Computer Vision Based Method for Identifying Grouting Defects of Prefabricated Building Sleeves

In order to accurately detect the grouting defects of construction sleeves, a method for identifying the grouting defects of prefabricated construction sleeves based on computer vision is proposed. The 3D target detection algorithm is used for feature point detection, and the corner detection model of assembled building is constructed to obtain the feature angle detection results and match the feature angles. The constraint relationship is extracted from the data in the image sequence of the grouting defects of the prefabricated construction sleeve, and the parameters of the filling defects of the construction sleeve are calibrated. Finally, the defect identification is completed through the sleeve grouting scanning of computer vision. The experimental results show that the method based on computer vision to identify the grouting defects of prefabricated construction sleeve has strong recognition ability and can well complete the defect identification.