Abstract: With the rapid development of BIM industry, the application of BIM combined with artificial intelligence has become a prevalent trend. In order to solve the problems of inconvenient carrying apparatus and unavoidable manual errors in building inspection, intelligent inspection glasses based on BIM and machine learning are designed, which enables the process to be portable and automated.Firstly, building component images are trained based on machine learning to realize component recognition; secondly, information of architectural components should be fetched quickly through the visualization program-Dynamo; finally, Raspberry Pi(RPi)is used to design intelligent inspection glasses, and the detection results can accurately identify the types of building components. This method can not only greatly improve the efficiency of inspection but also make the further integration of BIM technology and artificial intelligence realized.