Search

Automatic classification of buildings within port areas from synthetic aperture radar (SAR) images is crucial for effective port monitoring and planning. Yet, the unique challenges of SAR imaging, such as side-looking geometry, multi-bouncing scattering, and the compact arrangement of structures, often lead to incomplete building structures and blurred boundaries in classification results. To address these issues, this paper introduces SPformer, an efficient and lightweight pyramid transformer model tailored for semantic segmentation. The SPformer utilizes a pyramid transformer encoder with spatially separable self-attention (SSSA) to refine both local and global spatial information and to process multi-scale features, enhancing the accuracy of building structure delineation. It also integrates a lightweight all multi-layer perceptron (ALL-MLP) decoder to consolidate multi-scale information across various depths and attention scopes, refining detail processing. Experimental results on the Gaofen-3 (GF-3) 1 m port building classification dataset demonstrate the effectiveness of SPformer, achieving competitive performance compared to state-of-the-art models, with mean intersection over union (mIoU) and mean F1-score (mF1) reaching 77.14% and 87.04%, respectively, while maintaining a compact model size and lower computational requirements. Experiments conducted on the entire scene of SAR images covering port area also show the good capabilities of the proposed method. [ABSTRACT FROM AUTHOR]

The urbanization rate of India is 35.9 % by 2022 reports. In this 45.23 % of urbanization is happening in Maharashtra and it is the third most urbanized state of India after Tamil Nadu and Kerala. In metropolitan areas, the classification of land cover from satellite images has been the focus of remote sensing over the years. Due to complex architecture and a lack of labeled data, classifying buildings in metropolitan areas from very high resolution (VHR) satellite imagery is challenging. Traditional approaches for building classification include hand-crafted features and transfer learning methods. These methods often struggle with the variability in building shapes, orientation, and viewpoint, leading to low accuracy in densely populated urban areas and limited performance when dealing with high- resolution satellite images. A deep-learning based approach for semantic segmentation using U-Net with a backbone of ResNet-34 is proposed for building classification. Urban area Dataset with Images of 0.5 m resolution is prepared from SASPlanet. One hot Encoding is applied for classifying buildings. U-Net is trained with encoded data. The proposed model is evaluated on the Indian dataset, specifically, the urban areas of Nashik, Maharashtra state and the accuracy obtained for the classification dataset is 60 % and the accuracy of the building detection is about 85 %. Change detection is calculated from bi-temporal images. The GIS maps are updated to detect changes in buildings, represented by different colors to distinguish newly constructed buildings, existing structures and demolished ones.

Automatic classification of buildings within port areas from synthetic aperture radar (SAR) images is crucial for effective port monitoring and planning. Yet, the unique challenges of SAR imaging, such as side-looking geometry, multi-bouncing scattering, and the compact arrangement of structures, often lead to incomplete building structures and blurred boundaries in classification results. To address these issues, this paper introduces SPformer, an efficient and lightweight pyramid transformer model tailored for semantic segmentation. The SPformer utilizes a pyramid transformer encoder with spatially separable self-attention (SSSA) to refine both local and global spatial information and to process multi-scale features, enhancing the accuracy of building structure delineation. It also integrates a lightweight all multi-layer perceptron (ALL-MLP) decoder to consolidate multi-scale information across various depths and attention scopes, refining detail processing. Experimental results on the Gaofen-3 (GF-3) 1 m port building classification dataset demonstrate the effectiveness of SPformer, achieving competitive performance compared to state-of-the-art models, with mean intersection over union (mIoU) and mean F1-score (mF1) reaching 77.14% and 87.04%, respectively, while maintaining a compact model size and lower computational requirements. Experiments conducted on the entire scene of SAR images covering port area also show the good capabilities of the proposed method.

The urbanization rate of India is 35.9 % by 2022 reports. In this 45.23 % of urbanization is happening in Maharashtra and it is the third most urbanized state of India after Tamil Nadu and Kerala. In metropolitan areas, the classification of land cover from satellite images has been the focus of remote sensing over the years. Due to complex architecture and a lack of labeled data, classifying buildings in metropolitan areas from very high resolution (VHR) satellite imagery is challenging. Traditional approaches for building classification include hand-crafted features and transfer learning methods. These methods often struggle with the variability in building shapes, orientation, and viewpoint, leading to low accuracy in densely populated urban areas and limited performance when dealing with high- resolution satellite images. A deep-learning based approach for semantic segmentation using U-Net with a backbone of ResNet-34 is proposed for building classification. Urban area Dataset with Images of 0.5 m resolution is prepared from SASPlanet. One hot Encoding is applied for classifying buildings. U-Net is trained with encoded data. The proposed model is evaluated on the Indian dataset, specifically, the urban areas of Nashik, Maharashtra state and the accuracy obtained for the classification dataset is 60 % and the accuracy of the building detection is about 85 %. Change detection is calculated from bi-temporal images. The GIS maps are updated to detect changes in buildings, represented by different colors to distinguish newly constructed buildings, existing structures and demolished ones. [ABSTRACT FROM AUTHOR]

Recent improvements in remote sensing technologies have boosted building detection techniques from rough classifications using moderate resolution imagery to precise extraction from high-resolution imagery. Shadows frequently emerge in high-resolution urban images. To exploit shadow information, we developed a novel building detection and classification algorithm for images of urban areas with large-size shadows, employing only the visible spectral bands to determine the height levels of buildings. The proposed method, building general-classified by height (BGCH), calculates shadow orientation, detects buildings using seed-blocks, and classifies the buildings into different height groups. Our proposed approach was tested on complex urban scenes from Toronto and Beijing. The experimental results illustrate that our proposed method accurately and efficiently detects and classifies buildings by their height levels; the building detection rate exceeded 95%. The precision of classification by height levels was over 90%. This novel building-height-level detection method provides rich information at low cost and is suitable for further city scene analysis, flood disaster risk assessment, population estimation, and building change detection applications. [ABSTRACT FROM AUTHOR]

Remote sensing images are important data sources for land cover mapping. As one of the most important artificial features in remote sensing images, buildings play a critical role in many applications, such as population estimation and urban planning. Classifying buildings quickly and accurately ensures the reliability of the above applications. It is known that the classification accuracy of buildings (usually indicated by a comprehensive index called F1) is greatly affected by image quality. However, how image quality affects building classification accuracy is still unclear. In this study, Boltzmann entropy (an index considering both compositional and configurational information, simply called BE) is employed to describe image quality, and the potential relationships between BE and F1 are explored based on images from two open-source building datasets (i.e., the WHU and Inria datasets) in three cities (i.e., Christchurch, Chicago and Austin). Experimental results show that (1) F1 fluctuates greatly in images where building proportions are small (especially in images with building proportions smaller than 1%) and (2) BE has a negative relationship with F1 (i.e., when BE becomes larger, F1 tends to become smaller). The negative relationships are confirmed using Spearman correlation coefficients (SCCs) and various confidence intervals via bootstrapping (i.e., a nonparametric statistical method). Such discoveries are helpful in deepening our understanding of how image quality affects building classification accuracy. [ABSTRACT FROM AUTHOR]

Regional seismic risk assessment is necessary for designing effective seismic risk mitigation measures. In general, such risk assessment studies consist of three components: hazard, vulnerability, and exposure modelling. This paper lays the foundations for regional seismic risk assessment of the residential building stock in Serbia and addresses each of the three seismic risk assessment components, either by reviewing the existing or proposing novel models. First, a review of seismic hazard models and seismic design codes used in Serbia in the past 70 years was presented. Next, an overview of Serbia's population metrics and historical development of Serbian's residential building stock was presented to provide the context for the exposure model. Furthermore, the paper proposed a novel building classification for Serbia's residential building stock, which is based on the existing building taxonomies, but it has been adapted to account for the local building characteristics. Building damage patterns reported in past earthquakes in Serbia and neighbouring countries were reviewed as a basis for damage classification pertaining to building typologies included in the proposed classification. Finally, the results of a preliminary vulnerability model were presented in the form of expert-based fragility functions derived for buildings typical of Serbia's residential building stock. [ABSTRACT FROM AUTHOR]

There are high concentrations of urban spaces and increasingly complex land use types. Providing an efficient and scientific identification of building types has become a major challenge in urban architectural planning. This study used an optimized gradient-boosted decision tree algorithm to enhance a decision tree model for building classification. Through supervised classification learning, machine learning training was conducted using a business-type weighted database. We innovatively established a form database to store input items. During parameter optimization, parameters such as the number of nodes, maximum depth, and learning rate were gradually adjusted based on the performance of the verification set to achieve optimal performance on the verification set under the same conditions. Simultaneously, a k-fold cross-validation method was used to avoid overfitting. The model clusters trained in the machine learning training corresponded to various city sizes. By setting the parameters to determine the size of the area of land for a target city, the corresponding classification model could be invoked. The experimental results show that this algorithm has high accuracy in building recognition. Especially in R, S, and U-class buildings, the overall accuracy rate of recognition reaches over 94%. [ABSTRACT FROM AUTHOR]

Recent improvements in remote sensing technologies have boosted building detection techniques from rough classifications using moderate resolution imagery to precise extraction from high-resolution imagery. Shadows frequently emerge in high-resolution urban images. To exploit shadow information, we developed a novel building detection and classification algorithm for images of urban areas with large-size shadows, employing only the visible spectral bands to determine the height levels of buildings. The proposed method, building general-classified by height (BGCH), calculates shadow orientation, detects buildings using seed-blocks, and classifies the buildings into different height groups. Our proposed approach was tested on complex urban scenes from Toronto and Beijing. The experimental results illustrate that our proposed method accurately and efficiently detects and classifies buildings by their height levels; the building detection rate exceeded 95%. The precision of classification by height levels was over 90%. This novel building-height-level detection method provides rich information at low cost and is suitable for further city scene analysis, flood disaster risk assessment, population estimation, and building change detection applications.

Remote sensing images are important data sources for land cover mapping. As one of the most important artificial features in remote sensing images, buildings play a critical role in many applications, such as population estimation and urban planning. Classifying buildings quickly and accurately ensures the reliability of the above applications. It is known that the classification accuracy of buildings (usually indicated by a comprehensive index called F1) is greatly affected by image quality. However, how image quality affects building classification accuracy is still unclear. In this study, Boltzmann entropy (an index considering both compositional and configurational information, simply called BE) is employed to describe image quality, and the potential relationships between BE and F1 are explored based on images from two open-source building datasets (i.e., the WHU and Inria datasets) in three cities (i.e., Christchurch, Chicago and Austin). Experimental results show that (1) F1 fluctuates greatly in images where building proportions are small (especially in images with building proportions smaller than 1%) and (2) BE has a negative relationship with F1 (i.e., when BE becomes larger, F1 tends to become smaller). The negative relationships are confirmed using Spearman correlation coefficients (SCCs) and various confidence intervals via bootstrapping (i.e., a nonparametric statistical method). Such discoveries are helpful in deepening our understanding of how image quality affects building classification accuracy.

Despite extensive study, performing Rapid visual screening is still a challenging task for many countries. The challenges include the lack of trained engineers, limited resources, and a large building inventory to detect. One of the most important aspect in rapid visual screening is to establish the building classification based on the guidelines’ specific criteria. This study proposes a general framework based on Convolutional Neural Network to perform automated building classification for the rapid visual screening procedure. The method classifies buildings based on the Federal Emergency Management Agency (FEMA)-154 guidelines and uses transfer learning techniques from a pre-trained network. The Indonesian building portfolio is used as a case study and a dataset of building images generated through web-scraping on Google Search™ engines and Google StreetView™ website is used for the method validation. Results show that the proposed framework has promising potential to automate the building classification based on FEMA-154 guidelines.

To understand the community risk from severe weather threats, two components, including weather information and community assets, are crucial. Recently, probabilistic hazard information (PHI) from the National Oceanic and Atmospheric Administration (NOAA) Forecasting a Continuum of Environmental Threats (FACETs) program has been developed to provide dynamic weather-related information between the watch and warning systems to weather forecasters, emergency management agencies, and the public. To predict community physical risks on critical infrastructure and building properties using PHI, building type information is required. This study applied a machine learning technique to predict building types using building footprint and city zoning data. We collected Oklahoma county building property data to train and test a random forest model. The result of this study showed that building footprint and city zoning data can be applied to classify multiple building types with an accuracy of 96%. The machine learning–based building classification contributed to the acquisition of building type data in the Oklahoma City, Oklahoma, metropolitan area. This geodatabase will be utilized to predict real-time critical infrastructure and building damage assessment using PHI. In addition to their importance to physical building damage assessment, the results can be utilized to develop postdisaster responses and planning. [ABSTRACT FROM AUTHOR]

There are high concentrations of urban spaces and increasingly complex land use types. Providing an efficient and scientific identification of building types has become a major challenge in urban architectural planning. This study used an optimized gradient-boosted decision tree algorithm to enhance a decision tree model for building classification. Through supervised classification learning, machine learning training was conducted using a business-type weighted database. We innovatively established a form database to store input items. During parameter optimization, parameters such as the number of nodes, maximum depth, and learning rate were gradually adjusted based on the performance of the verification set to achieve optimal performance on the verification set under the same conditions. Simultaneously, a k-fold cross-validation method was used to avoid overfitting. The model clusters trained in the machine learning training corresponded to various city sizes. By setting the parameters to determine the size of the area of land for a target city, the corresponding classification model could be invoked. The experimental results show that this algorithm has high accuracy in building recognition. Especially in R, S, and U-class buildings, the overall accuracy rate of recognition reaches over 94%.

Despite extensive study, performing Rapid visual screening is still a challenging task for many countries. The challenges include the lack of trained engineers, limited resources, and a large building inventory to detect. One of the most important aspect in rapid visual screening is to establish the building classi- fication based on the guidelines’ specific criteria. This study proposes a general framework based on Convolutional Neural Network to perform automated building classification for the rapid visual screening procedure. The method classifies build- ings based on the Federal Emergency Management Agency (FEMA)-154 guidelines and uses transfer learning techniques from a pre-trained network. The Indonesian building portfolio is used as a case study and a dataset of building images gener- ated through web-scraping on Google Search™ engines and Google StreetView™ website is used for the method validation. Results show that the proposed frame- work has promising potential to automate the building classification based on FEMA-154 guidelines. [ABSTRACT FROM AUTHOR]

This paper is the result of the first-phase, inter-disciplinary work of a multi-disciplinary research project ("Urban pop-up housing environments and their potential as local innovation systems") consisting of energy engineers and waste managers, landscape architects and spatial planners, innovation researchers and technology assessors. The project is aiming at globally analyzing and describing existing pop-up housings (PUH), developing modeling and assessment tools for sustainable, energy-efficient and socially innovative temporary housing solutions (THS), especially for sustainable and resilient urban structures. The present paper presents an effective application of hierarchical agglomerative clustering (HAC) for analyses of large datasets typically derived from field studies. As can be shown, the method, although well-known and successfully established in (soft) computing science, can also be used very constructively as a potential urban planning tool. The main aim of the underlying multi-disciplinary research project was to deeply analyze and structure THS and PUE. Multiple aspects are to be considered when it comes to the characterization and classification of such environments. A thorough (global) web survey of PUH and analysis of scientific literature concerning descriptive work of PUH and THS has been performed. Moreover, out of several tested different approaches and methods for classifying PUH, hierarchical clustering algorithms functioned well when properly selected metrics and cut-off criteria were applied. To be specific, the 'Minkowski'-metric and the 'Calinski-Harabasz'-criteria, as clustering indices, have shown the best overall results in clustering the inhomogeneous data concerning PUH. Several additional algorithms/functions derived from the field of hierarchical clustering have also been tested to exploit their potential in interpreting and graphically analyzing particular structures and dependencies in the resulting clusters. Hereby, (math.) the significance 'S' and (math.) proportion 'P' have been concluded to yield the best interpretable and comprehensible results when it comes to analyzing the given set (objects n = 85) of researched PUH-objects together with their properties (n > 190). The resulting easily readable graphs clearly demonstrate the applicability and usability of hierarchical clustering- and their derivative algorithms for scientifically profound building classification tasks in Urban Planning by effectively managing huge inhomogeneous building datasets. [ABSTRACT FROM AUTHOR]

Copyright of Guangdong Architecture Civil Engineering is the property of Guangdong Architecture Civil Engineering Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Monthly energy benchmarking supports identifying trends, improving energy efficiency, and conducting cost management for building owners, managers, and policymakers better than annual or hourly benchmarking. Annual data cannot fully reflect operation utility status, and hourly data poses the issue of high-cost data mining and incomparability due to its minor scale. However, the primary challenges of monthly energy benchmarking are data limitation, "black-box" barrier, and building classification uncertainty. This study proposes a novel AI-powered Monthly Energy Benchmarking Approach (MEBA) to better assess building energy use patterns, benchmark end-use loads, and track utility bills. MEBA addresses two scenarios: (1) predict complete year-round monthly energy using partial monthly energy data; (2) estimate monthly energy loads from annual total energy data. The study collects monthly electricity and natural gas energy use from two U.S. cities. For the first scenario, the entire dataset is clustered into two primary groups by Gaussian Mixture Model (GMM). Then, the two groups are divided by Self-Organizing Map (SOM) models into five subclusters via energy use patterns. For the second scenario, an additional step is needed to locate the subcluster labels with advanced Light Gradient Boosting Machine (LGBM) classifications. All five subclusters have high prediction performance with an average accuracy of >95%. Both scenarios require the last stage to predict monthly electricity and natural gas by LGBM regressions. MEBA's prediction performance achieves R2s ranging from 0.50 to 0.73, with RMSEs between 0.15 and 2.35, outperforming the state-of-the-art XGBoost model. Each subcluster exhibits distinct energy use patterns, with EUIs, electricity loads, and year built as the most significant attributes. [Display omitted] • Classify buildings via intelligent clustering models based on energy use patterns. • Predict monthly energy use by partial monthly or total annual energy data scenarios. • Building types do not significantly correlate with energy use patterns. • EUI, electricity use, and year built are the critical attributes for energy predictions. • Outperform the state-of-the-art model in classification, explainability, and prediction. [ABSTRACT FROM AUTHOR]

This paper is the result of the first-phase, inter-disciplinary work of a multi-disciplinary research project (“Urban pop-up housing environments and their potential as local innovation systems”) consisting of energy engineers and waste managers, landscape architects and spatial planners, innovation researchers and technology assessors. The project is aiming at globally analyzing and describing existing pop-up housings (PUH), developing modeling and assessment tools for sustainable, energy-efficient and socially innovative temporary housing solutions (THS), especially for sustainable and resilient urban structures. The present paper presents an effective application of hierarchical agglomerative clustering (HAC) for analyses of large datasets typically derived from field studies. As can be shown, the method, although well-known and successfully established in (soft) computing science, can also be used very constructively as a potential urban planning tool. The main aim of the underlying multi-disciplinary research project was to deeply analyze and structure THS and PUE. Multiple aspects are to be considered when it comes to the characterization and classification of such environments. A thorough (global) web survey of PUH and analysis of scientific literature concerning descriptive work of PUH and THS has been performed. Moreover, out of several tested different approaches and methods for classifying PUH, hierarchical clustering algorithms functioned well when properly selected metrics and cut-off criteria were applied. To be specific, the ‘Minkowski’-metric and the ‘Calinski-Harabasz’-criteria, as clustering indices, have shown the best overall results in clustering the inhomogeneous data concerning PUH. Several additional algorithms/functions derived from the field of hierarchical clustering have also been tested to exploit their potential in interpreting and graphically analyzing particular structures and dependencies in the resulting clusters. Hereby, (math.) the significance ‘S’ and (math.) proportion ‘P’ have been concluded to yield the best interpretable and comprehensible results when it comes to analyzing the given set (objects n = 85) of researched PUH-objects together with their properties (n > 190). The resulting easily readable graphs clearly demonstrate the applicability and usability of hierarchical clustering- and their derivative algorithms for scientifically profound building classification tasks in Urban Planning by effectively managing huge inhomogeneous building datasets.

Building function labelling plays an important role in understanding human activities inside buildings. This study develops a method of function label classification using integrated features derived from remote sensing and crowdsensing data with an extreme gradient boosting tree (XGBoost). The classification framework is verified based on a dataset from Shenzhen, China. An extended label system for six building types (residential, commercial, office, industrial, public facilities, and others) was applied, and various social functions were considered. The overall classification accuracies were 88.15% (kappa index = 0.72) and 85.56% (kappa index = 0.69). The importance of features was evaluated using the occurrence frequency of features at decision nodes. In the six-category classification system, the basic building attributes (22.99%) and POIs (46.74%) contributed most to the classification process; moreover, the building footprint (7.40%) and distance to roads (11.76%) also made notable contributions. The result shows that it is feasible to extract building environments from POI labels and building footprint geometry with a dimensional reduction model using an autoencoder. Additionally, crowdsensing data (e.g., POI and distance to roads) will become increasingly important as classification tasks become more complicated and the importance of basic building attributes declines.

The detection of buildings in the city is essential in several geospatial domains and for decision-making regarding intelligence for city planning, tax collection, project management, revenue generation, and smart cities, among other areas. In the past, the classical approach used for building detection was by using the imagery and it entailed human–computer interaction, which was a daunting proposition. To tackle this task, a novel network based on an end-to-end deep learning framework is proposed to detect and classify buildings features. The proposed CNN has three parallel stream channels: the first is the high-resolution aerial imagery, while the second stream is the digital surface model (DSM). The third was fixed on extracting deep features using the fusion of channel one and channel two, respectively. Furthermore, the channel has eight group convolution blocks of 2D convolution with three max-pooling layers. The proposed model’s efficiency and dependability were tested on three different categories of complex urban building structures in the study area. Then, morphological operations were applied to the extracted building footprints to increase the uniformity of the building boundaries and produce improved building perimeters. Thus, our approach bridges a significant gap in detecting building objects in diverse environments; the overall accuracy (OA) and kappa coefficient of the proposed method are greater than 80% and 0.605, respectively. The findings support the proposed framework and methodologies’ efficacy and effectiveness at extracting buildings from complex environments.

Due to poor building conditions, most of UAE's existing public schools do not meet the latest educational governmental requirements. For this reason, the government elaborated a new generation of public schools that soon will replace half the existing buildings. While some schools are already fully abandoned because of typological unsuitability, many others are unoccupied or underused due to social and urban changes. Regardless of the reason of the buildings' conditions, the new schools will empty over 300 existing ones, producing a large waste of resources and embodied-energy. Moving within the framework of the UAE Vision 2021, this study promotes refurbishment practices in the local school sector to minimize the use of new resources by extending the life cycle of existing ones. A multi-criteria assessment, including socio-demographic, pedagogical, typological-functional, and energy performance analyses, helped evaluating the feasibility of refurbishing existing schools as an alternative to the common practice of demolishing and building new facilities. The paper presents the results of the 3-year research conducted for UAE public schools: a deep survey campaign allowed mapping, analyzing, and classifying the existing school stock. Then, through a bottom-up approach a sample building was selected to simulate detailed refurbishment actions, addressing typological renovation and energy upgrade. The results proved the financial feasibility of the proposal. Finally, thanks to the bottom-up approach, the benchmark outcomes have been extended to the broader UAE scenario showing significant financial benefit that would derive from the application of the proposed refurbishment. [ABSTRACT FROM AUTHOR]

This study presents a building classification scheme for residential houses in Malawi by focusing upon informal construction, which accounts for more than 90% of housing in the country with the highest urbanisation rate in the world. The proposed classification is compatible with the Prompt Assessment of Global Earthquakes for Response (PAGER) method and can be used for seismic vulnerability assessments of building stock in Malawi. To obtain realistic proportions of the building classes that are prevalent in Malawi, a building survey was conducted in Central and Southern Malawi between 10th and 20th July 2017. The results from the survey are used to modify the PAGER-based proportions of main housing typologies by reflecting actual housing construction in the surveyed areas. The results clearly highlight the importance of using realistic building stock data for seismic risk assessment in Malawi; relying on global building stock information can result in significant bias of earthquake impact assessment. [ABSTRACT FROM AUTHOR]

In the context of rapid urbanization, the refined management of cities is facing higher requirements. In improving urban population management levels and the scientific allocation of resources, fine-scale population data plays an increasingly important role. The current population estimation studies mainly focus on low spatial resolution, such as city-scale and county scale, without considering differences in population distributions within cities. This paper mines and defines the spatial correlations of multi-source data, including urban building data, point of interest (POI) data, census data, and administrative division data. With populations mainly distributed in residential buildings, a population estimation model at a subdistrict scale is established based on building classifications. Composed of spatial information and attribute information, POI data are spaced irregularly. Based on this characteristic, the text classification method, frequency-inverse document frequency (TF-IDF), is applied to obtain functional classifications of buildings. Then we screen out residential buildings, and quantify characteristic variables in subdistricts, including perimeter, area, and total number of floors in residential buildings. To assess the validity of the variables, the random forest method is selected for variable screening and correlation analysis, because this method has clear advantages when dealing with unbalanced data. Under the assumption of linearity, multiple regression analysis is conducted, to obtain a linear model of the number of buildings, their geometric characteristics, and the population in each administrative division. Experiments showed that the urban fine-scale population estimation model established in this study can estimate the population at a subdistrict scale with high accuracy. This method improves the precision and automation of urban population estimation. It allows the accurate estimation of the population at a subdistrict scale, thereby providing important data to support the overall planning of regional energy resource allocation, economic development, social governance, and environmental protection.

Precise measuring of urban façade color is necessary for urban color planning. The existing manual methods of measuring building façade color are limited by time and labor costs and hardly carried out on a city scale. These methods also make it challenging to identify the role of the building function in controlling and guiding urban color planning. This paper explores a city-scale approach to façade color measurement with building functional classification using state-of-the-art deep learning techniques and street view images. Firstly, we used semantic segmentation to extract building façades and conducted the color calibration of the photos for pre-processing the collected street view images. Then, we proposed a color chart-based façade color measurement method and a multi-label deep learning-based building classification method. Next, the field survey data were used as the ground truth to verify the accuracy of the façade color measurement and building function classification. Finally, we applied our approach to generate façade color distribution maps with the building classification for three metropolises in China, and the results proved the transferability and effectiveness of the scheme. The proposed approach can provide city managers with an overall perception of urban façade color and building function across city-scale areas in a cost-efficient way, contributing to data-driven decision making for urban analytics and planning.

Deep learning paradigm has been shown to be a very efficient classification framework for many application scenarios, including the analysis of Light Detection and Ranging (LiDAR) data for building detection. In fact, deep learning acts as a set of mathematical transformations, encoding the raw input data into appropriate forms of representations that maximize the classification performance. However, it is clear that mathematical computations alone, even highly nonlinear, are not adequate to model the physical properties of a problem, distinguishing, for example, the building structures from vegetation. In this letter, we address this difficulty by augmenting the raw LiDAR data with features coming from a physical interpretation of the information. Then, we exploit a deep learning paradigm based on a convolutional neural network model to find out the best input representations suitable for the classification. As test sites, three complex urban study areas with various kinds of building structures through the LiDAR data set of Vaihingen, Germany were selected. Our method has been evaluated in the context of “ISPRS Test Project on Urban Classification and 3-D Building Reconstruction.” Comparisons with traditional methods, such as artificial neural networks and support vector machine-based classifiers, indicate the outperformance of the proposed approach in terms of robustness and efficiency. [ABSTRACT FROM AUTHOR]

Utilising satellite images for planning and development is becoming a common practice as computational power and machine learning capabilities expand. In this paper, we explore the use of satellite image derived building footprint data to classify the residential status of urban buildings in low and middle income countries. A recently developed ensemble machine learning building classification model is applied for the first time to the Democratic Republic of the Congo, and to Nigeria. The model is informed by building footprint and label data of greater completeness and attribute consistency than have previously been available for these countries. A GIS workflow is described that semiautomates the preparation of data for input to the model. The workflow is designed to be particularly useful to those who apply the model to additional countries and use input data from diverse sources. Results show that the ensemble model correctly classifies between 85% and 93% of structures as residential and nonresidential across both countries. The classification outputs are likely to be valuable in the modelling of human population distributions, as well as in a range of related applications such as urban planning, resource allocation, and service delivery.

Semantic segmentation, especially for buildings, from the very high resolution (VHR) airborne images is an important task in urban mapping applications. Nowadays, the deep learning has significantly improved and applied in computer vision applications. Fully Convolutional Networks (FCN) is one of the tops voted method due to their good performance and high computational efficiency. However, the state-of-art results of deep nets depend on the training on large-scale benchmark datasets. Unfortunately, the benchmarks of VHR images are limited and have less generalization capability to another area of interest. As existing high precision base maps are easily available and objects are not changed dramatically in an urban area, the map information can be used to label images for training samples. Apart from object changes between maps and images due to time differences, the maps often cannot perfectly match with images. In this study, the main mislabeling sources are considered and addressed by utilizing stereo images, such as relief displacement, different representation between the base map and the image, and occlusion areas in the image. These free training samples are then fed to a pre-trained FCN. To find the better result, we applied fine-tuning with different learning rates and freezing different layers. We further improved the results by introducing atrous convolution. By using free training samples, we achieve a promising building classification with 85.6% overall accuracy and 83.77% F1 score, while the result from ISPRS benchmark by using manual labels has 92.02% overall accuracy and 84.06% F1 score, due to the building complexities in our study area. [ABSTRACT FROM AUTHOR]

An integrated web application, referred to as ER² for rapid risk evaluator, is under development for a user-friendly seismic risk assessment by the non-expert public safety community. The assessment of likely negative consequences is based on pre-populated databases of seismic, building inventory and vulnerability parameters. To further accelerate the computation for near real-time analyses, implicit building fragility curves were developed as functions of the magnitude and the intensity of the seismic shaking defined with a single intensity measure, input spectral acceleration at 1.0 s implicitly considering the epicentral distance and local soil conditions. Damage probabilities were compared with those obtained with the standard fragility functions explicitly considering epicentral distances and local site classes in addition to the earthquake magnitudes and respective intensity of the seismic shaking. Different seismic scenarios were considered first for 53 building classes common in Eastern Canada, and then a reduced number of 24 combined building classes was proposed. Comparison of results indicate that the damage predictions with implicit fragility functions for short (M ≤ 5.5) and medium strong motion duration (5.5 < M ≤ 7.5) show low variation with distance and soil class, with average error of less than 3.6%. [ABSTRACT FROM AUTHOR]

In 2015, space heating and domestic hot water production accounted for around 40% of the Swiss final energy consumption. Reaching the goals of the 2050 energy strategy will require significantly reducing this share despite the growing building stock. Renewables are numerous but subject to spatial–temporal constraints. Territorial planning of energy distribution systems enabling the integration of renewables requires having a spatial–temporal characterization of the energy demand. This paper presents two bottom-up statistical extrapolation models for the estimation of the geo-dependent heat and electricity demand of the Swiss building stock. The heat demand is estimated by means of a statistical bottom-up model applied at the building level. At the municipality level, the electricity load curve is estimated by combining socio-economic indicators with average consumption per activity and/or electric device. This approach also allows to break down the estimated electricity demand according to activity type (e.g., households, various industry, and service activities) and appliance type (e.g., lighting, motor force, fridges). The total estimated aggregated demand is 94 TWh for heat and 58 TWh for electricity, which represent a deviation of 2.9 and 0.5%, respectively compared to the national energy consumption statistics. In addition, comparisons between estimated and measured electric load curves are done to validate the proposed approach. Finally, these models are used to build a geo-referred database of heat and electricity demand for the entire Swiss territory. As an application of the heat demand model, a realistic saving potential is estimated for the existing building stock; this potential could be achieved through by a deep retrofit program. One advantage of the statistical bottom-up model approach is that it allows to simulate a building stock that replicates the diversity of building demand. This point is important in order to correctly account for the mismatch between gross and net energy saving potential, often called performance gap. The impact of this performance gap is substantial since the estimated net saving potential is only half of the gross one.

Classifying building occupancy types helps build inventory databases, which are important in disaster management. These databases are used to estimate disaster loss and vulnerable to disaster. Oftentimes, this information is gathered via remote sensing data, which can help determine the shape of buildings. However, in order to be helpful, such records require precise occupancy information to support pre-disaster preparation and post-disaster recovery planning. Collecting this information is time-consuming and costly, especially in developing countries with limited resources, such as Thailand. This study offers a potential solution to these hurdles, using a combination of building construction laws and regulations on building occupancy and geospatial data to develop a new methodology for developing a database of building occupancy classification data. The results reveal that this method is effective in classifying urban building occupancy, with more than 75% overall accuracy. Therefore, these methods can be useful in the disaster management of national or local governments in countries where the means to collect such data is limited. [ABSTRACT FROM AUTHOR]

The main goal of this study is to understand the patterns in which commercial buildings consume energy, rather than evaluating building energy use based on aggregate utility bills typically linked to building principal tenant activity or occupancy type. The energy consumption patterns define buildings as externally-load, internally-load, or mixed-load dominated buildings. Penn State and Harvard campuses serve as case studies for this particular research project. The buildings in these two campuses use steam, chilled water, and electricity as energy commodities and maintain databases of different resolutions to include minute, hourly, daily, and monthly data instances depending on the commodity and available data acquisition system. The results of this study show monthly steam consumption directly correlates to outdoor environmental conditions for 88% of the studied buildings, while chilled water consumption has negligible correlation to the outdoor environmental conditions. Thus, in terms of monthly chilled water consumption, 86% of buildings are internally-load and mixed-load dominated, respectively. Chilled water consumption is better suited for the daily analyses compared to the monthly and hourly analyses. While the influence of building operation schedules affects the analyses at the hourly level, the monthly chilled water consumptions are not good indicators of the building energy consumption patterns. Electricity consumption at the monthly (or seasonal) level can support the building energy simulation tools for the building energy retrofits. A granular electricity data, e.g. interval hourly or 15-min, could reveal building operation schedules. Overall, the steam consumption for these campus buildings located in the Northeastern United States indicated a stronger correlation to the outdoor conditions than the operation schedules. The energy consumption patterns for the reviewed campus buildings suggest opportunities for the campuses to benefit from a better: (i) interactions of the steam generation/distribution systems with the building heating demand, (ii) chilled water data collection systems to optimize the cooling demand and operation schedule of the buildings, and (iii) electricity consumption with relying on daylighting and variable operation schedules during summer and winter seasons. [ABSTRACT FROM AUTHOR]

Recent years have shown a shift from pure geometric 3D city models to data with semantics. This is induced by new applications (e.g. Virtual/Augmented Reality) and also a requirement for concepts like Smart Cities. However, essential urban semantic data like building use categories is often not available. We present a first step in bridging this gap by proposing a pipeline to use crawled urban imagery and link it with ground truth cadastral data as an input for automatic building use classification. We aim to extract this city-relevant semantic information automatically from Street View (SV) imagery. Convolutional Neural Networks (CNNs) proved to be extremely successful for image interpretation, however, require a huge amount of training data. Main contribution of the paper is the automatic provision of such training datasets by linking semantic information as already available from databases provided from national mapping agencies or city administrations to the corresponding façade images extracted from SV. Finally, we present first investigations with a CNN and an alternative classifier as a proof of concept. [ABSTRACT FROM AUTHOR]

An integrated web application, referred to as ER2 for rapid risk evaluator, is under development for a user-friendly seismic risk assessment by the non-expert public safety community. The assessment of likely negative consequences is based on pre-populated databases of seismic, building inventory and vulnerability parameters. To further accelerate the computation for near real-time analyses, implicit building fragility curves were developed as functions of the magnitude and the intensity of the seismic shaking defined with a single intensity measure, input spectral acceleration at 1.0 s implicitly considering the epicentral distance and local soil conditions. Damage probabilities were compared with those obtained with the standard fragility functions explicitly considering epicentral distances and local site classes in addition to the earthquake magnitudes and respective intensity of the seismic shaking. Different seismic scenarios were considered first for 53 building classes common in Eastern Canada, and then a reduced number of 24 combined building classes was proposed. Comparison of results indicate that the damage predictions with implicit fragility functions for short (M ≤ 5.5) and medium strong motion duration (5.5 < M ≤ 7.5) show low variation with distance and soil class, with average error of less than 3.6%.

Knowledge of the German building stock is largely based on census data and annual construction statistics. Despite the wide range of statistical data, they are constrained in terms of temporal, thematic and spatial resolution, and hence do not satisfy all requirements of spatial planning and research. In this paper, we describe a new workflow for data integration that allows the quantification of the structure and dynamic of national building stocks by analyzing authoritative geodata. The proposed workflow has been developed, tested and demonstrated exemplarily for the whole country of Germany. We use nationwide and commonly available authoritative geodata products such as building footprint and address data derived from the real estate cadaster and land use information from the digital landscape model. The processing steps are (1) data preprocessing; (2) the calculation of building attributes; (3) semantic enrichment of the building using a classification tree; (4) the intersection with spatial units; and finally (5) the quantification and cartographic visualization of the building structure and dynamic. Applying the workflow to German authoritative geodata, it was possible to describe the entire building stock by 48 million polygons at different scale levels. Approximately one third of the total building stock are outbuildings. The methodological approach reveals that 62% of residential buildings are detached, 80% semi-detached and 20% terraced houses. The approach and the novel database will be very valuable for urban and energy modeling, material flow analysis, risk assessment and facility management. [ABSTRACT FROM AUTHOR]

Virtual 3D cities are becoming increasingly important as a means of visually communicating diverse urban-related information. To get a deeper understanding of a human's cognitive experience of virtual 3D cities, this paper presents a user study on the human ability to perceive building categories (e.g. residential home, office building, building with shops etc.) from geometric 3D building representations. The study reveals various dependencies between geometric properties of the 3D representations and the perceptibility of the building categories. Knowledge about which geometries are relevant, helpful or obstructive for perceiving a specific building category is derived. The importance and usability of such knowledge is demonstrated based on a perception-guided 3D building abstraction process. [ABSTRACT FROM AUTHOR]

A frequent map revision is required in GIS applications, such as disaster prevention and urban planning. In general, airborne photogrammetry and LIDAR measurements are applied to geometrical data acquisition for automated map generation and revision. However, attribute data acquisition and classification depend on manual editing works including ground surveys. In general, airborne photogrammetry and LiDAR measurements are applied to geometrical data acquisition for automated map generation and revision. However, these approaches classify geometrical attributes. Moreover, ground survey and manual editing works are finally required in attribute data classification. On the other hand, although geometrical data extraction is difficult, SAR data have a possibility to automate the attribute data acquisition and classification. The SAR data represent microwave reflections on various surfaces of ground and buildings. There are many researches related to monitoring activities of disaster, vegetation, and urban. Moreover, we have an opportunity to acquire higher resolution data in urban areas with new sensors, such as ALOS2 PALSAR2. Therefore, in this study, we focus on an integration of airborne LIDAR data and satellite SAR data for building extraction and classification. [ABSTRACT FROM AUTHOR]

This study uses cluster analysis to examine simulated energy consumption of 134 U.S. LEED NC office buildings to classify buildings into high, medium, and low energy use intensity clusters. The analysis uses energy simulation results from the LEED database, as a comparably large data set of energy end uses from sub-meter data does not yet exist. The difference between the low energy use cluster and other clusters is explained mostly by lower process loads and lower heating energy intensity, and partly by lower intensities of other HVAC related end-uses. Lighting energy use shows the least variation between clusters. The lower heating energy intensity in the low intensity cluster is largely explained by lower roof U-values, lower window-to-wall ratio, and smaller building size. Unregulated process loads are the most significant contributor to total building energy use, accounting for 36%, 33%, and 31% of the energy use in the high, medium, and low intensity clusters, respectively. This analysis provides a quantitative evaluation of the large difference in energy intensities in high-performance office buildings, showing that these buildings are dominated by internal loads, especially unregulated process loads. [ABSTRACT FROM AUTHOR]

Buildings, the basic unit of an urban landscape, host most of its socio-economic activities and play an important role in the creation of urban land-use patterns. The spatial arrangement of different building types creates varied urban land-use clusters which can provide an insight to understand the relationships between social, economic, and living spaces. The classification of such urban clusters can help in policy-making and resource management. In many countries including the UK no national-level cadastral database containing information on individual building types exists in public domain. In this paper, we present a framework for inferring functional types of buildings based on the analysis of their form (e.g. geometrical properties, such as area and perimeter, layout) and spatial relationship from large topographic and address-based GIS database. Machine learning algorithms along with exploratory spatial analysis techniques are used to create the classification rules. The classification is extended to two further levels based on the functions (use) of buildings derived from address-based data. The developed methodology was applied to the Manchester metropolitan area using the Ordnance Survey's MasterMap®, a large-scale topographic and address-based data available for the UK. [ABSTRACT FROM AUTHOR]