Point cloud data classification method, device and equipment and readable storage medium

Abstract

The invention discloses a point cloud data classification method, device and equipment, and a readable storage medium. The method comprises the steps of extracting a point cloud feature vector of preset point cloud data; performing dimension raising on the point cloud feature vector based on a preset dimension raising algorithm to obtain a point cloud feature vector after dimension raising; and classifying the point cloud feature vectors after dimension raising to obtain a classification result. The point cloud feature vectors are subjected to dimension raising through the preset dimension raising algorithm, the point cloud feature vectors subjected to dimension raising are obtained, the dimension of the point cloud feature vectors subjected to dimension raising is higher than that of thepoint cloud feature vectors, and therefore input features during classification of the point cloud data are enriched, the feature representation capability of the point cloud feature vector after dimension raising is stronger than the feature representation capability of the point cloud feature vector, so that the accuracy of a point cloud data classification result is improved.

Description

technical field [0001] The present invention relates to the technical field of point cloud data processing, in particular to a point cloud data classification method, device, equipment and readable storage medium. Background technique [0002] With the improvement of computing power and the substantial increase of data volume, deep learning technology has achieved breakthrough development in recent years, and has achieved remarkable results in computer vision, natural language processing and other fields. In terms of computer vision, in image classification, face recognition, object detection, semantic segmentation and other tasks, deep neural networks have achieved results that exceed traditional methods and human capabilities. In recent years, GNN (Graph Neural Network, graph neural network) has developed rapidly. Data such as two-dimensional digital images, social networks, and three-dimensional point sets can be expressed in the form of graphs, and the powerful fitting a...

AI Technical Summary

Problems solved by technology

[0005] The main purpose of the present invention is to provide a point cloud data classification method, device, equipment and readable storage medium, aiming at solving the technical problem of low accuracy of the existing point cloud data classification results

Images