Individual stem detection in residential environments with MLS data
Published in Remote Sensing Letters, 2017
Recommended citation: Xu Sheng, Ye Ning and Xu Shanshan(*). "Individual stem detection in residential environments with MLS data." Remote Sensing Letters. vol. 9(1), pp.51-60 ,2018, doi: 10.1080/2150704X.2017.1384588. https://www.tandfonline.com/doi/full/10.1080/2150704X.2019.1569277
Nowadays, mobile laser scanning (MLS) system succeeds to collect plentiful side information of roadside trees. This letter aims to propose a method for the individual stem detection in residential environments using mobile LiDAR point clouds. The first step is to use the proposed point removal method for filtering ground points based on the elevation histogram. The second step is to localize trees by using a circle fitting method on the projection points of trunks. The third step is to use the voxel-based representation approach to organize points and calculate the voxel value for the following optimization. The last step is to minimize the formulated stem energy function by the dynamic programming technique. The main contribution of our work is that the cost of finding a stem is calculated by a penalty function and minimized by a dynamic programming model. Test on residential environments shows that our method achieves the completeness of 94.2% and correctness of 95.7%, which are competitive results in the stem detection.
Recommended citation: ‘S. Xu, N. Ye, S. Xu(*). "Individual stem detection in residential environments with MLS data." Remote Sensing Letters. vol. 9(1), pp.51-60, 2018.’