Rubber Composition Image Segmentation for Accurate Formulation Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques fail to accurately define regions corresponding to formulations in microscopic images of rubber compositions, making it difficult to extract quantitative data on physical properties and structure.
Innovation Solution
A machine learning-based analysis method using a trained model, such as a TransUNet, to segment and define regions in microscopic images, enabling precise region division and feature extraction of rubber compositions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image analysis techniques are used to calculate indices from microscopic images, then processing simplicity is maintained, but the ability to accurately define regions corresponding to formulations is insufficient
Solution Approach 1:
The patent applies segmentation by dividing the microscopic image into distinct regions corresponding to different formulations. The trained machine learning model segments the image to identify and separate regions of interest, enabling accurate definition of formulation-specific areas for subsequent analysis.
Solution Approach 2:
The trained machine learning model serves as an intermediary between the raw microscopic image and the region definition process. Instead of directly analyzing pixel data, the model acts as a mediator that interprets image features and outputs precise region boundaries, resolving the contradiction between accuracy and complexity.
2Measurement precision
If machine learning models are introduced to define regions in microscopic images, then region definition accuracy is improved, but processing complexity increases
Solution Approach 1:
The machine learning model is trained in advance on a dataset of microscopic images with labeled regions. This preliminary training phase allows the model to learn formulation patterns and region characteristics before actual analysis, so that during operation, region definition can be performed accurately without real-time complexity.
Solution Approach 2:
The patent uses copied and transformed versions of input images (e.g., different color spaces, scaled versions, or augmented variants) as intermediate representations for the machine learning model. This allows the model to process multiple views of the same data, improving region definition accuracy while keeping the base processing pipeline manageable.
3Reliability
If quantitative data extraction is attempted without proper region definition, then processing speed is maintained, but the reliability of physical property estimation deteriorates
Solution Approach 1:
The patent replaces manual or rule-based region definition methods with a machine learning-based system. This substitution automates the complex task of identifying formulation regions, improving reliability of subsequent physical property estimates while managing complexity through algorithmic approaches rather than manual intervention.
Data Source
Figure 1A~1B
Figure 1C
Figure 2
AI summary
Provided is an analysis method and the like for appropriately defining a region corresponding to a formulation contained in a rubber composition in a microscopic image of the rubber composition. An analysis method for a rubber composition includes the followings: acquiring input data generated from a microscopic image obtained by image-capturing a rubber composition with a microscope, the microscopic image showing a formulation contained in the rubber composition; inputting the input data to a trained machine learning model; and deriving output data from the trained machine learning model. Note that the output data is data defining a region that appears in the microscopic image corresponding to the formulation.