Spatial Disorder in China

Protocol for assessing neighborhood physical disorder using the YOLOv8 deep learning model

Neighborhood physical disorder (PD), characterized by disruptions and irregularities in spatial elements, is associated with negative economic, social, and public health outcomes. Here, we present a protocol to quantitatively assess PD utilizing a range of metrics. We describe steps for collecting street views, constructing detection models using the YOLOv8 deep learning model, calculating PD scores, and quantifying changes in PD across streets and cites. This protocol serves as a methodological foundation for assessing PD in different countries and regions. For complete details on the use and execution of this protocol, please refer to Chen et al.
Download
Li et al 2024 Protocols_SpatialDisorder.
Adobe Acrobat Document 5.6 MB
Download
Code.rar
compressed file archive 6.7 KB
Download
YOLOv8_model.rar
compressed file archive 5.4 MB

Using street view imagery to examine the association between urban neighborhood disorder and the long-term recurrence risk of patients discharged with acute myocardial infarction in central Beijing, China

Background: To examine the association between urban neighborhood disorder and the recurrence risk of patients  with acute myocardial infarction (AMI) in central Beijing, China. 

 

Methods: Recurrent AMI was identified by the Beijing Monitoring System for Cardiovascular Diseases through the end of 2019 for patients discharged with AMI between 2007 and 2017. Cox proportional hazards models were performed to estimate associations between neighborhood disorder and AMI recurrence.

 

Results: Of 66,238 AMI patients, 11,872 had a recurrent event, and 3117 died from AMI during a median followup of 5.92 years. After covariate adjustment, AMI patients living in the high tertile of neighborhood disorder had a higher recurrence risk (hazard ratio [HR] 1.08, 95 % confidence interval [CI], 1.03–1.14) compared with those in the low tertile. A stronger association was noted for fatal recurrent AMI (HR 1.21, 95 % CI 1.10–1.34). The association was mainly observed in females (HR 1.04, 95 % CI: 1.02 to 1.06).

 

Conclusions: Serious neighborhood disorder may contribute to higher recurrence risk, particularly fatal recurrence, among AMI patients. Policies to eliminate neighborhood disorders may play an important role in the secondary prevention of cardiovascular disease. 

Download
Zhang et al 2023 Cities_AMI.pdf
Adobe Acrobat Document 5.2 MB

Measuring Physical Disorder in Urban Street Spaces: A Large-Scale Analysis Using Street View Images and Deep Learning

Physical disorder is associated with negative outcomes in economic performance, public health, and social stability, such as the depreciation of property, mental stress, fear, and crime. A limited but growing body of literature considers physical disorder in urban space, especially the topic of identifying physical disorder at a fine scale. There is currently no effective and replicable way of measuring physical disorder at a fine scale for a large area with low cost, however. To fill the gap, this article proposes an approach that takes advantage of the massive volume of street view images as input data for virtual audits and uses a deep learning model to quantitatively measure the physical disorder of urban street spaces. The results of implementing this approach with more than 700,000 streets in Chinese cities—which, to our knowledge, is the first attempt globally to quantify the physical disorder in such large urban areas—validate the effectiveness and efficiency of the approach. Through this large-scale empirical analysis in China, this article makes several theoretical contributions. First, we expand the factors of physical disorder, which were previously neglected in U.S. studies. Second, we find that urban physical disorder presents three typical spatial distributions—scattered, diffused, and linear concentrated patterns—which provide references for revealing the development trends of physical disorder and making spatial interventions. Finally, our regression analysis between physical disorder and street characteristics identified the factors that could affect physical disorder and thus enriched the theoretical underpinnings.
Download
Chen et al 2022 AAAG_Disorder.pdf
Adobe Acrobat Document 6.3 MB

Measuring individuals’ mobility-based exposure to neighborhood physical disorder with wearable cameras

To date, most studies have assessed individual exposure to neighborhood physical disorder (NPD) through the static residence-based approach, which ignores elements of human mobility and may lead to inaccurate estimates. This study assessed individual exposure to neighborhood physical disorder through the mobility-based approach using wearable cameras. The use of this approach allowed us to leverage innovative tools to accurately assess exposure to NPD in individuals’ activities in space-time. We assessed the volunteers’ exposure to neighborhood physical disorder by manually auditing pictures taken by wearable cameras on an online browserbased assessment platform. The results illustrated that wearable cameras can clearly capture the exposure while volunteers were engaged in travel behaviors. We also compared the proposed approach (mobility-based, using wearable cameras to take photos) with other approaches (with consideration of travel behaviors to varying degrees, using street view images) to demonstrate that wearable cameras can record individual exposure to neighborhood physical disorder accurately and conveniently, and the assessment results might be significantly different from those obtained by other approaches. Thus, the proposed approach is of great significance.

Download
Li et al 2022 AG_SpatialDisorder.pdf
Adobe Acrobat Document 8.2 MB

Quality Measurement of Urban Street Space from the Perspective of Spatial Disorder

In contrast to a clean, orderly and safe environment, neighborhood physical disorder refers to poor space quality or even decayed urban landscapes caused by lack of repair and management or even long-term abandonment, and interferes with the normal use of public space by residents. With the increasing demand for urban space quality, the phenomenon of physical disorder, which is a negative feature of the built environment, has gradually attracted attention, and has been proved to have a negative impact on individuals' health, public safety, and even the urban decay. To respond and intervene in the follow-up, it is necessary to find the place where the disorder occurs first. However, there is lack of reproducible, low-cost and high-efficiency large-scale measurement method for physical disorder. Taking the street space with Beijing's Fifth Ring Road as an example, this study deconstructs the phenomenon of physical disorder in the context of construction and quality improvement of Chinese urban space and reveal the negative characteristics of street space and its potential externalities. Based on massive street view image data, combined with virtual audit and deep learning models, a reliable and efficient automatic measurement model of spatial quality is proposed quantify the physical disorder phenomena of urban streets, aiming at supporting city managers and planners to understand space quality and its changes from a more comprehensive perspective, and providing an important basis for further fine management, decision-making and intervention.

 

Based on scoping review and field research, we build a quantitative checklist for physical disorder and a standard handbook for auditing that includes the characteristics of Chinese urban landscapes. Via the self-developed virtual audit online system, the validity of the checklist and the off-site audit method is verified through small-scale manual audit, and a sample library of disordered street view images is constructed. Model training and optimization are carried out applying Faster R-CNN, SSD object detection algorithm and SegNet segmentation algorithm. We finally select the optimal deep learning model for each physical disorder factor (with F2-score above 80%).

 

With street view image data for multiple years, we further applied this model to carry out an empirical study on the street space within the Fifth Ring Road in Beijing and estimated physical disorder levels throughout the city, providing evidence for understanding the characteristics of urban physical disorder in China. The deep learning results show that although the overall urban spatial quality of the city is moderate, physical disorder is still common and spread to varying degrees in Beijing urban area, where ratio of the points with disorder has reached 69.8% among the 71,165 street view points. Also, disordered areas performed concentrated in the north within the Second 

Ring Road. Stores with poor signboards, garbage/litter on street and graffiti/illegal advertisement are the main factors of disorder that affect the quality of urban space in Beijing, leaving a negative impact on the vitality of the space. For the multiyear scenario, the overall street space quality tends to improve (50.4% of street view points have witnessed the decreasing or even disappearing of disorder, while 16.1% remained basically the same). For areas within the Second Ring Road and Chang'an Avenue and the extension area, the space quality has improved significantly, indicating the certain effects of the urban space renovation. This study also sorted out various intervention methods in the urban street design guidelines at home and abroad, proposed sustainable strategies for specific disorder factors which significantly improved the space quality. 

Download
陈婧佳 2021 硕士论文_空间失序.pdf
Adobe Acrobat Document 4.5 MB

Element Identification, Measurement, Impacts Evaluation and Spatial Intervention of Disordered Urban Public Space

Under the call of the construction and quality improvement of urban space, the current uneven quality of the space caused by extensive urban development is worthy of attention, given the lack of maintenance of the old city core, the brown filed and vacant land, as well as the decay of the urban environment that is common in many cities. Drawing on a concept from sociology, this phenomenon of poor space quality and chaotic space order is defined as a physical disorder of urban space. With a new method of off-site built environment audits based on street view images, this study measures and evaluates the physical disorder of urban space within Beijing's Fifth Ring Road area. It finds that varying degrees of physical disorder is spread in this area, where the disorder has reached 50.1% among the 70,436 street points. The unkempt façades of buildings and roads lacking maintenance were the main factors of disorder that affect the quality of urban space in Beijing, leaving a negative impact on the vitality of the space. This study also carries out spatial intervention experiments that target specific disorder factors and can significantly improve the space. The large scale measurement of public spaces of poor quality or even disordered space provides an important basis for refined management and intervention of cities in the future.

Download
陈婧佳和龙瀛 2021 时代建筑_北京空间失序.pdf
Adobe Acrobat Document 2.6 MB

Strategies for improving the quality of urban street space oriented to promoting public health: Perspective from spatial disorder

In recent years, space quality and design of streets have received increasing attention. In the field of public health, the insufficient quality of urban spatial characteristics or even disorder have been proved to directly or indirectly affect the physical and mental health of individuals, implying high-risk influence on individual behavior and delivering negative health outcomes. The improvement of micro-scale spatial features is beneficial to enhancing the activity-friendliness of public space and shaping positive psychological perceptions, thereby promoting public health. Focusing on the core goal of creating high-quality urban street space, this study takes the street space of the public space as the research object, and pays attention to the phenomenon of insufficient local space quality. Based on the current quality and significant characteristics of street space quality in China, it sorts out relevant design strategies in various street design guidelines for cities at home and abroad, and proposes design response strategies for different space quality factors, so as to explore the practical points to solve the problem of poor street space quality or physical disorder in Chinese cities, to pave a way for public health-oriented environmental maintenance, improvement and organic renewal, and to further serve the refined urban space management and design.

Download
陈婧佳等 2020 城市规划_空间品质提升.pdf
Adobe Acrobat Document 11.8 MB

Large-scale Measurement of the Quality of Urban Street Space Based on Physical Disorder Theory

A Case Study of Area within the Second Ring of Hefei City

With improvements and optimization in the field of urban construction and people's pursuit of a higher quality of life, spatial quality has become an important aspect of urban research. However, rapid developments in the Chinese economy in recent years have caused disordered local urban space. In this study, area within the second ring of Hefei City was used as the research object, and multi-source data (e.g., street view images) were applied as carriers. On this basis, the physical disorder phenomenon, the relationship between different street types and the degree of physical disorder in Hefei City were explored through technical methods such as virtual built-environment audit. For area within the second ring road of Hefei City, the results revealed as following: (1) for the overall spatial quality, the degree of physical disorder was 35.11%; (2) among the spatial features explored, the commercial elements along the street presented the highest disorder degree; and (3) the quality of the space along the streets of land servicing commercial-industry facilities (Class B) was the worst, while the quality along the streets of logistics and storage land (Class W) was the best. Based on physical disorder theory, this study quantitatively measured the quality of street space. In practice, these findings provide important insights toward improving the urban management of cities in the future. In theory, they address the limitations of current research into physical disorder in China's urban space.

Download
陈纯等 南方建筑 2020_合肥空间失序.pdf
Adobe Acrobat Document 18.4 MB