A three-dimensional model data analysis method for high-precision extraction and fast classification

Abstract

The invention relates to a three-dimensional model data analysis method for high-precision extraction and fast classification, and belongs to the technical field of computer information processing. The method includes: designing a data parsing scheme based on binary tree for three-dimensional model, searching element tags of the model, analyzing the nesting relationship of the data, and forming the simplest document. When a three-dimensional dae model is received, the model data is read, and an analysis algorithm flow is provided. By searching every element mark of the model, the nesting relation of the data is analyzed step by step, the redundant data is eliminated, and then the improved limit learning machine is used to classify the data, and finally the simplest document is formed As that binary tree and the improve limit learning machine simultaneously use the parse three-dimensional model data, the invention reduces the stack of redundant data, accelerates the processing speed ofthe data and improve the parsing efficiency.

Description

technical field [0001] The invention belongs to the technical field of computer information processing, and in particular relates to a three-dimensional model data analysis method. Background technique [0002] The 3D data model is a kind of semi-structured data. Users can arbitrarily set the element tags and the nesting relationship between them. Taking dae as an example, the data model of the dae file is a very important piece of information. Only one The data pattern of a dae file is conducive to identifying the true meaning of the file, and provides a good foundation for subsequent data classification, clustering, and data mining. [0003] So far, many studies have proposed various algorithms for the extraction of 3D model data, which provide important ideas for related research. Jun-Ki Min and others proposed an extraction method based on the element content model, which restricts the element content model. In the element content model, sub-elements can only appear onc...

AI Technical Summary

Problems solved by technology

However, these algorithms also have corresponding shortcomings, which are mainly reflected in two aspects: ① not accurate enough

② Algorithm complexity is high

[0004] There are also many studies on the inductive classification of the data proposed by the model. The traditional machine learning classification models include decision trees, Bayesian, artificial neural networks, support vector machines, etc., but these models suffer from the curse of dimensionality and relatively long training time. Disadvantages of waiting

Images