Ground three-dimensional laser point cloud segmentation method based on unsupervised fuzzy clustering

The invention discloses a ground three-dimensional laser point cloud segmentation method based on unsupervised fuzzy clustering. The method comprises the following steps: acquiring site ground three-dimensional laser radar point cloud data; the data points are classified into surface categories, the data points are classified into different surface features by calculating feature vectors of the data points, and similar data points are grouped together; classifying the points on the surface features into different planes on the basis of dividing the points into the surface categories, classifying the points on the surface features into different planes on the basis of dividing the points into the surface categories, and distinguishing different plane categories in a single category; plane categories are refined, and on the basis of identifying different plane categories, boundaries or features of plane regions are further refined by refining segmentation of the planes, so that the accuracy of point cloud segme