Defect Detection in IC Patterns Using Feature Space Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current defect detection methods for objects with integrated circuit patterns, such as photolithography masks and wafers, are inefficient due to high computational demands, limited accuracy, and inability to handle images of different appearances.
Innovation Solution
A computer-implemented method that generates input and reference representations of imaging datasets in a feature space, allowing for defect detection by comparing these representations without the need for image alignment, thereby reducing computational time and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional defect detection methods are used, then defect detection can be performed, but computational time is excessive and accuracy is limited
Solution Approach 1:
The patent extracts only the essential features from the imaging datasets by transforming them into a reduced feature space. This is achieved by projecting the high-dimensional image data onto a lower-dimensional feature space that captures only the most relevant defect-related information, thereby reducing computational complexity while maintaining detection accuracy.
Solution Approach 2:
The patent segments the defect detection task into distinct stages: feature extraction, representation transformation, and defect detection. By dividing the complex task into manageable segments, the system can process data more efficiently and reduce overall computational time while improving accuracy through focused analysis of critical features.
2Adaptability or versatility
If conventional methods are used, then defect detection is possible, but the method cannot handle images of different appearances
Solution Approach 1:
The patent changes the parameter space in which defect detection is performed by transforming images into a normalized feature space. This transformation adjusts and standardizes the representation of images with different appearances, allowing the detection system to reliably identify defects across diverse imaging conditions without requiring separate optimization for each appearance type.
3Ease of operation
If conventional defect detection methods are used, then basic defect identification can be achieved, but expert knowledge is required to define defects
Solution Approach 1:
The patent enables the system to automatically learn and adapt to defect patterns by training on imaging datasets without requiring manual definition of defect criteria. The system self-calibrates by analyzing the data distribution and automatically identifies defect characteristics, eliminating the need for expert intervention in defect definition while maintaining high detection accuracy.
Data Source
AI summary
The invention relates to a computer implemented method for defect detection in an object comprising integrated circuit patterns comprising: obtaining an imaging dataset and a reference dataset of the object; generating an input representation of a subset of the imaging dataset and a reference representation of a corresponding subset of the reference dataset in a feature space; and detecting defects in the object by comparing the input representation to the reference representation in the feature space. The invention also relates to a corresponding computer-readable medium, computer program product and system for defect detection.