Wafer Defect Inspection Using Shift Calibration and Interference Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing defect inspection methods for wafer surfaces are prone to false alarms due to issues like color cast, color shift, uneven brightness, and noise interference, leading to inaccurate image alignment and detection of false defects.
Innovation Solution
The method involves pre-processing the images by identifying and filtering out interfering sub-blocks, obtaining a shift calibration parameter, and calibrating the test block image to reduce false alarms by comparing the calibrated images for defect inspection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If image comparison is performed directly without pre-processing, then defect inspection can be conducted quickly, but false defects are detected due to color cast, color shift, uneven brightness, and noise interference
Solution Approach 1:
The patent applies preliminary action by performing image pre-processing operations (noise filtering, brightness normalization, color correction) before the defect inspection comparison. This prepares the images in advance to eliminate interfering factors, ensuring that the subsequent defect detection compares only genuine defects rather than artifacts from lighting or noise variations.
Solution Approach 2:
The patent segments the image comparison process into distinct stages: first pre-processing each image independently (noise filtering, brightness normalization, color correction), then performing the defect detection comparison. This segmentation allows each processing stage to be optimized separately, maintaining speed while improving accuracy.
2Measurement precision
If image pre-processing is performed to reduce false alarms, then defect detection accuracy is improved, but processing time and computational complexity increase
Solution Approach 1:
The patent applies partial action by selectively applying pre-processing operations only to the extent necessary for defect detection. Rather than performing exhaustive processing on entire images, the method focuses computational effort on identifying and correcting specific interfering factors (noise, brightness variations, color shifts) that directly impact defect detection accuracy, avoiding unnecessary processing steps.
3Device complexity
If simple image subtraction is used for defect inspection, then the process is simple and fast, but inaccurate alignment results in false defect detection
Solution Approach 1:
The patent introduces intermediary processing steps (noise filtering, brightness normalization, color correction) that mediate between the raw captured images and the final defect detection comparison. These intermediary operations prepare the images by removing interfering factors, ensuring that the subsequent subtraction-based defect detection operates on cleaned, aligned data, thereby reducing false alarms without requiring complex alignment algorithms.
Data Source
AI summary
A method, image processing system, and computer-readable recording medium for item defect inspection are provided. The method is as follows. A test image and a reference image of a test item are received. A test block and a corresponding reference block are obtained from the test image and the reference image to generate a test block image and a reference block image. The test block image and the reference block image are respectively partitioned into multiple sub-blocks. All interfering sub-blocks are identified and filtered out from the test block image and the reference block image, and a shift calibration parameter is obtained accordingly. The test block in the test image is calibrated based on the shift calibration parameter to generate a calibrated test block image. The calibrated test block image and the reference block image are compared to obtain defect information of the test item corresponding to the test block.