Automated measurement of similar structures
By employing machine learning models for deformable image registration and automated measurement point selection, the method addresses the inefficiencies and errors in manual measurement techniques, enhancing accuracy and throughput in the measurement of structurally similar images.
WO2025056927A1 Publication Date: 2025-03-20APPLIED MATERIALS INC +1
2 Cites 1 Cited by
Patent Information
- Application Number
- PCT/GR2023/000048
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-09-13
- Publication Date
- 2025-03-20
AI Technical Summary
Technical Problem
Existing measurement techniques for structurally similar images, such as those in semiconductor manufacturing, are time-consuming, labor-intensive, and prone to errors due to the need for manual measurement and registration of images.
Method used
A method using machine learning models, specifically trained on historical image data, to perform deformable image registration and automate the selection of measurement points across similar images, reducing the reliance on manual marking and increasing efficiency.
Benefits of technology
The method significantly reduces the time and labor required for measurement, enhances measurement accuracy and consistency, and improves the throughput of measurement processes in semiconductor manufacturing and other fields.
✦ Generated by Eureka AI based on patent content.
Abstract
A method includes training a machine learning model using a training dataset including historical image data of a structure, the historical image data including a plurality of images of the structure. The training includes labeling regions o f a source image of the plurality of images to create a source mask for the source image and providing as training input to the machine learning model the source image and a target image of the plurality of images. The training further includes receiving as output from the machine learning model a deformation field between the source image and the target image. The training further includes applying the deformation field to the source mask to create a warped source mask. The training further includes computing a loss associated with the warped source mask. The training further includes updating weights of the machine learning model based on the loss.
Need to check novelty before this filing date? Find Prior Art