Scholarlink

<p>Urban residents' visual perception of greenery contributes to enhancing their happiness index. However, in the complex visual environment of the urban street, street signboards occupy residents’ attention, thereby affecting their visual perception of greenery. To address this issue, this study is based on street-view images from 1009 sampling points in the study area, utilizing the saliency detection model (MSI-Net) to simulate human visual observation combined with the semantic segmentation model (DeepLabv3) to calculate the average saliency of greenery in the street-view images. Subsequently, statistical methods are employed to analyze the impact of the presence or absence of street signboards, as well as the size, color, and position of the signboards on the saliency of greenery. The results indicate: (1) there is a significant difference in greenery saliency between the signboard group and the no-signboard group (<i>p</i> &lt; 0.01). (2) The average greenery saliency of the signboard group is lower than that of the no-signboard group (<i>t</i> &lt; 0). (3) The proportion of blue pixels and the size of signboards have a significant negative impact on greenery saliency (<i>p</i> &lt; 0.01, <i>β</i> &lt; 0). The research methods and results offer references for urban designers to improve the quality of greenery.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Visual conflict in urban landscape design: A study on the influence of street signboards on the saliency of greenery

  • 期刊论文

  • 发表日期：

    2025年10月7日

查看全文
  • Qianlong Shi,
  • Jing Zhang,
  • Xinping Zhang,
  • Yan Wei

摘要

Urban residents' visual perception of greenery contributes to enhancing their happiness index. However, in the complex visual environment of the urban street, street signboards occupy residents’ attention, thereby affecting their visual perception of greenery. To address this issue, this study is based on street-view images from 1009 sampling points in the study area, utilizing the saliency detection model (MSI-Net) to simulate human visual observation combined with the semantic segmentation model (DeepLabv3) to calculate the average saliency of greenery in the street-view images. Subsequently, statistical methods are employed to analyze the impact of the presence or absence of street signboards, as well as the size, color, and position of the signboards on the saliency of greenery. The results indicate: (1) there is a significant difference in greenery saliency between the signboard group and the no-signboard group (p < 0.01). (2) The average greenery saliency of the signboard group is lower than that of the no-signboard group (t < 0). (3) The proportion of blue pixels and the size of signboards have a significant negative impact on greenery saliency (p < 0.01, β < 0). The research methods and results offer references for urban designers to improve the quality of greenery.