Wafer Scratch Detection Using Adaptive ROI Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting scratches on wafer surfaces after chemical mechanical polishing (CMP) are inefficient, time-consuming, and prone to human error, especially in high-resolution imaging, which is critical for semiconductor manufacturing quality control.
Innovation Solution
An automated method and device for detecting scratches on wafer surfaces using adaptive image binarization and equalization, edge detection, and line detection, which includes preprocessing steps like noise reduction and adaptive thresholding to handle non-uniform contrast distributions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution images are taken and partially enlarged for examination, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The wafer image is divided into multiple regions of interest (ROIs) based on predetermined criteria. By segmenting the full wafer image into smaller manageable regions, the system can process and analyze each region independently and in parallel, reducing the overall inspection time while maintaining high detection accuracy for scratches in each segment
Solution Approach 2:
Instead of processing the entire wafer image at full resolution, the system applies adaptive binarization and processing only to identified regions of interest where scratches are likely to occur or where contrast changes indicate potential defects. This partial processing approach reduces computational load and time while maintaining measurement precision in critical areas
2Measurement precision
If multiple images captured in small sizes need to be analyzed one by one, then measurement precision is improved, but productivity decreases
Solution Approach 1:
Multiple processed regions or images are merged into a comprehensive analysis framework. The system combines results from multiple ROIs and integrates them with the full wafer image context, enabling parallel processing of multiple regions while maintaining the ability to detect scratches with high precision across the entire wafer surface
Solution Approach 2:
The system performs preliminary processing steps such as adaptive binarization, noise reduction, and contrast enhancement on individual regions before final scratch detection. By preparing regions in advance with appropriate preprocessing tailored to each region's characteristics, the system enables faster and more accurate scratch detection when analyzing multiple images or regions
3Measurement precision
If adaptive image binarization and equalization are performed on detailed regions, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system applies different image processing techniques (adaptive binarization, adaptive equalization, noise reduction) to different regions of the wafer image based on local characteristics such as contrast distribution, noise levels, and scratch probability. By tailoring the processing method to each region's specific quality requirements, the system achieves high measurement precision without uniformly applying complex algorithms to the entire image
Solution Approach 2:
The system dynamically selects and adjusts processing parameters based on real-time analysis of each region's characteristics. The adaptive nature of the binarization and equalization algorithms allows them to automatically adjust thresholds and parameters according to local contrast distribution, making the complexity adaptive rather than static, thereby improving precision where needed while managing overall system complexity
Data Source
AI summary
A method for detecting defects on the wafer surface includes receiving a wafer image obtained by capturing a wafer in grayscale or monochromatic expression, determining a uniformity of contrast distribution of the wafer image, detecting n detailed regions separated within the wafer image and performing adaptive image binarization on the n detailed regions when the contrast distribution of the wafer image is determined to be uniform, where n is a natural number, detecting m detailed regions separated within the wafer image and performing adaptive image equalization on the m detailed regions when the contrast distribution of the wafer image is determined to be non-uniform, where m is a natural number, performing edge detection on the wafer image to detect an edge pixel, and performing line detection on the wafer image based on the edge pixel to detect a scratch on the surface of the wafer.