Three-dimensional oral cavity model Angle's classification method based on multi-view convolutional neural network

Abstract

The invention relates to a three-dimensional oral cavity model Angle's classification method based on a multi-view convolutional neural network, and the method comprises the following steps: 1, collecting Angle's classification data, extracting the three-dimensional grid data of teeth at a plurality of preset positions for each digital three-dimensional oral cavity model, obtaining two groups of Angle's classification triangular grid data through combination, and acquiring a corresponding occlusion dislocation category; 2, generating a multi-view two-dimensional diagram based on Angle's classification triangular mesh data; 3, preprocessing the generated multi-view two-dimensional diagram to obtain a preprocessed multi-view two-dimensional diagram; 4, based on the preprocessed multi-view two-dimensional graph, constructing an ANNET model of the oral cavity digital model; 5, training the ANNET model based on the multi-view two-dimensional graph, and after the training is completed, performing ANNET prediction on the input three-dimensional oral cavity model by using the ANNET model.

Description

technical field [0001] The invention belongs to the field of image processing, in particular to a three-dimensional oral cavity model Angle's classification method based on a multi-view convolutional neural network. Background technique [0002] Angle's classification was proposed by Dr. Angle in 1899 and is currently the most widely used classification of malocclusions. Angle believes that the maxilla is fixed on the skull, and its position is relatively constant. The upper first permanent molar is located under the maxillary zygomatic process, which is stable and not easy to dislocate. Therefore, based on the above first permanent molar, according to the anteroposterior relationship between the upper and lower dental arches, Malocclusions were divided into 3 categories. In the prior art, because the classification method needs to consider the relationship and position of the teeth to identify and judge the teeth, if the Angle classification of the teeth in the oral cavity...

AI Technical Summary

Problems solved by technology

In the prior art, because the classification method needs to consider the relationship and position of the teeth to identify and judge the teeth, if the Angle classification of the teeth in the oral cavity is to be performed, the identification and judgment are usually performed manually against the images of the oral teeth, and the efficiency is low. And it is easy to make mistakes when negligent, and it is still impossible to realize computer automated intelligent judgment and classification recognition

Images