Multimode Defect Detection Using Difference Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current multimode defect detection methods in semiconductor inspection face challenges such as high false defect rates, limited detectability, and inefficiencies in data processing and storage, particularly when dealing with smaller defect sizes and varying inspection tool modes.
Innovation Solution
A system and method that generates first and second mode test images, reference images, and difference images using an inspection subsystem, with a computer system configured to combine these images for defect detection, effectively utilizing multimode data to enhance detection sensitivity and suppress nuisances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multimode defect detection is performed by scanning the specimen with multiple combinations of parameter values, then defect detection sensitivity is improved, but data processing complexity and storage requirements increase
Solution Approach 1:
The patent segments the multimode defect detection process into distinct functional modules: image acquisition module that captures test images in multiple modes, reference image generation module that creates mode-specific reference images, difference image generation module that subtracts references from tests, and defect detection module that analyzes the difference images. This segmentation allows each module to handle specific tasks independently, reducing overall system complexity while maintaining high detection sensitivity through coordinated multimode analysis.
2Object-generated harmful factors
If rule-based filters are applied to defect attributes from multiple modes, then nuisance filtering is achieved, but detection performance deteriorates when defect attributes are not well-behaved
Solution Approach 1:
The patent introduces difference images as an intermediary representation between the raw test images and the final defect detection results. By subtracting mode-specific reference images from test images to generate difference images, the system creates a standardized intermediate format that highlights defects while suppressing nuisances. This intermediary step transforms variable-quality defect attributes into consistent difference image data, enabling reliable defect detection even when original attributes are not well-behaved.
3Quantity of substance
If a substantially hot scan is performed to detect an extremely large number of defects, then defect detection coverage is improved, but throughput and efficiency decrease
Solution Approach 1:
The patent performs preliminary generation of mode-specific reference images before the actual defect detection scan. These reference images capture the expected specimen appearance under each inspection mode, allowing the system to quickly subtract them from test images during the hot scan. This preliminary preparation enables the hot scan to focus solely on detecting deviations from the reference, significantly increasing throughput while maintaining comprehensive defect detection coverage.
4Adaptability or versatility
If multiple mode scans are performed to detect different defect types, then detection versatility is improved, but inspection time increases
Solution Approach 1:
The patent merges multiple mode scans into a unified defect detection framework where difference images from all modes are combined and analyzed together by a single defect detection module. This integration allows the system to maintain the detection versatility of multiple modes while performing defect detection in a consolidated step, reducing the total inspection time compared to separate detection processes for each mode.
Data Source
AI summary
Methods and systems for detecting defects on a specimen are provided. One method includes generating first and second mode test, reference, and difference images of a specimen for first and second modes of an inspection subsystem, respectively. The method also includes combining the first and second mode test images, the first and second mode reference images, and the first and second mode difference images as an input for defect detection. In addition, the method includes detecting defects on the specimen based on at least the first and second mode difference images.