggregate data. The integration of this information (locations and volumes) facilitates a better understanding and optimization of the supply chain. This methodology enables the tackling of the three main sources of variation in the perishable food supply chain, including Farm Location Variability, Annual Production Volume Variability, and Demand Distribution Variability. Despite the synthetic generation of data, this methodology accurately mirrors the stochastic patterns, real-life complexities, and realities of the perishable food supply chain in Jordan, allowing for variations. This paper showcases an effective solution to overcome data limitations, offering necessary insights into the potential for strategic supply chain optimization and strategic decision-making in perishable food sectors.
Homicide's impact transcends the individual, shattering social trust, damaging community cohesion, and casting a shadow of fear over entire nations, demanding urgent efforts to address root causes and prevent further loss. Although Thailand's homicide rates are comparatively lower than those of certain neighboring Southeast Asian countries, this issue continues to pose a significant threat to public safety within the nation. To gain deeper insights into their spatial distribution and potential clustering, this research employed Geographic Information Systems (GIS) to conduct a spatial analysis of homicide report cases across Thailand from 2019 to 2021. Utilizing national data sourced from police reports, the study investigated the existence of spatial autocorrelation, indicating whether homicide occurrences are statistically dependent on their geographic location. This analysis shed light on whether homicides are randomly distributed or exhibit clustering tendencies, suggesting potential underlying spatial factors at play. Furthermore, the research employed hotspot mapping techniques to identify statistically significant clusters of homicide cases. These hotspots revealed areas with an elevated concentration of homicides, providing crucial information for targeted prevention strategies. By pinpointing high-risk areas, law enforcement agencies can allocate resources more effectively, focusing on factors specific to each hotspot area. This research leverages GIS to unlock new insights into Thailand's homicide patterns, leading to a more comprehensive understanding. The identification of spatial autocorrelation and hotspots provided valuable insights for policymakers and law enforcement agencies to develop targeted prevention strategies and enhance public safety across the country.
Industrial Hygiene has been a science that is concerned with the anticipation, recognition up to the control of the factors or stressors that can be found in different workplaces that may have adverse effect on the workers. A part of the control aspect is to be able to plan management strategies in the cases of accidents that occur in hazardous workplaces. Geographic Information System (GIS) is an emerging tool that stores, evaluate, and represents data on a spatial scale for mitigation and emergency response procedures. This research aimed to identify the location of the different hazardous workplaces in the city of Manila using GIS. Maps representing locations of workplaces and establishments that are exposed to flammable substances, dock works, hand-driven tools, and biological hazards were produced and analyzed as to their distribution and proximity from the nearest hospital. Outputs of this study can help give managers and workplace owners an idea on the planning and control for their establishments in the instance of accidents.
Typhoon is the most destructive marine disastrous weather system. The strong winds and waves, heavy rainfall and poor visibility caused by typhoon have a significant impact on marine activities. With the progress of science and technology, the theory and technical means of retrieving typhoon by satellite remote sensing have made great progress. This paper collects and compiles the data and products of multi-source full polarization microwave radiometer, studies and establishes the theory of typhoon retrieval by satellite remote sensing under high sea conditions (strong wind, rainstorm, huge waves, low visibility, etc.), verifies the algorithm, and analyses the retrieval accuracy of the algorithm under different wind speeds and different precipitation; The visual demonstration software of typhoon retrieval by satellite remote sensing is developed, which realizes the historical analysis and real-time monitoring of typhoons in the global sea area, and can provide new theoretical and technical means for typhoon ensuring
Human society is developing rapidly and the demand for the earth's resources is increasing. At the same time, global environmental quality issues are becoming increasingly serious. China is a vast country with abundant water resources, and sustainable development strategy has become a national strategic goal. Therefore, in the process of geological exploration, we need to use advanced digital technology to establish a complete, efficient and sustainable development path. The so-called Geographic Information System (GIS) is a system formed under the role of information technology, computer networks, professional theoretical knowledge in surveying and mapping, referred to as GIS, which can be applied to achieve efficient collection, processing and storage of geospatial information, etc. The potential application value is large and has good market application prospects. At the same time, GIS has the characteristics of high mapping efficiency, high accuracy and good applicability in practice, and can provide professional support for basin geomorphology investigation and research. Therefore, in the process of basin geomorphology survey and research, the efficient use of GIS should be paid attention to. Based on the theoretical research of spatial information technology (DSP) and GIS technology, this paper first gives a brief introduction to digital GIS, then analyses and discusses relevant algorithms and technologies, proposes a series of software system design and optimisation and GIS networked management platform construction, realises data collection and storage of simulated topographic landscapes, and finally discusses and studies the digital GIS-based The specific application of digital GIS-based geological landscape survey is discussed and studied.
Understanding what people need during disasters and how many people are exposed to disasters are critical in effective disaster management especially in urban megacities where high population density poses greater disaster risk. More importantly, analyzing how disaster needs and population vary through time is becoming as critical for modelling population exposure to hazards, which can aid disaster risk estimation and mitigation. Although traditional data collection methods such as remote sensing data are available, it is still a challenge to estimate exposure and analyze dynamic changes in a high temporal resolution. This paper investigates the use of spatio-temporal big data as an input in population exposure assessment across multiple disaster scenarios in Tokyo. Specifically, we demonstrate this through case studies on natural disasters typhoon and earthquake, as well as abnormal scenarios such as heavy snowfall in the city. We utilize geoinformation (e.g., GPS traces) from mobile phone users in Japan, extract trajectory and search query data, and analyze population changes and trends at hourly temporal resolution during disasters. Moreover, we compare the intensity of changes with normal times to delineate extent of exposure. In addition, we collect geo-tagged social media data from Twitter in the same location to analyze hourly trend of tweet volume. By utilizing this method, we are able to get better understanding of the intensity and dynamic trend of the population affected by the disaster at a high temporal resolution (i.e., hourly) which can aid population exposure assessment for disaster risk management.
Background: Malnutrition is a major public health concern worldwide; a recent study found that 22% of children under the age of five were stunted worldwide in 2020. Stunting in Rwanda has decreased dramatically over the last 15 years, from 51% in 2005 to 33% by 2020. However, because few geospatial studies have been conducted, geographical survey data analysis is required to effectively focus stakeholders' efforts in response to successful stewardship of health programs in the eradication of all types of malnutrition. The study's goal is to map the prevalence distribution of stunted children under the age of five, make Projections, and provide exceedance probability maps for each stunting at the 30% threshold value, as well as identify risk factors associated with stunting in Rwanda. Methods: This study makes use of Rwandan Demographic and Health Surveys (R-DHS)2019/2020. To obtain the marginal posterior distribution of stunting prevalence at each location in Rwanda, the Bayesian model was developed using an integrated nested Laplace approximation technique. The risk factors for stunting were identified using multivariate logistic regression. Results: This research finds the prevalence of stunting in 500 clusters. Kigali city clusters had the lowest prevalence, ranging from 0% to 25%. The Western Province, specifically the Congo Nile Divide, has the highest rate where some clusters have more than 60%. The 30% probability threshold enables the identification of Rwandan areas and communities most vulnerable to stunting. In Rwanda, regional disparities in childhood stunting are significant. There is statistically significant child, maternal, and socio-demographic characteristics with p-values less than 5% in a 95% confidence interval (CI). Among the risk factors are the baby's birth weight (OR:2.18, CI: [1.613-2.95], intestinal parasites (OR:1.544, 95% CI: [1.253-1.903], and the baby's age (OR:1.095, 95% CI: [1.024-1.171]. Province and altitude have ORs of 1.088 and 1.575, respectively, in socio-demographic factors. Conclusions: Finally, geospatial studies should be conducted to identify locations that require more attention in disease and epidemic control and multivariate logistic regression to identify risk factors. Therefore, more attention of malnutrition eradication should be paid to Rwanda's Western and Northern Provinces. The findings of this study may be useful to program managers and decision makers working to reduce the burden of stunting.
The aim of this work is to analyse greenhouse gases (GHGs), their emissions in the agricultural sector, and the possibility of monitoring them through remote sensing (RS), and data-driven solutions. In this paper using GIS software analysed the GHG emission reports of 43 member countries that are regularly submitted to the Secretariat of the United Nations Framework Convention on Climate Change. The analysis highlighted Sweden's leadership in the energy sector, where up to 66% of its electricity comes from renewable sources. New Zealand also stood out in the context of all countries, with very high methane (CH4) emissions due to the country's large livestock population and poor emission controls in the agricultural sector. The article also provided an overview of the satellites currently available on the market for monitoring GHG emissions and a partial analysis of their characteristics. Future work is planned to further investigate the applicability of satellites for monitoring GHG emissions, to provide a detailed analysis of the characteristics of public, commercial, and hybrid satellites, to carry out practical applications of satellite data for the determination of agricultural emissions, and to develop a methodology for continuous emissions monitoring.
The importance of elderly care institutions in city public service planning has increased due to rapid population aging. This study used the point-of-interest coordinates of elderly care institutions for kernel density,hotspot analyses and Getis–Ord Gi* Analysis. The spatial distribution of elderly care institutions in Meizhou, China, was analyzed, and related hotspots were identified. Through the analysis of computer software, it can be indicated that: (i) the distribution of elderly care institutions in Meizhou City shows agglomeration characteristics, and the overall distribution shows the characteristics along residential areas, traffic arteries and commercial centers. (ii) The hot spots of nursing institutions in Meizhou City are prominent, forming two significant hot spots and almost no cold spots. The findings of this study can be used as a reference by those involved in optimizing the spatial layout of elderly care institutions in cities.
Environmental pollution is a major problem facing modern society and is closely related to human survival activities. Traditional air quality monitoring models mainly explain the mechanisms and sources of pollutant dispersion in the atmosphere, and it is difficult to reveal the intuitive process and results of air pollution dispersion. In order to solve this problem, this paper combines GIS technology and environmental models applied to air pollution dispersion simulation, and designs an intelligent system that can be used for dynamic monitoring of atmospheric environmental quality. On the basis of systematic analysis of a large amount of spatial information, the dynamic simulation model and spatial analysis model of urban air environmental quality are established, which realises the visualisation of urban air quality prediction and dynamic simulation and helps The system not only makes use of the powerful GIS technology, but also provides a good living environment. The system not only brings into play the powerful spatial visualisation management and analysis functions of GIS, but also provides an efficient decision support platform for the simulation, prediction and analysis of environmental models, adaptive collection of air environment monitoring data and real-time, reliable and efficient transmission, providing good technical support and reliable data support for the monitoring and management of urban air environment quality. The experiment shows that the combination of GIS can complete the simulation of air pollution dispersion, and complete the visualization and analysis of simulation results, providing a set of spatial decision support simulation platform of air pollution dispersion with friendly interface and simple operation for relevant environmental management departments.
As a meeting point of multiple transportation modes such as land, water, and pipeline, the wharf has always been an important transportation hub in my country. With the development of the economy, the excessive use of wharves has caused many hidden dangers in quality and safety. However the traditional quality evaluation methods are complicated and cumbersome, and the results are not intuitive. Therefore, by analyzing the demand for wharf quality evaluation services and combining it with web geographic information systems, this article uses the Yangtze River Delta and Lianyungang wharf's data as research data, uses Angular framework, PostgreSQL, Leaflet, and other technologies to build a wharf quality evaluation service platform based on WebGIS that is compatible with all major browsers. This platform realizes the functions of spatial information visualization, viewing of attribute information, multi-mode data query, calculation of reinforced concrete strength, and concrete corrosion potential evaluation services. The platform can not only visually display the quality level of the wharf, but also provide quality evaluation services through dynamic interaction, which is of great significance to the management and evaluation of wharf quality information.
The use of indoor positioning technology is crucial for the security of firefighters. For ordinary fire rescue, this paper offers a quick introduction of several indoor positioning technology methods. We show that the firefighting and rescue operations' technical bottleneck is indoor positioning technology. The demand for indoor positioning technology among firefighters is analyzed and summarized in this article.
The Internet of Vehicles (IoV) is a use case of the Internet of Things (IoT), where the urban vehicle fleet forms a worldwide network. Quality of Service (QoS) optimization in IoV is a challenging task, since vehicles are in continuous movement, which causes the network to be unstable and in continuous topology change, this creates a hostile environment with many variables that should be taken into consideration. Location- based routing protocols have proven to be the most performing in such an environment, they offer high packet delivery with low delay, but they also encounter some issues such as routing loops and bandwidth overload. In the present paper, we study existing solutions and propose a novel optimal location based routing algorithm named Minimum Hops Routing (MHR). For experimental study, we compared the proposed MHR algorithm with existing solutions from the state of the art. The experimental results show that the proposed MHR solution is more efficient than existing solutions in terms of reliability enhancement and delay.
Glacier is the product of climate, which is highly sensitive to global climate change, and is the most rapid and significant response to environmental change and climate change. Therefore, in the context of global warming, the study of glacier change is of great significance to the global climate change, global warming and the sustainable development of human society. Different GPS monitoring is the main method for the study of glacier movement. GPS technology was proposed by European and American countries in the 1990s, and its application in industrial and agricultural production has been quite mature, and it has become an important information technology means in the world. It not only promotes industrial automation but also provides data processing functions. Based on the research of GPS positioning technology, this paper establishes a new type of precise geospatial measurement system. The system is built with visualization and data processing as the core content; it has the characteristics of automatic identification and three-dimensional coordinate transformation. It can realize the effective combination of traditional map and satellite remote sensing network, and this can complete elevation control and highly precise positioning. It can also display digital information in real time when the precision requirement is lower. The results of this paper are instructive for data processing using high-density based GPS dynamic single-point localization measurement methods for ice surface topographic measurements.
