The invention relates to the technical field of
point cloud de-noising, and discloses a substation equipment
laser radar point cloud de-noising method based on space
grid density, comprising the following steps: S1, acquiring
point cloud data, and preprocessing the point
cloud data; s2, analyzing point cloud features, and performing point cloud type division by utilizing the features of equipment and ground point cloud on elevation distribution; and S3, removing
noise points through a
density difference method by using the
relative distribution characteristics of the equipment point cloud and the ground point cloud with the
noise. The multi-dimensional subspace
grid density difference segmentation method based on the three-dimensional
laser point cloud suitable for the extra-
high voltage transformer substation is provided for overcoming the defects that an existing method is too high in dependence degree on a
big data technology, and the method can efficiently and accurately remove noisy three-dimensional point
cloud data according to different scanning
modes and different data types, so that the accuracy of the three-dimensional subspace
grid density difference segmentation method based on the three-dimensional
laser point cloud is improved. And on the basis of
noise processing of massive three-dimensional point
cloud data, automatic extraction of the extra-
high voltage substation equipment can be completed more accurately and efficiently.