Honeycomb Body Strength Prediction Through Image Abstraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for predicting the structural characteristics of honeycomb bodies, such as isostatic strength, are often destructive and inefficient, particularly for green ware, and machine learning algorithms struggle to analyze extremely small geometric features in high-resolution images.
Innovation Solution
A method using image abstraction and machine learning algorithms, including convolutional neural networks, to classify honeycomb bodies based on geometric features, generating an abstracted image with reduced resolution to enhance feature detection and prediction of structural characteristics like isostatic strength.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution images are used to capture honeycomb body features, then measurement precision is improved, but processing time and computational resources increase
Solution Approach 1:
The patent extracts only the critical geometric features from the high-resolution image data rather than processing the entire image. By identifying and isolating specific features that correlate with structural characteristics, the system maintains measurement precision while significantly reducing the data volume that requires computational processing, thereby decreasing processing time.
Solution Approach 2:
The patent segments the image analysis process into distinct stages: capturing the full high-resolution image, detecting specific geometric features, abstracting those features into simplified representations, and then analyzing only the abstracted data. This segmentation allows the system to benefit from high-resolution capture while minimizing the computational burden by processing only the essential extracted features.
2Measurement precision
If machine learning algorithms analyze high-resolution images directly, then measurement precision is improved, but productivity is reduced due to computational complexity
Solution Approach 1:
The patent performs preliminary action by detecting and abstracting geometric features from the high-resolution image before the machine learning algorithm processes the data. This pre-processing step transforms the complex high-resolution image into a simplified set of geometric feature representations, allowing the machine learning model to work with reduced computational complexity while maintaining the precision needed for accurate structural characteristic prediction.
Solution Approach 2:
The patent introduces an intermediary abstraction layer between the high-resolution image capture and the machine learning analysis. This intermediary step converts detailed image data into geometric feature representations that serve as a bridge, preserving the essential information needed for accurate prediction while reducing the computational burden on the machine learning algorithm, thereby improving processing speed without sacrificing precision.
3Manufacturing precision
If detailed geometric features are detected in high-resolution images, then manufacturing precision is improved, but device complexity increases
Solution Approach 1:
The patent extracts only the essential geometric features needed for manufacturing precision assessment from the high-resolution images, rather than attempting to analyze all image details. By selectively identifying and isolating specific geometric characteristics that correlate with structural properties, the system achieves high manufacturing precision while avoiding the complexity of processing the entire image dataset.
Solution Approach 2:
The patent creates simplified graphical representations (copies) of the detected geometric features that capture the essential information needed for manufacturing precision assessment. These abstracted feature representations serve as simplified models that maintain the critical geometric information while reducing the complexity of the data structure, making the system more manageable without sacrificing detection accuracy.
Data Source
AI summary
A method and system for inspecting a honeycomb body. The method includes capturing a first image. Instances of at least one feature in the first image that correlates to a structural characteristic of the honeycomb body are detected. One or more detected instances of the at least one feature identified in the first image are abstracted by creating a graphical representation of each of the one or more detected instances of the at least one feature. A second image is generated by augmenting the first image with the graphical representation in place of or in addition to each of the one or more detected instances of the at least one feature identified in the first image. The second image is analyzed using a machine learning algorithm to classify the honeycomb body with respect to the structural characteristic of the honeycomb body.