What is the landscape that residents want? -Clarify with a landscape formation support system using artificial intelligence

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K04657
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 22050:Civil engineering plan and transportation engineering-related
• Research Institution: Aichi Institute of Technology
• Principal Investigator: Yamamoto Yoshiyuki 愛知工業大学, 工学部, 准教授 (10580508)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: 景観形成支援システム / 人工知能

Outline of Final Research Achievements

This study aimed to develop a landscape formation support system using artificial intelligence that makes it possible to infer the landscape intention of residents. In order to do that, we selected a deep learning model. Furthermore, we have developed artificial intelligence that recognizes the emotions brought about by images. We have developed a landscape formation support system that can infer landscape intention from the selected deep learning model and artificial intelligence of emotion recognition.
Using Image Inpainting, a deep learning model, we attempted to create a landscape prediction image without utility poles. As a result, we created a landscape prediction image in which the location of the utility pole was harmonized with the surroundings and seemed naturally restored. We also developed an artificial intelligence for emotion recognition from the images (GAPED,etc) with CNN. Finally, we created a landscape formation support system.

Free Research Field

地理空間情報工学

Academic Significance and Societal Importance of the Research Achievements

日本での景観形成は，諸外国と比較して目に余る景観の乱雑さを減らし，美しく調和を保つことに主眼が置かれてきた．かつての景観を無視した行政主導の一方的なプロジェクトへの批判から，近年では，市民の合意形成の位置づけが向上してきた．しかし，合意形成には多くの時間を費やす場合が多く，それにかかる費用や事業の遅れなど重要な問題をはらんでいる．学術的にも，住民の志向の合意を積み重ねるプロセスに関わる研究がほとんどであり，合意形成に要するコストを短縮する方策に力点を置いた研究はわずかであった．これにおいて，本研究で構築した景観形成支援システムは，合理的な合意形成支援において学術的・社会的意義を有す．