Mapping the Vanishing and Transformation of Urban Villages in China

TL;DR

This study tackles the problem of how urban villages (UVs) in China disappear and transform after demolition by developing a deep learning framework that maps UV boundaries over multiple years using multi-temporal very-high-resolution remote sensing data. The authors compareU-Net, DeepLab-v3+, and UV-SAM for semantic segmentation, finding UV-SAM generally provides the best boundary delineation, while cross-city generalization remains challenging. They model post-demolition land-use with a Remained–Demolished–Redeveloped lifecycle and six land-use subtypes, revealing three transformation pathways: synchronized redevelopment, delayed redevelopment, and gradual optimization. The framework yields scalable, cross-regional insights into spatial restructuring and policy implementation, supporting more inclusive and sustainable urban renewal and offering a transferable approach for monitoring informal settlements globally.

Abstract

Urban villages (UVs), informal settlements embedded within China's urban fabric, have undergone widespread demolition and redevelopment in recent decades. However, there remains a lack of systematic evaluation of whether the demolished land has been effectively reused, raising concerns about the efficacy and sustainability of current redevelopment practices. To address the gap, this study proposes a deep learning-based framework to monitor the spatiotemporal changes of UVs in China. Specifically, semantic segmentation of multi-temporal remote sensing imagery is first used to map evolving UV boundaries, and then post-demolition land use is classified into six categories based on the "remained-demolished-redeveloped" phase: incomplete demolition, vacant land, construction sites, buildings, green spaces, and others. Four representative cities from China's four economic regions were selected as the study areas, i.e., Guangzhou (East), Zhengzhou (Central), Xi'an (West), and Harbin (Northeast). The results indicate: 1) UV redevelopment processes were frequently prolonged; 2) redevelopment transitions primarily occurred in peripheral areas, whereas urban cores remained relatively stable; and 3) three spatiotemporal transformation pathways, i.e., synchronized redevelopment, delayed redevelopment, and gradual optimization, were revealed. This study highlights the fragmented, complex and nonlinear nature of UV redevelopment, underscoring the need for tiered and context-sensitive planning strategies. By linking spatial dynamics with the context of redevelopment policies, the findings offer valuable empirical insights that support more inclusive, efficient, and sustainable urban renewal, while also contributing to a broader global understanding of informal settlement transformations.

Figures (8)

• Figure 1: Study area. Four cities from China's four economic divisions are selected, including Guangzhou (East), Zhengzhou (Central), Xi’an (West), and Harbin (Northeast). The major ring roads are indicated with dashed lines, including the Fourth Ring Road in Zhengzhou, the Xi’an Ring Expressway (G3001, national expressway code), and the Harbin Ring Expressway (G1001, national expressway code).
• Figure 2: Representative urban village types in the study area. (a) High-density area with blurred boundaries and active commerce; (b) Suburban village with clear edges and traditional structure; (c) Compact, independent core-area village with dense commercial activity; (d) Loosely organized settlement with semi-urban characteristics. The red boundaries of UVs in the remote sensing imagery are delineated manually. (Data source: Baidu Map (as of July 2024))
• Figure 3: The proposed framework for urban village spatiotemporal change analysis, including three major steps: (a) urban village data collection and preprocessing, (b) urban village mapping, (c) spatiotemporal change analysis.
• Figure 4: Spatiotemporal distribution of urban villages in the four cities. The results are derived from post-processed multi-temporal segmentation. Urban villages are denoted in red, while demolished sites are marked with hatching, overlaid on urban administrative boundaries (gray). The total UV area and the corresponding demolished area relative to the prior year are indicated below each map.
• Figure 5: Spatial distribution and typical examples of different land use types after demolition of urban villages. (a) Spatial distribution of land use transitions from demolished UVs, categorized by type and overlaid with road networks. (b-j) Representative cases of different transformation types across years.