Among various building information model (BIM) reconstruction methods for existing building, image-based method can identify building components from scanned as-built drawings and has won great attention due to its lower cost, less professional operators and better reconstruction performance. However, this kind of method will cost a great deal of time to design and extract features. Moreover, the manually extracted features have poor robustness and contain less non-geometric information. In order to solve this problem, this paper proposes a deep learning-based method to detect building components from scanned 2D drawings. Taking structural drawings as an example, in this article, 1500 images of structural drawings were firstly collected and preprocessed to guarantee the quality of data. After that, the neural network model—You Only Look Once (YOLO) was trained, verified and tested. In addition, a series of metrics were utilized to evaluate the performance of recognition. The results of test experiments show that the components in structural drawings (e.g., grid reference, column and beam) can be successfully detected, while the average detection accuracy of the whole image is over 80% and the average detection time for each image is 0.71 s. The experimental results demonstrate that the proposed method is robust and timesaving, which provides a good basis for the reconstruction of BIM from 2D drawings.

During data sharing and exchange of building projects, the particular business task generally requires a part of the complete model. This paper adopted XML schema to develop a generic language to extract the partial model from an Industry Foundation Classes (IFC) model based on the proposed Selection Set (called PMESS). In this method, the Selection Set was used to integrate users’ requirements, which could be mapped into IFC data. To ensure the validity of the generated partial IFC models in syntax and semantics, seven rules—including three basic rules for a valid IFC file, three extraction rules based on the Selection Set, and a processing rule for redundant information—were defined. Through defining PMESS-based configuration files, the required data can be extracted and formed as a partial IFC model. Compared with the existing methods, the proposed PMESS method can flexibly extract the user-defined required information. In addition, these PMESS-based configuration files can be stored as templates and reused in other tasks, which prevents duplicated work for defining extraction requirements. Finally, a practical project was used to illustrate the utility of the proposed method.