Scholarlink

As cities grow and develop, their cultural heritage is at risk of being lost due to modernization. Historic areas are losing their unique characteristics, making it crucial to implement systematic color planning to preserve their cultural significance. Japan has been a pioneer in urban color planning, with a systematic approach that is worth studying. This paper focuses on seven merchant towns of Japan’s Preservation Districts, exploring their color structure and tourists’ Internet reviews. The study uses image semantic segmentation to identify buildings, roads, and landscapes, extracting their color data for analysis. Natural Language Processing is used to extract Internet reviews for keyword analysis, which helps to identify correlations between color psychology and keywords. The study also uses Generative Adversarial Networks to generate predicted images of color structure improvements and color planning. The methods and conclusions of this study can provide innovative ideas for the digital management of environmental colors in historic districts, contributing to the preservation of local traditional building culture.

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

Based on Deep Learning Methods to Preserve the Traditional Building Color of Historic Districts

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Dongqi Li,
  • Yunjia Huang,
  • Tomo Inoue,
  • Kohei Inoue,
  • Qiaoling Fang

摘要

As cities grow and develop, their cultural heritage is at risk of being lost due to modernization. Historic areas are losing their unique characteristics, making it crucial to implement systematic color planning to preserve their cultural significance. Japan has been a pioneer in urban color planning, with a systematic approach that is worth studying. This paper focuses on seven merchant towns of Japan’s Preservation Districts, exploring their color structure and tourists’ Internet reviews. The study uses image semantic segmentation to identify buildings, roads, and landscapes, extracting their color data for analysis. Natural Language Processing is used to extract Internet reviews for keyword analysis, which helps to identify correlations between color psychology and keywords. The study also uses Generative Adversarial Networks to generate predicted images of color structure improvements and color planning. The methods and conclusions of this study can provide innovative ideas for the digital management of environmental colors in historic districts, contributing to the preservation of local traditional building culture.