An Intelligent Learning Algorithm for Improving BIM Object Classification and Recognition  

一种改进BIM对象分类和识别的智能学习算法

作　　者：WANG Ru BENMANSOUR Oussama XING Ying 王茹;BENMANSOUR Oussama;邢颖(西安建筑科技大学土木工程学院,陕西西安710055;Trine University)

机构地区：[1]School of Civil Engineering,Xi’an University of Architecture and Technology,Xi’an,Shaanxi 710055,China [2]Trine University

出　　处：《施工技术（中英文）》2024年第20期86-93,共8页Construction Technology

基　　金：陕西省杰出青年科学基金(2018JC-025);西安市社发科技创新项目(23SFSF0013)。

摘　　要：Building information modeling(BIM)object classification takes a lot of time and energy.Misclassification or omission of any object may lead to the emergence of abnormal results,which have a great impact on the project workflow and results.Roundly understanding BIM object classification,by improving Swin Transformer classifier algorithm parameters,using the model primitives extracted from IFC format BIM model file,deep learning of 7 types of BIM object categories is taken.Through the performance and evaluation indicators obtained in training,the results improve the classification accuracy.建筑信息模型(BIM)对象分类需耗费大量时间和精力,任何对象的错误分类或遗漏均可能导致异常结果的出现,从而对项目工作流程及结果产生极大影响。全面理解BIM对象分类,通过改进Swin Transformer分类器算法参数,使用从IFC格式BIM模型文件中提取的模型图元,对7类BIM对象类别进行深度学习,通过训练中得到性能和评估指标的结果,提高分类准确性。

关 键 词：building information modeling(BIM) object classification deep learning model primitive performance 

分 类 号：TU17[建筑科学—建筑理论]