3D Model Triangular Facet Classification Using Deep CNN and Bag-of-Words
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 3D model triangular facet feature learning and classifying techniques using unsupervised deep learning methods fail to ensure accurate feature extraction and classification due to insufficient capability in describing triangular facets, leading to inaccurate results.
Innovation Solution
A deep learning based method utilizing a deep convolutional neural network (CNN) with specific layer configurations and feature reconstruction techniques, including a bag-of-words algorithm and K-means clustering, to enhance the capability of describing 3D model triangular facets and improve classification accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If unsupervised deep learning method is used for 3D model triangular facet feature learning and classifying, then automation is improved, but measurement precision deteriorates
Solution Approach 1:
The patent applies preliminary action by performing feature reconstruction using bag-of-words algorithm and K-means clustering before the main deep learning classification process. This preprocessing step creates more accurate initial features that improve the subsequent supervised learning accuracy while maintaining automation.
Solution Approach 2:
The patent uses feedback by implementing a supervised learning approach where labeled triangular facets are used to train the deep CNN model. The model learns from feedback in the form of labeled examples, continuously improving its feature extraction and classification accuracy through backpropagation and weight adjustment.
2Manufacturing precision
If feature reconstruction using bag-of-words algorithm is applied, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the feature extraction process into distinct stages: initial feature extraction from triangular facets, feature reconstruction using bag-of-words algorithm, and final classification using deep CNN. This segmentation allows each stage to be optimized independently, improving overall precision while managing complexity.
Solution Approach 2:
The patent uses an intermediary approach by introducing K-means clustering as a mediator between the initial feature extraction and the final deep learning classification. The clustering process creates intermediate feature representations that bridge the gap between raw features and classified results, improving description capability while structuring the complexity in a manageable way.
3Measurement precision
If deep CNN model with multiple layers is used, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent applies local quality by designing the deep CNN model with different layer configurations optimized for specific functions: convolutional layers for feature extraction, pooling layers for dimensionality reduction, and fully connected layers for classification. Each layer has specialized structures tailored to its local task, improving overall classification accuracy while managing complexity through functional specialization.
Data Source
AI summary
The invention discloses a deep learning based method for three dimensional (3D) model triangular facet feature learning and classifying and an apparatus. The method includes: constructing a deep convolutional neural network (CNN) feature learning model; training the deep CNN feature learning model; extracting a feature from, and constructing a feature vector for, a 3D model triangular facet having no class label, and reconstructing a feature in the constructed feature vector using a bag-of-words algorithm; determining an output feature corresponding to the 3D model triangular facet having no class label according to the trained deep CNN feature learning model and an initial feature corresponding to the 3D model triangular facet having no class label; and performing classification. The method enhances the capability to describe 3D model triangular facets, thereby ensuring the accuracy of 3D model triangular facet feature learning and classifying results.