Vehicle-mounted road scene point cloud automatic classification method based on deep learning

Abstract

The invention discloses a vehicle-mounted road scene point cloud automatic classification method and system based on deep learning, and the method comprises the steps: building a training sample set:collecting the three-dimensional point cloud data of a road scene vehicle-mounted laser, labeling a category label, and packaging the three-dimensional point cloud data and corresponding category label information as a point cloud data set; Designing a three-dimensional deep learning network model based on an octree data structure, wherein the three-dimensional deep learning network model adopts aU-shaped full convolutional neural network model; A loss function for weighting each category is utilized to solve the problem that the number difference of different categories of points of the vehicle-mounted laser point cloud of the road scene is large, a network model is trained based on a training sample set, and a trained point cloud classification model is obtained; And inputting to-be-classified vehicle-mounted laser three-dimensional point cloud data. By adopting the technical scheme of the invention, the laser point cloud under the vehicle-mounted road scene with different types, numbers and sizes can be automatically classified stably and accurately.

Description

technical field [0001] The invention belongs to the field of automatic classification of laser point clouds of vehicle-mounted road scenes, and in particular relates to an automatic classification method of laser point clouds of vehicle-mounted road scenes based on deep learning. This method learns and trains the 3D laser point cloud data set that has been marked with category labels, and the deep learning model obtained through the final training can stably and accurately classify the laser beams in the vehicle-mounted road scene with different types, quantities and sizes. Point clouds are automatically classified. Background technique [0002] Laser point cloud automatic classification technology refers to the process of automatically classifying each three-dimensional point into a specific category by using appropriate feature extraction, learning and classification methods for the original laser point cloud with information such as three-dimensional space coordinates. E...

AI Technical Summary

Problems solved by technology

However, this method projects three-dimensional space points to a series of two-dimensional planes, which will inevitably lose a certain amount of spatial detail information.

At the same time, there are still a series of problems such as projection viewing angle selection and self-occlusion.

Therefore, it cannot fundamentally solve the difficult problem of using deep learning to process three-dimensional data

Images