Automated computer vision-based detection of components of under-construction indoor partitions

详细信息    查看全文

• 作者：Hesam Hamledari ; ^{h.hamledari@mail.utoronto.ca} ; Brenda McCabe ^{brenda.mccabe@utoronto.ca} ; Shakiba Davari ^{sdavari@cs.toronto.edu}
• 关键词：Computer vision ; Interior construction ; Machine learning ; Image processing ; Digital images ; Indoors
• 刊名：Automation in Construction
• 出版年：2017
• 出版时间：February 2017
• 年：2017
• 卷：74
• 期：Complete
• 页码：78-94
• 全文大小：3386 K
• 卷排序：74

文摘

Four novel vision-based algorithms were introduced to detect partition components. Studs, electrical outlets, insulation, and three states for drywall were detected. The modules were validated using three image databases (UAV, smartphone, and online). The image processing results can be used in progress tracking systems. The modules can also operate robustly in automated data collection contexts.