Defect Inspection Device Multi-Condition Image Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current defect inspection methods face challenges in detecting minute defects on sample surfaces with high sensitivity due to limitations in simultaneous detection across multiple optical conditions, leading to gaps in throughput and sensitivity, especially when combining bright-field and dark-field inspections, and difficulties in integrating results from multiple detection systems.
Innovation Solution
A defect inspection method and device that image the same region under multiple conditions, extract defect candidates, clip partial images, and associate them in a multi-dimensional feature space to enhance defect detection sensitivity without increasing data volume, utilizing an image acquiring unit, defect candidate extracting unit, partial image clipping unit, and feature quantity calculating unit to output defect information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple optical conditions are used for defect detection, then detection sensitivity is improved, but system complexity and data processing requirements increase
Solution Approach 1:
The patent segments the defect detection process into multiple independent detection units, each handling a specific optical condition. This allows parallel processing of multiple optical conditions without requiring a single complex system to handle all conditions simultaneously, thereby improving detection sensitivity while managing system complexity through modular architecture
Solution Approach 2:
The patent introduces a new dimension for data integration by combining defect detection results from multiple optical conditions in a multi-dimensional feature space. This dimensional expansion allows the system to differentiate between actual defects and noise by analyzing patterns across multiple detection dimensions, thereby improving sensitivity without proportionally increasing system complexity
2Reliability
If multiple detection systems are integrated, then defect capture ratio is improved, but data transfer rate and processing performance requirements increase
Solution Approach 1:
The patent extracts only the essential defect detection results from each detection system rather than processing complete image data. By extracting and integrating only the critical defect information from multiple optical conditions, the system improves defect capture ratio while significantly reducing data transfer rates and processing requirements compared to processing full images from multiple systems
Solution Approach 2:
The patent applies partial action by performing complete defect detection analysis only on regions where defects are detected in at least one optical condition. For regions detected as defect-free in all conditions, the system accepts the partial detection result without requiring exhaustive analysis across all optical conditions, thereby maintaining high defect capture ratio while improving processing throughput
3Measurement precision
If bright-field and dark-field inspections are performed simultaneously, then detection coverage is improved, but throughput gap increases due to sensor imaging period constraints
Solution Approach 1:
The patent merges the detection results from bright-field and dark-field inspections by integrating defect detection outcomes from both optical conditions. This merging approach allows the system to achieve comprehensive detection coverage by combining the complementary strengths of both inspection methods while maintaining high throughput through efficient result integration rather than sequential processing
Data Source
AI summary
In an defect inspection method and device, in order to detect a minute defect present on a surface of a sample with a high degree of sensitivity, a defect inspection method includes imaging the same region of a sample in a plurality of image acquisition conditions and acquiring a plurality of images, processing the plurality of acquired images and extracting a defect candidate, clipping a partial image including the extracted defect candidate and a neighboring image of the defect candidate from the acquired images based on position information of the extracted defect candidate, obtaining feature quantities of the defect candidates in the plurality of clipped partial images, associating the defect candidates that have the same coordinates on the sample and are detected in different image acquisition condition, extracting a defect from among the associated defect candidates in a multi-dimensional feature quantity space, and outputting information of the extracted defect.