Quantifying aspects of urban design on a massive scale is crucial to help develop a deeper understanding of urban designs elements that contribute to the success of a public space. In this study, we further develop the Street-Frontage-Net (SFN), a convolutional neural network (CNN) that can successfully evaluate the quality of street frontage as either being active (frontage containing windows and doors) or blank (frontage containing walls, fences and garages). Small-scale studies have indicated that the more active the frontage, the livelier and safer a street feels. However, collecting the city-level data necessary to evaluate street frontage quality is costly. The SFN model uses a deep CNN to classify the frontage of a street. This study expands on the previous research via five experiments. We find robust results in classifying frontage quality for an out-of-sample test set that achieves an accuracy of up to 92.0%. We also find active frontages in a neighbourhood has a significant link with increased house prices. Lastly, we find that active frontage is associated with more scenicness compared to blank frontage. While further research is needed, the results indicate the great potential for using deep learning methods in geographic information extraction and urban design.

As a vital indicator for measuring urban development, urban areas are expected to be identified explicitly and conveniently with widely available data sets, thereby benefiting planning decisions and relevant urban studies. Existing approaches to identifying urban areas are normally based on midresolution sensing data sets, low-resolution socioeconomic information (e.g., population density) in space (e.g., cells with several square kilometers or even larger towns or wards). Yet, few of these approaches pay attention to defining urban areas with high-resolution microdata for large areas by incorporating morphological and functional characteristics. This article investigates an automated framework to delineate urban areas at the block level, using increasingly available ordnance surveys for generating all blocks (or geounits) and ubiquitous points of interest (POIs) for inferring density of each block. A vector cellular automata model was adopted for identifying urban blocks from all generated blocks, taking into account density, neighborhood condition, and other spatial variables of each block. We applied this approach for mapping urban areas of all 654 Chinese cities and compared them with those interpreted from midresolution remote sensing images and inferred by population density and road intersections. Our proposed framework is proven to be more straightforward, time-saving, and fine-scaled compared with other existing ones. It asserts the need for consistency, efficiency, and availability in defining urban areas with consideration of omnipresent spatial and functional factors across cities.

Under China’s Rural Revitalization program, it is essential to accurately determine the prospects of revival or decay for the villages alongside specific goals and paths, which existing literature lacks a systematic coverage. Based on rural typology theories, this paper proposes an analytical framework to determine the type of village revitalization from the perspective of factor endowments. Utilizing five groups of 45 indicators characterizing the natural, socio-economic, and cultural endowments of villages, this paper applies the Self-Organizing Mapping neural network to cluster 2,388,579 natural villages in 48,322 townships across the nation into the four basic types of rural revitalization as directed by China’s Strategic Plan for Rural Revitalization (2018–2022): (1) Agglomerative Promotion, (2) Suburban Annexation, (3) Special Endowment-based Development, and (4) Out-migration and Relocation. The results of cluster analysis are spatially visualized to form a national rural revitalization zoning map at the township level, the first attempt to our knowledge. We conclude the paper with discussions on the revitalization paths of the various types of villages, particularly the seemingly gloomy prospect of 2/3 of the villages falling into the fourth category, and ways to interpret the deterministic nature of the conclusion. The paper expands the understanding of rural typology to a national scale with both innovative categorization processes and strong linkages to revitalization practices.

Well-connected urban areas defined by different types of urban flow define the boundaries reflecting the relatedness of places in terms of their functionality. Yet, attempts to define cities and their communities based on aggregated data normally neglect the inherent differences between different groups of people. Based on the disaggregated flow data, this study detects community structures in the London Metropolitan Area perceived by different occupations by using the multi-level modularity optimisation algorithm. The results show the difference between our perceptions of different functional regions across occupations. The higher managerial groups have a more global sense than the lesser managerial occupations who have more segmented and local perceptions regarding their functional regions. This is well illustrated by the shift of the documented modularity scores across groups. Although the transport network and various natural boundaries do play a part in the locational patterns of the derived communities, it is found that the relative self-containment of functional regions is interpreted differently by different occupations. This kind of representation is of great value in advancing our knowledge regarding how different places are perceived by different occupations with all the implications for future commuting that future planning for housing and employment will bring.