Building extraction from remote sensing imagery using SegFormer with post-processing optimization

Accurate extraction of buildings using high spatial resolution imagery is essential to a wide range of urban applications. However, it is difficult to extract semantic features from a variety of complex scenes (e.g., suburban, urban and urban village areas) because various complex man-made objects usually appear heterogeneous with large intra-class and low inter-class variations. The automatic extraction of buildings is thus extremely challenging. The fully convolutional neural networks (FCNs) developed in recent years have performed well in the extraction of urban man-made objects due to their ability to learn state-of-the-art features and to label pixels end-to-end. One of the most successful FCNs used in building extraction is U-net. However, the commonly used skip connection and feature fusion refinement modules in U-net often ignore the problem of feature selection, and the ability to extract smaller buildings and refine building boundaries needs to be improved. In this paper, we propose a trainable chain fully convolutional neural network (CFCN), which fuses high spatial resolution unmanned aerial vehicle (UAV) images and the digital surface model (DSM) for building extraction. Multilevel features are obtained from the fusion data, and an improved U-net is used for the coarse extraction of the building. To solve the problem of incomplete extraction of building boundaries, a U-net network is introduced by chain, which is used for the introduction of a coarse building boundary constraint, hole filling, and "speckle" removal. Typical areas such as suburban, urban, and urban villages were selected for building extraction experiments. The results show that the CFCN achieved recall of 98.67%, 98.62%, and 99.52% and intersection over union (IoU) of 96.23%, 96.43%, and 95.76% in suburban, urban, and urban village areas, respectively. Considering the IoU in conjunction with the CFCN and U-net resulted in improvements of 6.61%, 5.31%, and 6.45% in suburban, urban, and urban village areas, respectively. The proposed method can extract buildings with higher accuracy and with clearer and more complete boundaries.

The application of Geographic Information Systems (GIS) in the work of government bodies, economic subjects and units of regional and local self-government presents the real situation today in every European country, as well as in its regions. The organisation of collecting, storing, distribution and usage of spatial data, in accordance with the principals of spatial data infrastructure (SDI) presents the next (higher) level in the application of spatial data and GIS. Rapid development of GIS application in the last decade highlights two problems in the efficient usage of GIS technologies, entailing the usage of spatial data in general. The first problem is the lack of the concept in organising the spatial data that has been solved basically through the development of IPP concept. The other problem lies in the fact that the subjects and the users do not have enough knowledge of GIS and IPP. Therefore, special attention has been paid within the frame of INSPIRE project to the development of education concept for the subjects and users of IPP, in order to provide the possibility for them to use the advantages offered by the concept of enhanced technologies and IPP itself. The paper presents the current situation in the education referring to GIS and IPP, and describes the activities carried out for the purpose of creating the environment that would provide quicker and more efficient implementation of National SDI (NSDI) based on better information given to the subjects and users.

For many years, modeling and GIS applications have been used as separate systems for planning, design and operation of water distribution systems. As digital information becomes increasingly available, the potential benefits of an integrated model and GIS system become more obvious. These benefits of integration and resulting automation include: better service to internal and external customers, improved work processes and higher productivity, reduced data entry errors and duplication, improved data security and reliability, and improved emergency response. Las Vegas Valley Water District (LVVWD) is developing an integrated system that includes a hydraulic and water quality model. Geographic Information System (GIS), Supervisory Control and Data Acquisition (SCADA), and other database systems. Model data is dynamically linked to the GIS information with unique identification features. The model utilizes GIS and other linked data to compute hydraulic and water quality results for master planning, operational planning, facility outage scenarios, and other what-if scenarios. The integrated system enables the data process to be automated to the maximum extent possible so that manpower and time is saved. This paper describes the significance of such integration and demonstrates its value with example applications. The examples encompass master planning, water quality planning and operational planning for rapid-growing communities. This paper was presented at the 8th Annual Water Distribution Systems Analysis Symposium which was held with the generous support of Awwa Research Foundation (AwwaRF).

Geographic Information Systems (GIS) is one of the modern methods that contributed extensively to facilitating the tasks and performance of planners during decisions related to the development or analysis of a particular problem with a spatial dimension, and this technique has been used in many different scientific fields such as the study of the distribution of services, analysis of land uses, or the division of transportation. The paper aims to highlight the benefits of Geographic Information Systems (GIS) in tourism planning, identify the basic stages of Geographic Information Systems (GIS) applications in tourism planning, and to identify the importance of using Geographic Information Systems (GIS) in the tourism planning of archaeological sites in Fatimid Cairo. This study was based on a descriptive analytical approach and the field survey method was applied in the sample method to reach conclusions that serve the objectives of the study. This study was finalized by several results, most notably: The need to use Geographic Information Systems (GIS) in documenting the monuments by transferring all images, data and written inscriptions on them and illustrating them through pictures and videos, to form a database to be available to the tourist, researcher and planner. The planning of archaeological sites using Geographic Information Systems (GIS) in the study area (Fatimid Cairo), saves time and effort by relying on aerial and space images and topographic maps with high accuracy and conducting analysis and reaching accurate results in a very short period, compared to the time required using traditional methods.

The application of Geographic Information System (GIS) technology in education is becoming increasingly vital for enhancing digital literacy and spatial analysis skills. Rahmatullah Lempake Islamic Boarding School in Samarinda is one of the educational institutions with significant potential to utilize this technology in supporting the management of its environment and local resources. The initiative aims to provide students with training in GIS, enhancing their knowledge and skills in spatial data management and geographic analysis. The training employed a combination of lectures, discussions, and practical sessions using GIS software. It was attended by students aged 12 to 15 and covered foundational topics, such as an introduction to GIS, mapping applications, and data analysis techniques. The results of this training demonstrated a notable improvement in the students' understanding and basic GIS skills, as well as increased awareness of the technology's potential applications in their local environment. The integration of GIS into the educational process offers significant benefits, particularly in enhancing students' technological literacy and spatial analysis capabilities, which can be applied to environmental management activities within the Islamic boarding school. In conclusion, the GIS training at Rahmatullah Lempake Islamic Boarding School has effectively developed the students' foundational skills in technology and geographic analysis, offering practical benefits for future resource management and environmental planning endeavors.

Building extraction has attracted much attentions for decades as a prerequisite for many applications and is still a challenging topic in the field of photogrammetry and remote sensing. Due to the lack of spectral information, massive data processing, and approach universality, building extraction from point clouds is still a thorny and challenging problem. In this paper, a novel deep-learning-based framework is proposed for building extraction from point cloud data. Specifically, first, a sample generation method is proposed to split the raw preprocessed multi-spectral light detection and ranging (LiDAR) data into numerous samples, which are directly fed into convolutional neural networks and completely cover the original inputs. Then, a graph geometric moments (GGM) convolution is proposed to encode the local geometric structure of point sets. In addition, a hierarchical architecture equipped with GGM convolution, called GGM convolutional neural networks, is proposed to train and recognize building points. Finally, the test scenes with varying sizes can be fed into the framework and obtain a point-wise extraction result. We evaluate the proposed framework and methods on the airborne multi-spectral LiDAR point clouds collected by an Optech Titan system. Compared with previous state-of-the-art networks, which are designed for point cloud segmentation, our method achieves the best performance with a correctness of 95.1%, a completeness of 93.7%, an F-measure of 94.4%, and an intersection over union (IoU) of 89.5% on two test areas. The experimental results confirm the effectiveness and efficiency of the proposed framework and methods.

Deep convolutional neural networks have promoted significant progress in building extraction from high-resolution remote sensing imagery. Although most of such work focuses on modifying existing image segmentation networks in computer vision, we propose a new network in this paper, Deep Encoding Network (DE-Net), that is designed for the very problem based on many lately introduced techniques in image segmentation. Four modules are used to construct DE-Net: the inception-style downsampling modules combining a striding convolution layer and a max-pooling layer, the encoding modules comprising six linear residual blocks with a scaled exponential linear unit (SELU) activation function, the compressing modules reducing the feature channels, and a densely upsampling module that enables the network to encode spatial information inside feature maps. Thus, DE-Net achieves state-of-the-art performance on the WHU Building Dataset in recall, F1-Score, and intersection over union (IoU) metrics without pre-training. It also outperformed several segmentation networks in our self-built Suzhou Satellite Building Dataset. The experimental results validate the effectiveness of DE-Net on building extraction from aerial imagery and satellite imagery. It also suggests that given enough training data, designing and training a network from scratch may excel fine-tuning models pre-trained on datasets unrelated to building extraction.

With the development of deep learning technology, an enormous number of convolutional neural network (CNN) models have been proposed to address the challenging building extraction task from very high-resolution (VHR) remote sensing images. However, searching for better CNN architectures is time-consuming, and the robustness of a new CNN model cannot be guaranteed. In this paper, an improved boundary-aware perceptual (BP) loss is proposed to enhance the building extraction ability of CNN models. The proposed BP loss consists of a loss network and transfer loss functions. The usage of the boundary-aware perceptual loss has two stages. In the training stage, the loss network learns the structural information from circularly transferring between the building mask and the corresponding building boundary. In the refining stage, the learned structural information is embedded into the building extraction models via the transfer loss functions without additional parameters or postprocessing. We verify the effectiveness and efficiency of the proposed BP loss both on the challenging WHU aerial dataset and the INRIA dataset. Substantial performance improvements are observed within two representative CNN architectures: PSPNet and UNet, which are widely used on pixel-wise labelling tasks. With BP loss, UNet with ResNet101 achieves 90.78% and 76.62% on IoU (intersection over union) scores on the WHU aerial dataset and the INRIA dataset, respectively, which are 1.47% and 1.04% higher than those simply trained with the cross-entropy loss function. Additionally, similar improvements (0.64% on the WHU aerial dataset and 1.69% on the INRIA dataset) are also observed on PSPNet, which strongly supports the robustness of the proposed BP loss.

The role of environmental management is extremely important in human life because of its contribution to environmental protection and strategies. In modern society, nevertheless, facing the challenges in land, water, and air quality management in reality is unavoidable. Understanding the information on variability and features of environmental issues is necessary to support decision-makers in establishing environmental management planning. Spatial information and data should be considered as basic initial knowledge of environmental management. Among various technologies, Geographic Information Systems (GIS) is known as one of the most popular tools to store, analyze, and visualize data relating to geographically referenced information which can mitigate these challenges more effectively and accurately. The main purpose of this paper is to emphasize how GIS applications in environmental management, discuss the challenges of GIS technology in environmental management, and provide recommendations for future research directions. Many articles over the last decades were reviewed to highlight the application of GIS in environmental management. These results indicated GIS applications support land, water, air quality, and waste management. Moreover, it is a useful tool to monitor and assess environmental quality and in environmental planning and decision-making. For the limitation of GIS application, it is obvious that accuracy assessment and sensitivity analysis should be noticed and assessed although GIS application may strongly support decision-making in environmental management. Emerging GIS technologies and tendencies could help address these challenges. In future, GIS should be integrated along with new technologies and different decision-making algorithms in further steps for developing and supporting spatial decision-support systems more efficiently.

The application of Geographic Information Systems (GIS) in green project management has gained significant attention due to its potential to support sustainable construction. This review explored the application of GIS in green project management practices, focusing on the challenges and opportunities involved to propose a future research agenda for advancing the use of GIS. A systematic literature review of publications from 2000 to 2024 was conducted, focusing on the role of GIS in promoting sustainable construction and energy-efficient project management. The review used specific keywords, including "Geographic Information Systems (GIS)," "Green Projects," "Energy Optimization," "Sustainability Reporting," and "Waste Management." These terms were combined with Boolean operators (AND, OR) to refine and enhance search results, ensuring the inclusion of relevant studies related to GIS applications in sustainability and resource management within construction projects. The findings revealed that GIS enhances energy optimization by analyzing consumption patterns, guiding interventions like retrofitting and renewable energy integration. In water and waste management, GIS helps detect inefficiencies and optimize collection routes, although challenges remain in acquiring high-resolution data. GIS also supports real-time monitoring, providing continuous insights into energy and resource usage, though integration with advanced systems and infrastructure is essential. In sustainability reporting, GIS aids in visualizing indicators like carbon footprints and conducting Lifecycle Assessments (LCA), providing an in-depth evaluation of environmental impacts throughout a project’s lifecycle. This review recommends integrating GIS with real-time monitoring systems and standardized sustainability metrics to enhance its effectiveness in optimizing energy use, waste management, and sustainability reporting in green projects.

The article presents the results of the application of geographic information systems for environmental auditing, which allows you to monitor the ecological situation on the objects, food and chemical industries and to ensure environmental safety of natural and industrial areas. The authors propose a mathematical model that performs monitoring of environmental objects. The computational experiment is performed in Matlab for classifying the soil measurement conditions. The results of the analysis and the electronic cartograms constructed on their basis are used to predict the location of the points exceeding the parasitological indices in the soil. The assessment of the condition of the soil in different districts of the Voronezh city is made with the help of the created neural networks. The results are compared with the same indicators, measured in 2017. Data is taken from the information Bulletin «Evaluation of the influence of environmental factors on the health of the population of the Voronezh region on indicators of socio-hygienic monitoring». Information was prepared by the Office of Rospotrebnadzor in the Voronezh region. A study cited in the article was conducted at the Department of information and control systems of the Voronezh state University of engineering technologies. Geographic information system was built with the help of MapInfo software product. It can be used to simulate the environmental situation and to inform about its current state with the help of cartography. Thus, the geoinformation research method reviewed in this article allows predicting the emission of harmful substances into the environment and it can be used by regulatory agencies of the state authorities to control the environmental state of the territory.

Abstract—To date, the knowledge regarding the potential uses of Geographic Information Systems (GIS) in Supply Chain Management (SCM) and how GIS applications can deliver business benefits are limited. To explore the current uses of GIS in SCM, a systematic literature review (SLR) was conducted. Three databases (Web of Science, ProQuest, and Business Source Premier) were explored using the PreferredReporting Items for Systematic Reviews and Meta-Analyses(PRISMA) methodology. A total of 79 papers were included in the literature review. The combined findings showed that the GIS applications in SCM were a) network/transportation analysis/ routing algorithms (53 papers), b) Location-Allocation Site Search/ Selection (using multiple-criteria decision analysis [MCDA]) (45 papers), c) spatial analysis(people or land) (34 papers), d) connection with optimisation tools (32 papers) and e) visualisation/ monitoring or BuildingInformation Modelling (BIM) (8 papers). The industries and related Supply Chain (SC) domains that used GIS were a)Biomass Biofuel/Wood (33), b) Logistics (22), c)Humanitarian/ Emergency/ Healthcare (10), d) Food/ Agro-Industry (5), e) Petroleum/ Coal/ Shale Gas (3), f) FaecalSludge (2), g) Recycle & product footprint (2) and h)Construction (2). The research case studies took place in 26 countries, and the most prominent GIS software provider wasEsri’s ArcGIS. The findings of the first literature review ofthe GIS applications in SCM showed that many businessindustries have not utilised the GIS capabilities, such asvisualisation, spatial analysis, optimisation, and site research. However, GIS can significantly contribute to SCM decisionmaking by providing answers to cost minimisation, supplierselection, facility location, SC network configuration andasset management. Keywords— Supply Chain Management, Logistics, SystematicLiterature Review, GIS

Accurate building extraction from very high-resolution (VHR) remote sensing images plays an important role in urban dynamic monitoring, planning, and management. However, it is still a challenging task to achieve building extraction with high accuracy and integrity due to diverse building appearances and more complex ground background in VHR remote sensing images. Recently, Unity Networking (UNet) has been proven to be capable of feature extraction and semantic segmentation of remote sensing images. However, UNet cannot achieve sufficient multi-scale and multi-level features with larger receptive fields. To address these problems, an improved network based on UNet structure (Refine-UNet) is proposed for extracting buildings from the VHR images. The proposed Refine-UNet mainly consists of encoder module, decoder module, and refine skip connection scheme. The refine skip connection scheme is composed of an atrous spatial convolutional pyramid pooling (ASPP) module and several improved depthwise separable convolution (IDSC) modules. Experimental results on the Jilin-1 VHR datasets with a spatial resolution of 0.75 m demonstrate that compared with UNet, PSPNet, DeepLabV3+, and SegNet, the proposed Refine-UNet can obtain more accurate building extraction results and achieve the best Precision of 95.1% and intersection over union (IoU) of 87.0%, respectively, indicating the great practical potential.

The role of Geographic Information System (GIS) technology in tourism analysis and site selection is well established. The tourism industry have widely adopted information technology (IT) to reduce costs, enhance operational efficiency, and most importantly to improve service quality and customer experience. The paper explores the applications of Geographic Information System (GIS) that present travel patterns related data obtained from Petra city in Jordan. The paper demonstrates that Geographic Information System (GIS) is increasingly becoming critical for the competitive operations of the tourism sites as well as hospitality organizations in Jordan. As proposed by the system used in this study the paper provides various utilities and a proposal for the application of Geographic Information System (GIS) in marketing of tourist sites in Jordan. Furthermore, the paper will highlight the barriers the researchers encounter in applying Geographic Information System (GIS) in marketing of tourist sites in Petra city and Jordan in general.

This systematic review aims to explore the application of geographic information system (GIS)-based multi-criteria decision analysis (MCDA) techniques in addressing diverse environmental and socio-economic challenges in Bangladesh. The study fills the gap in consolidated evidence regarding the effectiveness and sectoral orientation, as well as the methodological variety of GIS-MCDA applications, especially in the context of developing countries. A ten-year time frame has been chosen for collecting literature using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework from various scholarly databases. Findings reveal that GIS-MCDA techniques have gained widespread acceptance across various sectors, including disaster management, water resource management, energy, waste management, urban planning, tourism, agriculture, healthcare, and industry. Analytical hierarchy process (AHP) has been found as most popular MCDA technique to utilize in various sectors efficiently. However, the findings also note some essential knowledge gaps in existing literature such as the excessive focus on some particular disaster (e.g., floods), minimal implementation in less-researched areas and industries, and insufficient field validation or stakeholder engagement in most studies. The review highlights the operational significance of GIS-MCDA and its ability to aid spatially informed and evidence-based decision-making. It also recommends the necessity of wider thematic coverage, a hybrid approach of methods and participatory structure to develop the practical application value of such tools. Combining the sector-specific knowledge with the spatiotemporal trends, the paper provides an original contribution by conducting the first comprehensive evaluation of GIS-MCDA implementation in Bangladesh, which can guide policymakers, planners and practitioners toward more participatory, sustainable, and resilient planning.