Automobile radar point cloud data processing method

Abstract

The invention provides an automobile radar point cloud data processing method, which comprises the following steps: S1, marking point cloud original data, selecting a corresponding axle number according to different automobile types, taking one part of the marked point cloud data as training data, and taking the other part of the marked point cloud data as test data; s2, performing model training by using the training data in the S1, firstly aligning the point cloud data and performing deep learning iteration training, then aggregating the extracted feature information, and simultaneously performing deep learning iteration training to obtain a trained model; s3, performing model reasoning by using the test data in the S1 and adopting the model trained in the S2, and comparing the predicted axle type output by the model with the marked truth value to obtain a correct rate; and S4, if the accuracy is greater than the set target accuracy, passing, and if the accuracy is less than or equal to the set target accuracy, increasing the number of the marked point cloud data, and circulating the steps S1-S4 until the accuracy is greater than the set target accuracy.

Description

technical field [0001] The present invention relates to the technical field of point cloud data processing, and more particularly, relates to a processing method of automotive radar point cloud data. Background technique [0002] The radar data of the car is stored in the form of point cloud, and the point cloud data is a collection, which is not sensitive to the order of the data. This means that models dealing with point cloud data need to be invariant to different arrangements of the data. Automobile point cloud data is composed of a certain number of point clouds in a specific space, that is to say, there is a spatial relationship between these point clouds. Objects represented by automotive point cloud data should be invariant to some spatial transformations, such as rotation and translation. Because of the above characteristics, the point cloud classification and recognition algorithm can be used to perform the five-classification task of the wheel and axle. [0003...

AI Technical Summary

Problems solved by technology

[0003] In order to maintain the invariance of different arrangements of data, the methods currently used in the literature include: (1) reordering the unordered data, this method does not directly process the three-dimensional point cloud data, but through projection and other methods, the calculation Complicated; (2) Use all the arrangements of the data for data enhancement and then use the RNN model. This method does not directly process three-dimensional point cloud data, but through projection and other methods, and the calculation is complicated

Images