Fourier transform-based hierarchical clustering of layout patterns.
Fourier transform-based hierarchical clustering addresses the inefficiencies in IC inspection by allowing customizable pattern grouping, enhancing defect detection and process optimization.
Patent Information
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2022-04-28
- Publication Date
- 2026-03-11
AI Technical Summary
Existing IC inspection systems face challenges in efficiently and accurately clustering vast amounts of data from SEM images due to predefined and fixed grouping parameters, leading to time-consuming pattern analysis and inefficient defect detection.
The system employs a method of grouping IC patterns using Fourier transform-based hierarchical clustering, allowing for customizable partitioning of patterns into subsets based on similarity, enhancing inspection efficiency and accuracy by recursively evaluating features.
This approach improves the efficiency and accuracy of IC inspection by enabling flexible pattern grouping and hot spot analysis, optimizing the IC manufacturing process.
Smart Images

Figure 0007828365000001 
Figure 0007828365000002 
Figure 0007828365000003
Abstract
Citation Information
Patent Citations
Method for sorting defect on basis of picked-up image and method for displaying the result
JP2001134763A
Domain-based virtual Fourier filter
JP2013542404A
Multi-stage image alignment method for large offset die-to-die inspection
JP2021500740A
Defect displaying method
US20190370950A1
Pattern grouping method based on machine learning
US20200020092A1