Page 18 - JCAU-6-1

Journal of Chinese

Architecture and Urbanism

ORIGINAL ARTICLE
Exploring the spatial attributes of streets in Lu

Xun’s hometown of Shaoxing, China, through
image semantic segmentation

Qingyuan Hong*

Department of Architecture, School of Architecture, Southeast University, Nanjing, China
(This article belongs to the Special Issue: Advanced Technologies and Practices in Built Environment
and Cultural Heritage)

Abstract

Image semantic segmentation, a deep learning algorithm, enables the recognition
of pixel collections that form distinct categories, allowing for the identification of
vehicles, pedestrians, traffic signs, pavement, and other road features. In urban and
architectural design domains, image semantic segmentation and related techniques
empower practitioners and researchers to efficiently analyze the distribution of
public spaces. This application facilitates a better understanding of how people
interact with urban environments, ultimately improving the design of functional and
inviting spaces. This paper presents an analysis of images of different streets within
the Lu Xun Heritage Area in Shaoxing, Zhejiang Province, China, which were obtained
through onsite photography. The images were sampled, segmented, and compared
*Corresponding author: to assess the spatial characteristics of distinct street types. A self-trained semantic
Qingyuan Hong segmentation model based on the Cityscapes dataset and the PaddlePaddle
(hongqingyuan@seu.edu.cn)
framework was employed to statistically analyze space variations across various
Citation: Hong, Q. (2024). dimensions. This analysis contributes to a better understanding of historical street
Exploring the spatial attributes of
streets in Lu Xun’s hometown of structure and provides insights into the integration of artificial intelligence in urban
Shaoxing, China, through image planning and design.
semantic segmentation. Journal of
Chinese Architecture and Urbanism,
6(1), 1736. Keywords: Lu Xun’s hometown; Semantic segmentation; Street space; Historic streets;
https://doi.org/10.36922/jcau.1736
Vibrant streets; Shaoxing, China
Received: August 31, 2023
Accepted: October 25, 2023
Published Online: January 5, 2024 1. Introduction
Copyright: © 2024 Author(s).
This is an open-access article Semantic segmentation has found extensive application in the conservation and digital
distributed under the terms of the reconstruction of historical sites and building blocks, as well as in the decision-making
Creative Commons Attribution- processes of urban and rural planning and design (Picon & Zhou, 2019). This advanced
Non-Commercial 4.0 International
(CC BY-NC 4.0), which permits all computer vision technology employs machine learning methods to partition various
non-commercial use, distribution, areas in street view images into different categories. It acquires semantic information
and reproduction in any medium,
provided the original work is for each element, counts the number and distribution of each element, and facilitates
properly cited. valuable evaluations of the image.
Publisher’s Note: AccScience In everyday life, streets are commonly encountered and can be broadly categorized
Publishing remains neutral with
regard to jurisdictional claims in into modern streets and historic streets. Modern streets encompass main thoroughfares
published maps and institutional in cities or communities, bustling with commercial, social, and cultural activities.
affiliations.

Volume 6 Issue 1 (2024) 1 https://doi.org/10.36922/jcau.1736