Large-scale 3-D building models are a fundamental data of many research and applications. The automatic reconstruction of these 3-D models in LoD2 garners much attention and many automatic methods have been proposed. However, most existing solutions require multiple and complicated substeps for reconstructing the structure of a single building. Meanwhile, most of them have not been applied to large-scale reconstruction to better support the practical applications. Furthermore, some of them rely on the input point clouds with building classification information, thereby affecting their generalization. To resolve these issues, in this letter, we propose a workflow to fully automatically reconstruct large-scale 3-D building models in LoD2. This workflow takes airborne laser scanning (ALS) point clouds as input and uses building footprints and digital terrain model (DTM) as assistance. LoD2 3-D building models are reconstructed by a three-module pipeline: 1) building and roof segmentation; 2) 3-D roof reconstruction; and 3) final top–down extrusion with terrain information. By proposing hybrid deep-learning-based and rule-based methods for the first two modules, we ensure the accurate structure output of reconstruction results as much as possible. The experimental results on point clouds covering the whole city of Trondheim, Norway, indicate that the proposed workflow can effectively reconstruct large-scale 3-D building models in LoD2 with the acceptable RMSE.

Accurately identifying urban building layouts plays a crucial role in understanding the complexity of urban construction and the level of economic development. Previous footprint segmentation methods have struggled to adapt to the diverse morphology of buildings, limiting the accurate extraction of building footprints from remote sensing imagery and impeding insights and understanding of urban areas. To this end, we propose a framework named CityInsight for analyzing urban building morphology from remote sensing imagery. First, we establish a semantic segmentation network, dual-condition diffusion network (DC-Net), based on a diffusion model to accurately identify building footprints from remote sensing images. Second, we use uncertainty attention and condition attention to generate spatial and semantic priors. Finally, we design a condition injection module to incorporate spatial and semantic information into the diffusion learning. Comprehensive experiments demonstrate the accuracy, robustness, and generalization of the proposed method. The$F_{1}$-scores of the DC-Net on the large-scale remote sensing datasets SpaceNet, WHU Building, Inria, and Massachusetts are 92.05%, 96.59%, 92.17%, and 92.86%, respectively. Furthermore, the footprint segmentation is utilized for subsequent urban function identification and urban analysis of the Zona Oeste of Rio de Janeiro, to emphasize the application value of CityInsight. Our code is available athttps://github.com/Ting-Devin-Han/CityInsight

The emergence of crowdsourced geographic information (CGI) has markedly accelerated the evolution of land-use and land-cover (LULC) mapping. This approach taps into the collective power of the public to share spatial information, providing a relevant data source for producing LULC maps. Through the analysis of 262 papers published from 2012 to 2023, this work provides a comprehensive overview of the field, including prominent researchers, key areas of study, major CGI data sources, mapping methods, and the scope of LULC research. Additionally, it evaluates the pros and cons of various data sources and mapping methods. The findings reveal that while applying CGI with LULC labels is a common way by using spatial analysis, it is limited by incomplete CGI coverage and other data quality issues. In contrast, extracting semantic features from CGI for LULC interpretation often requires integrating multiple CGI datasets and remote sensing imagery, alongside advanced methods such as ensemble and deep learning. The paper also delves into the challenges posed by the quality of CGI data in LULC mapping and explores the promising potential of introducing large language models to overcome these hurdles.

In the essential task of roof segmentation, the performance and generalization of deep learning-based methods for roof segmentation are much affected by the training datasets. Existing datasets for roof segmentation reveal the limitation of size and diversity. To address these issues, a framework is proposed to achieve fully automatic generation of point cloud roof segmentation datasets in this study. This framework fully leverages open 3-D building model data in LoD2, generating roof point clouds that contain real geographic information as well as accurate roof segment information. The point density and noise level can be defined by users in the framework, ensuring its flexibility and practicality across different environments. A pipeline is also proposed to utilize the generated point clouds for assisting in the training of deep learning-based roof segmentation methods. For validation purposes, a generated roof segmentation dataset including 50463 3-D building models, NRW3D, is created based on open geodata in North Rhine-Westphalia (NRW) state, Germany. The experimental results on NRW3D and the RoofNTNU dataset demonstrate the effectiveness and quality of the generated dataset as well as the proposed framework and pipeline. The source code for the proposed framework is available at Zenodo.

Extracting precise location information from microblogs is a crucial task in many applications, particularly in disaster response, revealing where damages are, where people need assistance, and where help can be found. A crucial prerequisite to location extraction is place name extraction. In this paper, we present GazPNE: a hybrid approach to place name extraction which fuses rules, gazetteers, and deep learning techniques without requiring any manually annotated data. The core of the approach is to learn the intrinsic characteristics of multi-word place names with deep learning from gazetteers. Specifically, GazPNE consists of a rule-based system to select n-grams from the microblogs that potentially contain place names, and a C-LSTM model that decides if the selected n-gram is a place name or not. The C-LSTM is trained on 388.1 million examples containing 6.8 million positive examples with US and Indian place names extracted from OpenStreetMap and 381.3 million negative examples synthesized by rules. We evaluate GazPNE against the SoTA on a manually annotated 4,500 tweet dataset which contains 9,026 place names from three foods: 2016 in Louisiana (US), 2016 in Houston (US), and 2015 in Chennai (India). GazPNE achieves SotA performance on the test data with an F1 of 0.84.

The concept of “human as sensors” defines a new sensing model, in which humans act as sensors by contributing their observations, perceptions, and sensations. This is crucial for the development of Social Internet of Things, which is an integral part of cyber-physical–social systems. Online social media platforms, as the most active places where users act as social sensors, are responsive to real-world events and are useful for gathering situational information in real time. Unfortunately, posts rarely contain structured geographic information, thus hindering their usage for contributing to various challenges, such as emergency response. We address this limitation by introducing a general approach for extracting place names from tweets, named GazPNE2. It combines global gazetteers (i.e., OpenStreetMap and GeoNames), deep learning, and pretrained transformer models (i.e., BERT and BERTweet), which requires no manually annotated data. It can extract place names at both coarse (e.g., city) and fine-grained (e.g., street and POI) levels and place names with abbreviations. To fully evaluate GazPNE2 and compare it with 11 competing approaches, we use 19 public tweet data sets, containing 38 802 tweets and 22 197 places across the world. The results show GazPNE2 achieves a much higher F1 (0.8) than the other approaches. Furthermore, we apply GazPNE2 to three large unannotated tweet data sets related to over 20 crisis events (e.g., coronavirus disease 2019), containing 560 040 tweets. An F1 of 0.84 is achieved on 3000 tweets, which are randomly selected from the three data sets and then manually annotated. Code and data are available on GitHub page:https://github.com/uhuohuy/GazPNE2.

Determining the location of a building’s entrance is crucial to location-based services, such as wayfinding for pedestrians. Unfortunately, entrance information is often missing from current mainstream map providers such as Google Maps. Frequently, automatic approaches for detecting building entrances are based on street-level images that are not widely available. To address this issue, we propose a more general approach for inferring the main entrances of public buildings based on the association between spatial elements extracted from OpenStreetMap. In particular, we adopt three binary classification approaches, weighted random forest, balanced random forest, and smooth-boost to model the association relationship. There are two types of features considered in the classification: intrinsic features derived from building footprints and extrinsic features derived from spatial contexts, such as roads, green spaces, bicycle parking areas, and neighboring buildings. We conducted extensive experiments on 320 public buildings with an average perimeter of 350 m. The experimental results showed that the locations of building entrances estimated by the weighted random forest and balanced random forest models have a mean linear distance error of 21 m and a mean path distance error of 22 m, ruling out 90% of the incorrect locations of the main entrance of buildings.

Façade parsing is an essential process before the 3-D modeling of digital or virtual 3-D city models. The existing grammar-based approaches for façade parsing rely on strong prior knowledge but can obtain façade parts with better structure. Pixelwise-segmentation-based approaches achieve façade parsing with much less knowledge but the resulting structure of façade parts is normally incomplete. Both these approaches are restricted by their high reliance on the data set. Therefore, they cannot be applied for façade parsing with complex scenes. To address this issue, we built a large street-level data set by taking Mapillary images as the training data for more general scenes. At the same time, we propose a new pipeline based on convolutional neural network (CNN) that combines pixelwise segmentation and global object detection to achieve better results for facade parsing. Our pipeline can be applied to façade images after rectification and street-level façade images with complex scenes. The result of the ablation study demonstrates that the design of our pipeline is effective. We test our pipeline on the classic ECP2011 data set and our new large street-level data set. Our pipeline achieves state-of-the-art results for both the data sets: an accuracy of 98.2% and the mean average precision (mAP) of 98.8% on the ECP2011 data set as well as the mAP of 81.1% for façade parts parsing on our street-level data set.

High-Accuracy and high-efficiency 3-D sensing and associated data processing techniques are urgently needed for today’s roadway inventory, infrastructure health monitoring, autonomous driving, connected vehicles, urban modeling, and smart cities. 3D geospatial data acquired by digital photogrammetry or laser scanning or LiDAR systems have become one of the most critical data sources to support the above-mentioned applications. While progress has been made to applying 3D sensory data to those applications related to intelligent transportation systems (ITS), such as road network extraction, platform localization, obstacle avoidance, high-definition map generation, and transportation infrastructure inventory, many essential questions remain regarding the processing and understanding such massive 3D datasets in ITS-related applications. The authors have selected four articles for review in this Special issue. A summary of these articles is outlined below.

More than 10 years have passed since the coining of the term volunteered geographic information (VGI) in 2007. This article presents the results of a review of the literature concerning VGI. A total of 346 articles published in 24 international refereed journals in GIScience between 2007 and 2017 have been reviewed. The review has uncovered varying levels of popularity of VGI research over space and time, and varying interests in various sources of VGI (e.g. OpenStreetMap) and VGI-related terms (e.g. user-generated content) that point to the multi-perspective nature of VGI. Content-wise, using latent Dirichlet allocation (LDA), this study has extracted 50 specific research topics pertinent to VGI. The 50 topics have been subsequently clustered into 13 intermediate topics and three overarching themes to allow a hierarchical topic review. The overarching VGI research themes include (1) VGI contributions and contributors, (2) main fields applying VGI, and (3) conceptions and envisions. The review of the articles under the three themes has revealed the progress and the points that demand attention regarding the individual topics. This article also discusses the areas that the existing research has not yet adequately explored and proposes an agenda for potential future research endeavors.

Satellite images are widely applied in humanitarian mapping that labels buildings, roads, and so on for humanitarian aid and economic development. However, the labeling now is mostly done by volunteers. In this paper, we utilize deep learning to solve humanitarian mapping tasks of a mobile software named MapSwipe. The current deep learning techniques, e.g., convolutional neural network (CNN), can recognize ground objects from satellite images but rely on numerous labels for training for each specific task. We solve this problem by fusing multiple freely accessible crowdsourced geographic data and propose an active learning-based CNN training framework named MC-CNN to deal with the quality issues of the labels extracted from these data, including incompleteness (e.g., some kinds of object are not labeled) and heterogeneity (e.g., different spatial granularities). The method is evaluated with building mapping in South Malawi and road mapping in Guinea with level-18 satellite images provided by Bing Map and volunteered geographic information from OpenStreetMap, MapSwipe, and OsmAnd. The results based on multiple metrics, including Precision, Recall, F1 Score, and area under the receiver operating characteristic curve, show that MC-CNN can fuse the crowdsourced labels for higher prediction performance and be successfully applied in MapSwipe for humanitarian mapping with 85% labor saved and an overall accuracy of 0.86 achieved.

Modeling the geographic distribution of tourists at a tourist destination is crucial when it comes to enhancing the destination’s resilience to disasters and crises, as it enables the efficient allocation of limited resources to precise geographic locations. Seldom have existing studies explored the geographic distribution of tourists through understanding the mechanisms behind it. This article proposes to couple maximum entropy modeling with geotagged social media data to determine the geographic distribution of tourists in order to facilitate disaster and crisis management at tourist destinations. As one of the most popular tourist destinations in the United States, San Diego was chosen as the study area to demonstrate the proposed approach. We modeled the tourist geographic distribution in the study area by quantifying the relationship between the distribution and five environmental factors, including land use, land parcel, elevation, distance to the nearest major road and distance to the nearest transit stop. The geographic distribution’s dependency on and sensitivity to the environmental factors were uncovered. The model was subsequently applied to estimate the potential impacts of one simulated tsunami disaster and one simulated traffic breakdown due to crisis events such as a political protest or a fire hazard. As such, the effectiveness of the approach has been demonstrated with specific disaster and crisis scenarios.

Matching road networks is an essential step for data enrichment and data quality assessment, among other processes. Conventionally, road networks from two datasets are matched using a line-based approach that checks for the similarity of properties of line segments. In this article, a polygon-based approach is proposed to match the OpenStreetMap road network with authority data. The algorithm first extracts urban blocks that are central elements of urban planning and are represented by polygons surrounded by their surrounding streets, and it then assigns road lines to edges of urban blocks by checking their topologies. In the matching process, polygons of urban blocks are matched in the first step by checking for overlapping areas. In the second step, edges of a matched urban block pair are further matched with each other. Road lines that are assigned to the same matched pair of urban block edges are then matched with each other. The computational cost is substantially reduced because the proposed approach matches polygons instead of road lines, and thus, the process of matching is accelerated. Experiments on Heidelberg and Shanghai datasets show that the proposed approach achieves good and robust matching results, with a precision higher than 96% and a F1-score better than 90%.

In the past two years, several applications of generating three-dimensional (3D) buildings from OpenStreetMap (OSM) have been made available, for instance, OSM-3D, OSM2World, OSM Building, etc. In these projects, 3D buildings are reconstructed using the buildings’ footprints and information about their attributes, which are documented as tags in OSM. Therefore, the quality of 3D buildings relies strongly on the quality of the building footprints data in OSM. This article is dedicated to a quality assessment of building footprints data in OSM for the German city of Munich, which is one of the most developed cities in OSM. The data are evaluated in terms of completeness, semantic accuracy, position accuracy, and shape accuracy by using building footprints in ATKIS (German Authority Topographic–Cartographic Information System) as reference data. The process contains three steps: finding correspondence between OSM and ATKIS data, calculating parameters of the four quality criteria, and statistical analysis. The results show that OSM footprint data in Munich have a high completeness and semantic accuracy. There is an offset of about four meters on average in terms of position accuracy. With respect to shape, OSM building footprints have a high similarity to those in ATKIS data. However, some architectural details are missing; hence, the OSM footprints can be regarded as a simplified version of those in ATKIS data.

This study proposes a novel approach for extracting multilane roads from urban road networks in OpenStreetMap (OSM) data sets as functional high-level roads, thereby allowing comparative analyses to determine the differences between this functional hierarchy and other hierarchies. OSM road networks have high levels of detail and complex structures, but they also have large numbers of duplicated lines for the same road features, which leads to difficulties and low efficiency when extracting multilane roads using conventional methods based on the analysis and operations of line segments. To overcome these deficiencies, a polygon-based method is proposed that is based on shape analysis and Gestalt theory, which treats polygons surrounded by roads as operating elements. First, shape descriptors are calculated for each polygon in networks and are used for classification. Second, candidate multilane polygons are classified as seeds based on all the polygons used as shape descriptors by a support vector machine. Finally, based on the seed polygons, a region-growing method is proposed that connects and fills the multilane features according to Gestalt theory. An experiment using OSM data from different urban networks verified the validity of the proposed method. The method achieved good and effective extraction performance, regardless of the complexity and duplication of data sets. Thus, a comparative analysis with high-level roads extracted based on road type attributes and structural analysis was performed to demonstrate the differences between the constructed road levels and other hierarchies.

This letter is dedicated to a generic approach for the automated detection and classification of man-made objects in urban corridors from point clouds acquired by vehicle-borne mobile laser scanning (MLS). The approach is designed based on a priori knowledge in urban areas: 1) man-made objects feature geometric regularity such as vertical planar structures (e.g., building facades), whereas vegetation reveals huge diversity in shape and point distribution and 2) different types of urban man-made objects can be characterized by the point extension and the height above the ground level. Therefore, MLS-based point clouds are first divided into three layers with respect to the vertical height. In each layer, seed points of man-made objects are indicated by a line filter in the footprints of off-ground objects, which is generated by binarizing the spatial accumulation map of the point clouds. These seed points are further classified by examining in which layers the seed points of objects are found. Finally, points belonging to respective objects can be retrieved based on the classified seed points. The experiments show that various man-made objects on both sides of the street can be well detected, with a detection rate of up to 83%. For the classification of detected urban objects, overall accuracy of 92.37% can be achieved.

CityGML (City Geography Markup Language), the OGC (Open Open Geospatial Consortium) standard on three-dimensional (3D) city modeling, is widely used in an increasing number of applications, because it models a city with rich geometrical and semantic information. The underlying building model differentiates four consecutive levels of detail (LoDs). Nowadays, most city buildings are reconstructed in LoD3, while few landmarks in LoD4. For visualization or other purposes, buildings in LoD2 or LoD1 need to be derived from LoD3 models. But CityGML does not indicate methods for the automatic derivation of the different LoDs. This article presents an approach for deriving LoD2 buildings from LoD3 models which are essentially the exterior shells of buildings without opening objects. This approach treats different semantic components of a building separately with the aim to preserve the characteristics of ground plan, roof, and wall structures as far as possible. The process is composed of three steps: simplifying wall elements, generalizing roof structures, and then reconstructing the 3D building by intersecting the wall and roof polygons. The first step simplifies ground plan with wall elements projected onto the ground. A new algorithm is developed to handle not only simple structures like parallel and rectangle shapes but also complicated structures such as non-parallel, non-rectangular shapes and long narrow angles. The algorithm for generalizing roof structure is based on the same principles; however, the calculation has to be conducted in 3D space. Moreover, the simplified polygons of roof structure are further merged and typified depending on the spatial relations between two neighboring polygons. In the third step, generalized 3D buildings are reconstructed by increasing walls in height and intersecting with roof structures. The approach has been implemented and tested on a number of 3D buildings. The experiments have verified that the 3D building can be efficiently generalized, while the characteristics of wall and roof structure can be well preserved after the simplification.