Optical Mode Selection for Semiconductor Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semiconductor inspection systems face challenges in identifying suitable measurement parameters and defect detection models, leading to variability in defect inspection sensitivity and efficiency, particularly in determining optimal optical modes for defect detection during runtime.
Innovation Solution
A system and method that perform joint defect discovery and optical mode selection, using a controller to accumulate images and defect data, select optimal optical modes, and train a defect detection model, allowing for scalable and efficient defect detection with improved sensitivity and stability during runtime.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple optical modes are used for defect detection, then detection sensitivity is improved, but system complexity and runtime overhead increase
Solution Approach 1:
The system performs optical mode selection and defect detection model training in advance before runtime inspection. During training, the system evaluates multiple candidate optical modes and identifies the optimal subset for different defect types, storing these pre-determined modes for rapid deployment during actual inspection, thereby avoiding complex real-time mode selection
Solution Approach 2:
The system automatically performs mode selection and model training without requiring manual intervention. The defect detection model self-optimizes by evaluating candidate optical modes against training data, automatically identifying the most effective modes for detecting specific defect types, and this self-determined configuration is then used during runtime inspection
2Measurement precision
If multiple optical modes are used for defect detection, then detection sensitivity is improved, but production throughput decreases
Solution Approach 1:
The system extracts and selects only the most effective optical modes from a larger set of candidate modes for each specific defect type. Instead of applying all available optical modes during inspection, the system identifies and uses only the necessary subset, reducing inspection time while maintaining high detection sensitivity for the target defect types
Solution Approach 2:
The system dynamically selects optical modes based on the specific defect type being detected. Different defect types are associated with different optimal optical mode parameters, and the system adjusts the optical configuration accordingly, allowing efficient detection with minimal modes rather than using all modes for all defect types
3Device complexity
If manual selection of measurement parameters is performed, then system complexity is reduced, but detection accuracy and efficiency deteriorate
Solution Approach 1:
The system automatically performs parameter optimization through machine learning model training. The defect detection model self-determines the optimal optical modes and measurement parameters by analyzing training data, eliminating the need for manual parameter selection while achieving high detection accuracy through data-driven optimization
Data Source
AI summary
A system may be configured for joint defect discovery and optical mode selection. Defects are detected during a defect discovery step. The discovered defects are accumulated into a mode selection dataset. The mode selection dataset is used to perform mode selection to determine a mode combination. The mode combination may then be used to train the defect detection model. Additional defects may then be detected by the defect detection model. The additional defects may then be provided to the mode selection dataset, for further performing mode selection and training the defect detection model. One or more run-time modes may then be determined. The system may be configured for mode selection and defect detection at an image pixel level.