SEM Pattern Matching Using White Band Width for Noise Isolation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current pattern matching processes in semiconductor inspection using scanning electron microscopes face inefficiencies due to SEM noise variability, requiring frequent threshold adjustments and specialized image processing knowledge, which decreases inspection throughput and accuracy.
Innovation Solution
A pattern matching apparatus that uses white band width information to accurately isolate noise from SEM images, converts patterns into vector data, and matches them with CAD data, allowing for intuitive parameter setting without specialized knowledge, thereby enhancing noise separation and inspection efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a threshold value is provided to isolate SEM noise from the pattern, then noise isolation is achieved, but the threshold value must be frequently changed during the inspection process which decreases inspection efficiency
Solution Approach 1:
The patent changes the parameter used for noise isolation from a fixed threshold value to a dynamic value derived from the white band width of the pattern itself. The system calculates the white band width by analyzing the actual pattern in the SEM image and uses this measured width to determine the appropriate threshold, eliminating the need for manual threshold adjustments while maintaining accurate noise isolation
Solution Approach 2:
The system performs self-service by automatically determining the appropriate threshold value through analysis of the white band width in the pattern. The inspection apparatus autonomously extracts the white band width from the image data and uses it to set the threshold, without requiring external manual intervention or frequent parameter changes by operators
2Measurement precision
If a threshold value is provided to isolate SEM noise from the pattern, then noise isolation is achieved, but the setting of the threshold value cannot be intuitively determined from the SEM image and requires specialized knowledge which makes it difficult for operators to make settings
Solution Approach 1:
The system performs self-service by automatically determining the appropriate threshold value through analysis of the white band width in the pattern. The inspection apparatus autonomously extracts the white band width from the image data and uses it to set the threshold, without requiring external manual intervention or frequent parameter changes by operators
Solution Approach 2:
The patent changes the parameter used for noise isolation from a fixed threshold value to a dynamic value derived from the white band width of the pattern itself. The system calculates the white band width by analyzing the actual pattern in the SEM image and uses this measured width to determine the appropriate threshold, eliminating the need for manual threshold adjustments while maintaining accurate noise isolation
3Productivity
If templates are created by using CAD data to eliminate the prior process, then throughput of the inspection process is increased, but accurate pattern extraction from SEM images becomes more difficult due to noise
Solution Approach 1:
The system performs preliminary action by pre-processing the SEM image data to extract and enhance the white band width information before conducting pattern matching. This preliminary extraction of characteristic dimensions from the noisy image data creates a cleaner, more reliable basis for subsequent CAD data comparison, maintaining both speed and accuracy
Data Source
AI summary
A pattern matching apparatus comprising: means for storing photographed image data of a semiconductor device; means for storing CAD data of said semiconductor device; an information input means for inputting information on the white band width contained in said image data; a pattern extracting means for extracting a pattern on the semiconductor device from said image data by using the white band width information; and a matching means for matching said pattern with the CAD data.


