Research on ancient town style construction strategies based on coupled quantitative analysis of AI visual recognition and scenic beauty evaluation

IF 3.1 1区艺术学 0 ARCHITECTURE

Frontiers of Architectural Research Pub Date : 2025-01-07 DOI:10.1016/j.foar.2024.11.004

Wu Jin , Bifeng Zhu , Hiroatsu Fukuda

{"title":"Research on ancient town style construction strategies based on coupled quantitative analysis of AI visual recognition and scenic beauty evaluation","authors":"Wu Jin , Bifeng Zhu , Hiroatsu Fukuda","doi":"10.1016/j.foar.2024.11.004","DOIUrl":null,"url":null,"abstract":"<div><div>How to create the scenery is the key issue in ancient towns. In this study, 50 photos were collected and distributed through the Internet. First, 456 online questionnaires with 25,080 data were got. Respondents' favoritism was affected by gender, age, region, profession, and education. Second, SAM computer model was applied to image recognition of Wuzhen style photos, analyzing their visual elements. Third, SPSS software was used to analyze the correlation between subjective beauty degree score and objective landscape elements. Based on the coupled quantitative analysis of AI visual recognition and beauty degree score, it is found that the landscape elements that tourists cared most about are: water bodies, ancient buildings and boats. The proportions of the best landscape elements for the spatial sense of the ancient town are the sky ranged from 26.4% to 38.2%, water body ranged from 19.7% to 34.3%, and buildings ranged from 10.4% to 38.2%. This study reveals the pattern of different types of tourists' evaluation of the landscape to summarize the landscape construction strategy of ancient towns in Jiangnan accordingly. The results are not only benefit to the cultural tourism of Wuzhen, but can also be applied to many ancient towns in Jiangnan.</div></div>","PeriodicalId":51662,"journal":{"name":"Frontiers of Architectural Research","volume":"14 3","pages":"Pages 654-671"},"PeriodicalIF":3.1000,"publicationDate":"2025-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Frontiers of Architectural Research","FirstCategoryId":"1087","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2095263524001730","RegionNum":1,"RegionCategory":"艺术学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"0","JCRName":"ARCHITECTURE","Score":null,"Total":0}

引用次数: 0

Abstract

How to create the scenery is the key issue in ancient towns. In this study, 50 photos were collected and distributed through the Internet. First, 456 online questionnaires with 25,080 data were got. Respondents' favoritism was affected by gender, age, region, profession, and education. Second, SAM computer model was applied to image recognition of Wuzhen style photos, analyzing their visual elements. Third, SPSS software was used to analyze the correlation between subjective beauty degree score and objective landscape elements. Based on the coupled quantitative analysis of AI visual recognition and beauty degree score, it is found that the landscape elements that tourists cared most about are: water bodies, ancient buildings and boats. The proportions of the best landscape elements for the spatial sense of the ancient town are the sky ranged from 26.4% to 38.2%, water body ranged from 19.7% to 34.3%, and buildings ranged from 10.4% to 38.2%. This study reveals the pattern of different types of tourists' evaluation of the landscape to summarize the landscape construction strategy of ancient towns in Jiangnan accordingly. The results are not only benefit to the cultural tourism of Wuzhen, but can also be applied to many ancient towns in Jiangnan.

查看原文本刊更多论文

基于人工智能视觉识别与景区美景评价耦合定量分析的古镇风貌建设策略研究

如何营造古镇风貌是古镇建设的关键问题。在这项研究中，收集了50张照片，并通过互联网分发。首先，获得456份在线问卷，25,080份数据。受访者的偏好受性别、年龄、地区、职业和教育程度的影响。其次，将SAM计算机模型应用于乌镇风格照片的图像识别，分析其视觉元素。第三，运用SPSS软件分析主观美度评分与客观景观要素的相关性。基于AI视觉识别与美度评分的耦合定量分析，发现游客最关心的景观要素为：水体、古建筑和船只。对古镇空间感影响最大的景观要素为天空（26.4% ~ 38.2%）、水体（19.7% ~ 34.3%）和建筑（10.4% ~ 38.2%）。本研究揭示了不同类型游客对江南古镇景观评价的格局，并据此总结出江南古镇景观建设策略。研究结果不仅有利于乌镇的文化旅游，也可应用于江南其他古镇。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Frontiers of Architectural Research ARCHITECTURE-

CiteScore

6.20

自引率

2.90%

发文量

430

审稿时长

30 weeks

期刊介绍： Frontiers of Architectural Research is an international journal that publishes original research papers, review articles, and case studies to promote rapid communication and exchange among scholars, architects, and engineers. This journal introduces and reviews significant and pioneering achievements in the field of architecture research. Subject areas include the primary branches of architecture, such as architectural design and theory, architectural science and technology, urban planning, landscaping architecture, existing building renovation, and architectural heritage conservation. The journal encourages studies based on a rigorous scientific approach and state-of-the-art technology. All published papers reflect original research works and basic theories, models, computing, and design in architecture. High-quality papers addressing the social aspects of architecture are also welcome. This journal is strictly peer-reviewed and accepts only original manuscripts submitted in English.