Image Segmentation Applied to Urban Surface and Aerial Constraints Analysis

Author:

Trani Marco Lorenzo1,Madaschi Federica1

Affiliation:

1. Politecnico di Milano, IT

Abstract

The rapid progress of artificial intelligence (AI) has prompted the exploration of its potential applications in the construction industry, although at a slower rate. Since the starting point of a design is the analysis of the site’s constraints, the purpose of the ongoing research is the application of artificial intelligence in risk assessment for site areas. The primary objective of this research project is to develop an interactive map that employs AI to identify potential surface and aerial interferences. This map aims to support planners, engineers, and architects during the site context analysis phase by providing real-time visualization of obstacles. The interactive map allows users to explore and analyze identified obstacles, enabling cluster markers and filtering of features. The results obtained from applying this approach in Milan, Italy, demonstrate its functionality and usability, highlighting the tool's ability to provide valuable information in both localized and citywide scenarios. Potential improvements such as size assessment and advanced marker generation are also being examined to enhance the management of surface and air interferences. The goal is to enhance the tool's functionality, accuracy, and planning efficiency in construction projects

Publisher

Firenze University Press

Reference9 articles.

1. Couto, João & José, Frederico & Santos, Barros & Rahnemay, Emilia & Kohlman Rabbani, Emilia. (2017). How Time Constraints Affect Safety Conditions at Construction Sites: Analysis of the Perception of Portuguese Construction Participants. EJGE. 2. 563.

2. Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., & Houlsby, N. (2020). An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale. http://arxiv.org/abs/2010.11929

3. A Tutorial on Geographic Information Systems: A ten-year update

4. Liu, Y. (2021). Efficient Fully Convolutional Networks for Dense Prediction Tasks. [University of Adelaide]. https://hdl.handle.net/2440/134023

5. Fully convolutional networks for semantic segmentation

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3