An intelligent identification and retrieval method for three-dimensional processing characteristics

Abstract

The invention discloses an intelligent identification and retrieval method for three-dimensional processing characteristics. The method comprises the following steps: firstly, obtaining a processing characteristic three-dimensional model; Performing voxelization processing on the three-dimensional model; Performing data enhancement processing on the voxelization model to obtain a three-dimensionalmodel data set; Then constructing a three-dimensional convolutional neural network, training a three-dimensional model and recording model parameters; Then inputting a three-dimensional model, and performing voxelization processing on the three-dimensional model of the processing characteristics; And calculating the similarity between the retrieval model and the data set model according to the recorded model parameters to obtain the probability that the model belongs to each category, and finally arranging and outputting a retrieval result according to a descending order. The method is smallin calculated amount and high in recognition speed, and has higher three-dimensional processing feature recognition and retrieval accuracy.

Description

technical field [0001] The invention belongs to the technical field of deep learning and three-dimensional model intelligent identification and retrieval, and specifically relates to an intelligent identification and retrieval method for three-dimensional processing features. Background technique [0002] In recent years, with the development of the manufacturing industry, the development mode of enterprises has changed from mass production to multi-variety and small-scale production, and the types of processing features of parts and components have increased and their shapes are complex and changeable. In addition, with the increasing maturity of 3D modeling technology, the number of 3D models of parts has shown an explosive growth trend, and 3D process design technology has gradually been applied in enterprises. In the process design of traditional machining parts, technicians often rely on coding technology or empirical knowledge to identify the shape and size of parts pr...

AI Technical Summary

Problems solved by technology

However, the existing 3D convolutional neural network architecture needs to learn many parameters, the number of network layers is small, the calculation speed is slow, and overfitting may occur. There are still many areas for improvement in the efficient and accurate identification and retrieval of

Images