Material Structure Data Upscaling for Nanoscale Feature Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for measuring the minute shape and structure of substances are limited by hardware restrictions, preventing a detailed understanding of features on a narrower scale.
Innovation Solution
An information processing device and method that utilize a data generation means and learning model to convert macroscale feature data into atomic-scale data, using algorithms like self-affinity and fractal laws to enhance resolution and remove noise, enabling a better index score for material structure analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional measurement techniques are used, then measurement capability is limited by hardware restrictions, but the ability to grasp features on narrower scale is insufficient
Solution Approach 1:
The invention creates a virtual copy of the measurement system through software simulation. A probe model replicates the characteristics of the physical probe, and measurement data is processed through algorithms that simulate higher-resolution measurements without requiring actual hardware upgrades. This allows features on narrower scales to be grasped by computing rather than physical measurement.
Solution Approach 2:
The invention replaces the mechanical measurement system with an information processing system. Instead of relying on physical probes and hardware to achieve higher resolution, the system uses data processing algorithms, machine learning models, and computational methods to enhance measurement capability and recover features that would otherwise be lost due to hardware limitations.
2Device complexity
If hardware restrictions are accepted, then device complexity is reduced, but the ability to measure minute features is compromised
Solution Approach 1:
The invention substitutes complex hardware requirements with software-based solutions. By using data processing algorithms, the system achieves high measurement precision without requiring sophisticated physical measurement devices. The complexity is shifted from the mechanical domain to the computational domain, where it can be managed through software development rather than hardware engineering.
Solution Approach 2:
The invention changes the parameters of the measurement system from physical hardware specifications to data processing parameters. Instead of improving resolution through better probe design or higher measurement precision hardware, the system adjusts computational parameters such as algorithm selection, data sampling rates, and processing methods to achieve the desired measurement capability.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The objective of the present invention is to provide a technique with which it is possible for a feature of a fine substance obtained on a certain scale to be grasped on a smaller scale. A server 1 is provided with a data generating means including a data generating unit 31 and a learning model 41. With regard to a statistical index indicating a characteristic of a material, the learning model 41 uses a predetermined algorithm that generates output data indicating the structure of a material, the output data exhibiting an index score that is better than that of input data indicating the structure of the material. The data generating unit 31 inputs structural data (for example, data relating to the macro-scale uneven shape of a material surface) indicating a structure of a predetermined material as input data into the learning model 41, generates, from the learning model 41, structural data (for example, data relating to the atomic-scale uneven shape of the material surface) having an image quality that has been enhanced compared with the input structural data, and outputs the resulting structural data to a user terminal 2.