Building Decoration Auxiliary Design Based on BIM Technology

摘要

In view of the problem that pixel analysis method is not suitable for high resolution images information extraction, this paper puts forward a kind of image analysis method based on object for urban construction information extraction. Using multi-resolution image segmentation method to get the image object, put forward the optimal scale of unsupervised decision method, so as to solve the under-segmentation and over-segmentation problem caused by single scale segmentation. In the process of object classification extraction, combined the terrain surface elevation information and spectral information of LiDAR data to extract of buildings, and use the information such as size, space position, etc. for correction of error classification. It extracts 18 constructions on experiment area, and the results show that the proposed method is effective and feasible.