TEM Image Crystallinity Mapping Using Pixel Variation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for evaluating the crystallinity of materials in transmission electron microscope (TEM) images, such as using image brightness values or deep learning approaches, are inadequate as they fail to accurately distinguish between high and low crystallinity regions, especially when regions have similar brightness values, and require extensive labeled training data for neural networks.
Innovation Solution
An information processing device and method that calculates the standard deviation of pixel values in partial regions of TEM images to determine the degree of crystallinity, allowing for precise visualization of crystallinity through discrete pixel values, without the need for labeled training data or relying solely on brightness values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If image brightness values are used to evaluate crystallinity, then the evaluation process is simple, but regions with similar brightness values cannot be distinguished by crystallinity
Solution Approach 1:
The patent changes the evaluation parameter from brightness values to pixel value variations (standard deviation). By calculating the standard deviation of pixel values within partial regions, the method can distinguish crystallinity differences even when brightness values are similar, as crystalline regions exhibit different pixel value variations compared to amorphous regions
Solution Approach 2:
The patent replaces the simple brightness thresholding mechanism with a statistical analysis mechanism (standard deviation calculation). This substitution enables the system to capture subtle differences in pixel value distributions that indicate crystallinity, overcoming the limitation of brightness-based methods
2Measurement precision
If deep learning approaches are used to determine crystallinity, then crystallinity can be determined, but extensive labeled training data is required
Solution Approach 1:
The patent replaces the expensive and time-consuming deep learning model with a simple, computationally inexpensive statistical calculation (standard deviation). This approach achieves crystallinity determination without requiring extensive labeled training data, making the method more accessible and easier to implement
Solution Approach 2:
The method uses the inherent pixel value variations within the image data itself to determine crystallinity, without requiring external training data. The standard deviation calculation naturally captures the statistical properties of crystalline versus amorphous regions, enabling self-sufficient crystallinity evaluation
Data Source
AI summary
An information processing device acquires an image captured by a transmission electron microscope. The information processing device, for each partial region in the image, calculates a variation in pixel values of pixels included in the partial region. The information processing device, for each partial region in the image, determines a degree of crystallinity of the partial region based on the calculated variation in the pixel values of the partial region.