Towards automatic flatness quality assessment for building indoor acceptance via terrestrial laser scanning

For building indoor acceptance, flatness is an important metric. Traditionally, the flatness quality assessment (FQA) relies on manual labor, but manual inspection is random, error-prone, inefficient and non-repeatable results. With the development of terrestrial laser scanning (TLS) technology, it is becoming possible to resolve the above issues. In this paper, a fully automatic building indoor acceptance approach for FQA via TLS is proposed, which consists of indoor segmentation and FQA. First, the indoor segmentation is performed using the geometric features of the indoor point cloud. Then, the FQA is carried out using the two-dimensional continuous wavelet transform and simulated manual method. Manual and simulated inspections are performed on the same inspection points. The results show that the simulated manual method is rigorous than the manual method. Furthermore, the validity and efficiency of the proposed approach is proved by running two tests on point cloud data obtained from two as-built dwellings.