Charged Particle Beam Image Retrieval From Operation Time Series
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current charged particle beam devices require significant manual effort for sample observation and image data collection, especially for training new experimental series, leading to inefficiencies in semiconductor process development and material informatics applications.
Innovation Solution
A charged particle beam device equipped with a sample stage, imaging unit, output unit, and computer system that semi-automatically generates training image data by analyzing operating status time series data to determine specific variation patterns and extract relevant image data, using machine learning for automated visual field recognition and imaging.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual observation and image data collection methods are used, then training image data can be obtained, but the time and effort required for visual field search and imaging work is excessive
Solution Approach 1:
The system automatically retrieves and selects training image data using its own stored operation logs and image database, without requiring external manual collection. The image retrieval unit queries the database using operation command data as keys, and the system autonomously identifies important images based on predefined criteria such as focus operations and stage movements.
Solution Approach 2:
The system uses previously recorded operation command data as feedback to automatically retrieve relevant training images. By analyzing past operator actions (focus adjustments, stage movements, luminance changes), the system identifies and retrieves images that are likely to be important, creating a closed-loop system where past operations inform future data collection.
2Extent of automation
If machine learning is used to automatically detect characteristic objects, then operator workload for visual field adjustment is reduced, but a large amount of training image data is required which generates significant work load for data collection
Solution Approach 1:
The system automatically retrieves training image data using its own stored operation logs and image database, without requiring external manual collection. The image retrieval unit queries the database using operation command data as keys, and the system autonomously identifies important images based on predefined criteria such as focus operations and stage movements.
Solution Approach 2:
Instead of manually creating new training images, the system copies and retrieves relevant images from the existing database based on operation command patterns. The image retrieval unit extracts and reuses previously captured images that match the criteria for training data, eliminating the need for new manual imaging.
3Extent of automation
If PTL 1 method is used to retrieve important images using operation command data, then image retrieval is automated, but no appropriate images are available in the database for new experimental series
Solution Approach 1:
The system performs preliminary data collection during the observation phase itself. By automatically retrieving and storing images as training data during routine operations, the system prepares training datasets in advance for future machine learning applications, ensuring data availability when new experimental series begin.
Solution Approach 2:
The system automatically retrieves and selects training image data using its own stored operation logs and image database, without requiring external manual collection. The image retrieval unit queries the database using operation command data as keys, and the system autonomously identifies important images based on predefined criteria such as focus operations and stage movements.
Data Source
AI summary
A charged particle beam device includes: a sample stage configured to move a sample; an imaging unit configured to acquire observation image data of the sample; an output unit configured to digitalize an operating status of the charged particle beam device and output operating status time series data; a display unit configured to display a graphical user interface for displaying the observation image data and inputting an observation setting parameter; and a computer system configured to store time series image data in which the observation image data is arranged in time series and execute arithmetic processing relating to the operating status time series data and the observation image data. The charged particle beam device automatically determines a time-point that matches a predetermined specific variation pattern based on the operating status time series data, and acquires observation image data corresponding to the time-point from the time series image data and outputs the observation image data as image data of interest.