Development of a City-Scale Approach for Façade Color Measurement with Building Functional Classification Using Deep Learning and Street View Images

Abstract

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.

Exploring the Impact of Built Environment Attributes on Social Followings Using Social Media Data and Deep Learning

“…A total of 43.96% of the restaurants received service ratings below 7.4 (9261), and more than half of the restaurants with a rating above 7.4 and were scattered in the study area. Lastly, 58.90% of restaurants reported a consumption per capita of 63 RMB or less (12,407), and 84.59% reported a consumption per capita of 142 RMB or less (17,819). In addition, no more than one-fifth of the high-consumption restaurants showed a trend of "large dispersion and small clustering," with a certain degree of clustering on the east and west sides of the Huangpu River and the intersection of Caoyang Road and Huashan Road.…”

“…For a long time, geography has mainly relied on field observations to obtain first-hand data for geographic research. The application of remote sensing technology has promoted the classification of land [10], calculation of a vegetation index [11], measurement of façade color [12], and the evaluation of built qualities such as the characteristics of urban green spaces based on aerial measurements. Remote sensing data cannot, however, be fully utilized in the field of urban observations due to the limitations of the spatial, spectral, and temporal resolution sensors [6].…”

Using google street view panoramas to investigate the influence of urban coastal street environment on visual walkability

“…With the development of sensing technologies and map services, related map service companies (Google Maps, Baidu Maps) have collected a wealth of street view images with GPS data from all over the world (Helbich et al 2019, Rzotkiewicz et al 2018. These freely available image data have become the main data source to explore the relationship between the urban environment and human perception (Zhang et al 2021). For example, identifying street composition variation.…”

Evaluating building color harmoniousness in a historic district intelligently: An algorithm-driven approach using street-view images

“…Following this trend, a series of physical features including pedestrian walkways (Weld et al, 2019), traffic infrastructure (Hanibuchi et al, 2019), greenness (Yang et al, 2021; Ye et al, 2019) and building facades (Zhang et al, 2022), etc., have been quantitatively measured for further study. These quantitatively-measured features are also integrated with other urban data sources to measure the perceived qualities of the built environment and explore their socio-economic effects, for example, walkability (Yang et al, 2020; Zhang et al, 2019), public health (Zhang et al, 2021), transportation and mobility (Hu et al, 2020), and property price (Law et al, 2019). Among them, the buildings features extracted from SVIs is one of the focuses among these studies, which can serve as supplementary information for many other extended studies, for example, estimating energy demand (Li et al, 2018), generating precise 3D models (Kraff et al, 2020), and measuring building façade colors (Zhong et al, 2021).…”