Urban Function as a New Perspective for Adaptive Street Quality Assessment

Abstract

Street networks are considered to be one significant component of urban structures that serve various urban functions. Assessing the quality of each street is important for managing natural and public resources, organizing urban morphologies and improving city vitality. While current research focuses on particular street assessment indices, such as accessibility and connectivity, they ignore biases in street assessment caused by differences in urban functions. To address this issue, an adaptive approach to assessing street quality from the perspective of the variation in urban functions is proposed. First, an adaptive urban function detection model is established, with street-level element segmenting using PSPNet and semantic urban function extraction using LDA topic modelling. On this basis, an urban function-driven street quality assessment is proposed to adaptively evaluate multilevel urban streets. Taking Tianhe District in Guangzhou, Guangdong Province, as the study area, experiments using street view images and points of interest (POIs) are applied to validate the proposed approach. The experiment results in a model for adaptive urban function detection with an overall accuracy of 64.3%, showing that streets with different urban functions, including traffic, commercial, and residential functions, can be assessed. The experimental results can facilitate urban function organization and urban land-use planning.

Influence of Built Environment on Street Vitality: A Case Study of West Nanjing Road in Shanghai Based on Mobile Location Data

“…The calculation of instability needs to be based on the division of spatial units to show the spatiotemporal characteristics of instability. The spatial units are divided by the road intersections and the building facades on both sides of the street as the boundaries [19,29]. Since the LBS data is based on GPS positioning, there is a positioning error of 10 m. Therefore, the spatial unit buffers 10 m outwards so as to select the active people in the street as much as possible.…”

“…With the help of mobile location data with massive samples to sense crowd activity, street research can dig out the behavior rules of space users in continuous time and space and is no longer limited to small sample sampling and small-scale research. In recent years, a research trend has been to perceive urban streets from a new perspective, evaluate the characteristics of vitality, and examine the street design, especially the concept of place and its related theories, which are closely related to users' behavior and feelings [19]. Several previous studies have started from the idea of quantitative verification and have used more detailed records of residents' behaviors, quantitative models, and statistical methods to deepen the design theories of how the built environment affects street vitality and to propose street design strategies to better promote vitality [20][21][22].…”

Exploring the Effects of Roadside Vegetation on the Urban Thermal Environment Using Street View Images

“…In this paper, the PSPNet semantic segmentation network was used to extract e ments, including grasses, shrubs, trees and other elements, and this network achieved e cellent performance (Figure 3). In particular, the model predicted results with an accura of up to 77.23% in Cityscapes (a large-scale dataset containing high-quality pixel-level a notations of 5000 images of 50 cities taken in different seasons) [42]. In the tenfold cro validation process, the mean IoU value derived for the independent test set was 83.8 and the mean IoU values derived for trees, shrubs and grasses reached 89.6%, 73.3% a 84.1%, respectively.…”

Exploring the Correlation between Block Vitality and Block Environment Based on Multisource Big Data: Taking Wuhan City as an Example

“…PSPNet is a novel deep convolutional neural network model that classifies each pixel of an image and divides the image into several visually meaningful regions. Reportedly, it can be used to precisely and efficiently segment street-view images and identify the percentage of greenery, sky, and other elements [53][54][55]. The model schematic is shown in Figure 8.…”

“…The image was segmented into five parts-building, sky, greenery, water, and road (Figure 9). The ratio of sky pixels to the total number of pixels in the image was calculated to represent the percentage of the sky in the image, and the green-looking ratio was calculated in the same way [55].…”