Automated Activity Identification and Productivity Improvement in Construction using Computer Vision

使用计算机视觉进行施工中的自动活动识别和生产力提高

• 批准号: 203368-2012
• 负责人: McCabe, Brenda
• 金额: $ 1.53万
• 依托单位: University of Toronto
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2015
• 资助国家: 加拿大
• 起止时间: 2015-01-01 至 2016-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=569938
• 关键词: Automated; Activity; Identification; Productivity; Improvement

Construction is regularly cited as lagging behind other industries in its effort to automate operations and improve productivity. Some of the challenges that contribute to this lapse are the rugged environment in which construction takes place, and the expanse of the site boundaries. The objective of this research is to create automated systems to track construction activities with a minimum of human intervention. This will facilitate the development of application modules that can: - Identify operational bottlenecks and use innovative optimization routines to suggest improvements - Identify inefficiencies and unnecessary idling to increase profitability and reduce emissions from equipment - Extend the technology to include night vision and operations in emergency zones Challenges in the research include ensuring that the technologies can withstand the the rugged environment of application areas (e.g. construction and open pit mining), and developing resilient algorithms that can self-correct due to camera movement or errors in identification. Ultimately, the system may be used to control unmanned equipment fleets in extreme environments.

建筑业经常被认为在自动化操作和提高生产力方面落后于其他行业。造成这种失误的一些挑战是施工环境的恶劣和场地边界的广阔。本研究的目的是创建自动化系统，以最少的人为干预来跟踪施工活动。这将有助于开发应用模块，这些模块可以： - 识别运营瓶颈并使用创新的优化程序提出改进建议 - 识别低效率和不必要的空转，以提高盈利能力并减少设备排放 - 将该技术扩展到包括夜视和紧急区域的操作 研究中的挑战包括确保这些技术能够承受应用领域（例如建筑和露天采矿）的恶劣环境，以及开发弹性算法，可以由于相机移动或识别错误而自我纠正。 最终，该系统可用于在极端环境中控制无人驾驶设备队。