Charged Particle Beam Imaging for Mixed Structure Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing charged particle beam devices face difficulties in detecting mixed types of structures, as trained models are not suitable for detecting particles with different shapes or foreign substances.
Innovation Solution
A charged particle beam device that includes an irradiation unit, a particle detection unit, an image generation unit, and a structure detection unit, which uses multiple trained models to detect and integrate the locations or regions of different structures within a sample, generating integration result information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single trained model is used to detect specific structures, then detection precision for that specific structure is improved, but the ability to detect mixed types of structures deteriorates
Solution Approach 1:
The detection system is segmented into multiple specialized trained models, each dedicated to detecting a specific type of structure. Instead of using one general model, the system divides the detection task into specialized sub-tasks handled by separate models (e.g., one model for particles, another for foreign substances), allowing each model to achieve high precision for its specific target while collectively covering diverse structure types
Solution Approach 2:
The structure detection unit is designed with multi-functionality by integrating multiple trained models that can detect different types of structures. This universal detection capability allows the same detection unit to handle various structure types (particles, foreign substances, defects) by switching between or combining results from different trained models, thus achieving both precision and versatility
2Adaptability or versatility
If multiple trained models are used to detect different structures, then adaptability for detecting mixed structures is improved, but device complexity increases
Solution Approach 1:
Multiple trained models are merged into a single structure detection unit that processes images collectively. The detection results from different models are integrated and combined to produce a unified detection output, reducing the need for separate physical detection systems for each structure type and thereby managing complexity while maintaining multi-detection capability
Solution Approach 2:
The structure detection unit acts as an intermediary that coordinates multiple trained models. It manages the input images, distributes them to appropriate models, integrates the detection results, and presents unified output. This intermediary layer simplifies the overall system architecture by providing a standardized interface for multi-model operations and result integration
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables the detection of multiple types of structures by utilizing multiple trained models, allowing for accurate identification and representation of various structures within a sample.
Implementation Method 1
an irradiation unit configured to irradiate a sample with a charged particle beam
Data Source
AI summary
A charged particle beam device 100 includes: an irradiation unit 110 configured to irradiate a sample S with a charged particle beam; a particle detection unit 130 configured to detect a particle caused by the irradiation of the sample with the charged particle beam; and a control unit 151 configured to generate an image of the sample based on an output from the particle detection unit, wherein the control unit 151 inputs the image of the sample S into models M1 and M2 for detecting a first structure 401 and a second structure 402, acquires a first detection result related to the first structure 401 and a second detection result related to the second structure 402 from the models M1 and M2, determines locations or regions of the first structure 401 and the second structure 402 based on the first detection result and the second detection result, and outputs an integration result image 203 representing the location or the region of the first structure 401 and the location or the region of the second structure 402.