Three-dimensional scanning modeling system for non-standard parts

Abstract

The invention discloses a three-dimensional scanning modeling system for non-standard parts, and relates to the technical field of paint spraying. The three-dimensional scanning modeling system for the non-standard part comprises applying a particle swarm algorithm to point cloud data, performing kernel density estimation on the optimal broadband, carrying out mean shift algorithm operation on thedensity function; removing outliers, performing weighted principal component analysis operation on the preprocessed point cloud data; performing operation of a three-dimensional contour surface extraction algorithm on the secondary processing point cloud data; carrying out contour surface extraction, component construction and optimization; obtaining a three-dimensional digital analog, carrying out first point cloud data processing through combination of a particle swarm algorithm and a mean shift algorithm; and then carrying out second point cloud data analysis through weighted principal component analysis, so as to enable a formed file to be more accurate and practical. Point cloud data obtained through scanning is directly utilized to form a digital model, establishment of a frameworkis avoided, a modeling process is simplified, a processing result is ensured, and doping of external impurities is avoided.

Description

technical field [0001] The invention relates to the technical field of paint spraying, in particular to a three-dimensional scanning modeling system for non-standard parts. Background technique [0002] The current 3D scanning technology is mostly used for parts inspection. Reverse modeling technology also requires the intervention of a large number of professionals. It uses professional knowledge to perform multiple processing on 3D point clouds, and the operation is cumbersome. Due to the limitations of the method and principle of 3D scanning, the scanning effect of objects with different shapes and structures cannot be guaranteed. Therefore, the quality of the generated 3D point cloud is also uneven, which leads to the problem of large amount of work in the later manual processing. [0003] At present, it is difficult to realize the overall automation of 3D scanning modeling, which still requires human intervention, and the intermediate process is separated. Although so...

AI Technical Summary

Problems solved by technology

But obviously these algorithms are not suitable for multi-variety, small-batch non-standard parts, and it is difficult to meet the quality requirements of 3D reconstruction for point cloud through the processed effect

In addition, for automatic modeling, there is currently a processing method that uses format conversion software to automatically call point cloud files and convert them into 3D digital model files, but the conversion success rate and conversion effect for multi-variety and small-batch non-standard parts are not ideal

Images