| Title |
Scan Matching based De-skewing Algorithm for 2D Indoor PCD captured from Mobile Laser Scanning |
| Authors |
Kang, Nam-woo ; Sa, Se-Won ; Ryu, Min Woo ; Oh, Sangmin ; Lee, Chanwoo ; Cho, Hunhee ; Park, Insung |
| DOI |
https://dx.doi.org/10.6106/KJCEM.2021.22.3.040 |
| Keywords |
Floor plan; Mobile Laser Scanning; Light Detection and Ranging; Point Cloud Data; De-skewing |
| Abstract |
MLS (Mobile Laser Scanning) which is a scanning method done by moving the LiDAR (Light Detection and Ranging) is widely employed to capture indoor PCD (Point Cloud Data) for floor plan generation in the AEC (Architecture, Engineering, and Construction) industry. The movement and rotation of LiDAR in the scanning phase cause deformation (i.e. skew) of PCD and impose a significant impact on quality of output. Thus, a de-skewing method is required to increase the accuracy of geometric representation. De-skewing methods which use position and pose information of LiDAR collected by IMU (Inertial Measurement Unit) have been mainly developed to refine the PCD. However, the existing methods have limitations on de-skewing PCD without IMU. In this study, a novel algorithm for de-skewing 2D PCD captured from MLS without IMU is presented. The algorithm de-skews PCD using scan matching between points captured from adjacent scan positions. Based on the comparison of the deskewed floor plan with the benchmark derived from TLS (Terrestrial Laser Scanning), the performance of proposed algorithm is verified by reducing the average mismatched area 49.82%. The result of this study shows that the accurate floor plan is generated by the de-skewing algorithm without IMU. |