Aviation part point cloud denoising method based on deep learning

Abstract

The invention discloses an aviation part point cloud denoising method based on deep learning. The method comprises the steps of: artificially increasing different degrees of Gaussian noise based on atheoretical data model of an aviation part, generating a height map for each point in the model, and constructing a deep learning training set; training a deep learning network based on the constructed deep learning training set to obtain a deep learning network model used for predicting normal information of each point in the point cloud data; scanning a real aviation part by a laser scanner to obtain actually measured point cloud data, and predicting normal information of the actually measured point cloud based on the trained deep learning network model; and further performing position updating on each point in the point cloud based on the predicted normal information, thereby realizing de-noising of the actually measured point cloud. According to the invention, the problem that the noise of scanned point cloud of an aviation part cannot be accurately and efficiently removed in the prior art is solved, and the efficiency and accuracy of aviation part detection and analysis are further improved.

Description

technical field [0001] The invention belongs to the technical field of aviation detection, and in particular relates to a method for denoising point clouds of aviation parts based on deep learning. Background technique [0002] In recent years, with the continuous improvement of domestic aviation research projects in technical combat indicators such as flight speed, stealth, and maneuverability, the structure of the aircraft itself has shown a clear trend of integration, lightweight, compactness, and precision. Aviation structural parts with complex structures and surface features have been widely used. The precision requirements of aviation structural parts are also increasing, and the requirements for assembly are gradually increasing, which makes the control and inspection of the manufacturing and processing quality of aviation parts particularly important. At the same time, as the use time of the aircraft increases, the wear and deformation of various parts will greatly ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a method for denoising the point cloud of aviation parts based on deep learning to solve the problem that the point cloud of aviation parts cannot be accurately and efficiently denoised in the prior art

Images