Abstract:Building use identification is crucial in urban planning and management. Current identification methods often rely on a single data source and neglect spatial proximity. In this paper, we propose a stepwise urban building use identification framework that integrates remote sensing and social sensing data with spatial constraints based on the association of buildings with Point of Interest (POI), Area of Interest (AOI) and remote sensing data. First, the study data are preprocessed using geometric correction and POI and AOI reclassification. Then, we identify buildings with the quantitative-density index of the POIs as well as the spatial relationships between the AOIs and the buildings. Next, we generate Traffic Analysis Zones (TAZs) from road network data and utilize the similarity in physical features of buildings from remote sensing data to identify building use within spatial constraints. Finally, POI kernel density estimation is used to determine the semantic features of the buildings, and the similarity of the features between the buildings is utilized to identify the remaining buildings. The specificity of our proposed framework lies not only in the combination of multiple source data at the building-level but also in the introduction of the spatial relationships of AOIs and spatial constraints. Shenyang is selected as an example. The proposed framework identifies buildings as commercial, residential, industrial, public service and scenic spots. The accuracy assessment indicates that the proposed framework performs well, with an Overall Accuracy (OA) of 87.1% and a kappa coefficient (kappa) of 73.4%. The results of the comparison experiments show that the consideration of spatial constraints and the integration of multiple data sources help to improve the accuracy of building use identification. The proposed framework provides a new tool for better identification of urban building use, and the generated data are suitable for in-depth analyses such as building-level urban heat islands.

Integrating the Eigendecomposition Approach and k-Means Clustering for Inferring Building Functions with Location-Based Social Media Data

Integrating multi-source big data to infer building functions

Research on identifying urban regions of different functions based on POI data

Delineating Urban Functional Areas with Building-Level Social Media Data: A Dynamic Time Warping (DTW) Distance Based K -Medoids Method

Type Identification of Land Use in Metro Station Area Based on Spatial–Temporal Features Extraction of Human Activities

Identification of urban functional areas based on social media location data:a case study of Shanghai

Identifying Urban Spatial Structure and Urban Vibrancy in Highly Dense Cities Using Georeferenced Social Media Data

Urban Spatial-Temporal Activity Structures: a New Approach to Inferring the Intra-urban Functional Regions Via Social Media Check-In Data

Identifying Building Functions From The Spatiotemporal Population Density And The Interactions Of People Among Buildings

A multi‐modal transportation data‐driven approach to identify urban functional zones: An exploration based on Hangzhou City, China

Latent spatio-temporal activity structures: a new approach to inferring intra-urban functional regions via social media check-in data.

A Graph Deep Learning Approach for Urban Building Grouping

Identifying Building Function Using Multisource Data: A Case Study of China's Three Major Urban Agglomerations

Urban building function classification based on multisource geospatial data: a two-stage method combining unsupervised and supervised algorithms

Integrating street view imagery and taxi trajectory for identifying urban function of street space

Characterizing Mixed-Use Buildings Based On Multi-Source Big Data

An Urban Building Use Identification Framework Based on Integrated Remote Sensing and Social Sensing Data with Spatial Constraints

Inferring Mixed Use of Buildings with Multisource Data Based on Tensor Decomposition

Large-scale urban building function mapping by integrating multi-source web-based geospatial data

Recognizing mixed urban functions from human activities using representation learning methods

Exploring the distribution and dynamics of functional regions using mobile phone data and social media data