Three-dimensional point cloud denoising method and system for complex underground roadway scene

The invention provides a three-dimensional point cloud denoising method and system for a complex underground roadway scene, and relates to the technical field of signal processing, and the method comprises the steps: obtaining three-dimensional roadway point cloud data of an underground roadway scene; calculating a normal vector of each data point in the three-dimensional roadway point cloud data by adopting a plane fitting method to obtain a normal vector set; according to the normal vector set, calculating the direction of the roadway axis by adopting a Gaussian mapping method; determining the type of each data point according to the direction of the roadway axis and the normal vector set, adding the data points of which the types are noise points into a noise point set, and adding the data points of which the types are roadway points and noise points into a roadway point set to obtain a noise point set and a roadway point set, the noise point set and the roadway point set can be further optimized, and the