Recognizing Symbols in Building Mechanical Drawings for Automated Building Information Models Reconstruction

60~85% of the total lifecycle cost of the building occurs after construction is completed. During the operation and maintenance (O&M) phase, every year 20 billion dollars are lost in the US due to inadequate information... [ view full abstract ]

60~85% of the total lifecycle cost of the building occurs after construction is completed. During the operation and maintenance (O&M) phase, every year 20 billion dollars are lost in the US due to inadequate information access. Among O&M activities, information-related activities account for 12% of the total maintenance time. The previous research studies have demonstrated that Building Information Models (BIM), if available with detailed asset information and properties, can enable rapid troubleshooting and execution of O&M tasks by providing the needed facility information. However, currently, there is still rarely BIM with Mechanical, Electrical, and Plumbing (MEP) assets and properties that are available for O&M. If BIM is not in possession, generating BIM manually is a time-consuming process. Hence, there is a need for an automated approach that can reconstruct the MEP systems in BIM. Previously research studies investigated automatic reconstruction of BIM using architectural drawings, structural drawings, or the combination with photos. But most of the previous studies focus on reconstructing the architectural and structural components. MEP systems are relatively more complex due to the difficulty in recognizing various symbols that are used to represent the different types of components. In order to address this challenge, the study discussed in this paper proposed a symbol recognition framework that can automatically recognize different types of symbols in the building mechanical drawings. The proposed approach utilizes vector-based computer vision techniques to identify various types of mechanical symbols and their properties (e.g., location, type, etc.) from drawings. With this automated approach, not only locating the mechanical component of interest but merging the updated information into the current BIM would be possible in the future and reduce the time of repeating the manual creation of BIM after every renovation project.