Wafer Inspection System Using Deep Learning for Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Semiconductor wafers often become contaminated during processing, and existing inspection methods are inefficient, leading to reduced throughput and increased manufacturing time due to the need for manual review of images and time-consuming generation of real-time images.
Innovation Solution
An integrated inspection and cleaning system utilizing an optical microscope and a scanning electron microscope with an image processing unit that applies deep learning algorithms, such as convolutional neural networks, to quickly identify and classify defects on wafers, allowing for automated defect detection and cleaning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual review of inspection images is used, then defect detection accuracy is maintained, but manufacturing time increases and throughput decreases
Solution Approach 1:
The patent replaces manual mechanical inspection with an automated inspection system that uses optical microscopes, scanning electron microscopes, and deep learning algorithms to detect and classify defects. This substitution eliminates the need for human operators to manually review images while maintaining high detection accuracy through automated image analysis and classification mechanisms.
Solution Approach 2:
The inspection system performs self-service by automatically detecting, classifying, and prioritizing defects without human intervention. The deep learning algorithms autonomously analyze inspection images, identify defect patterns, and generate classification results that guide subsequent cleaning operations, enabling the system to serve itself in the defect detection process.
2Speed
If real-time image generation is used for defect inspection, then inspection speed is improved, but system complexity and processing time increase
Solution Approach 1:
The patent segments the inspection system into multiple specialized components: optical microscopes for initial screening, scanning electron microscopes for detailed analysis, and deep learning algorithms for classification. Each component handles specific aspects of defect detection, allowing parallel processing and reducing overall inspection time while managing system complexity through functional decomposition.
Solution Approach 2:
The system performs preliminary actions by using optical microscopes to conduct initial defect screening before more detailed scanning electron microscope analysis. This preliminary inspection filters out obvious defects early in the process, reducing the burden on subsequent analysis stages and enabling faster overall inspection throughput.
3Productivity
If automated defect detection is implemented, then productivity is improved, but the need for sophisticated image processing increases complexity
Solution Approach 1:
The patent introduces deep learning algorithms as intermediaries between the inspection imaging systems and the defect classification process. These algorithms serve as mediators that automatically interpret complex image data, identify defect patterns, and generate classification results, thereby enabling automated defect detection without requiring complex manual image processing procedures.
Data Source
AI summary
A system includes an inspection device and an image processing unit. The inspection device is configured to scan a wafer to generate an inspected image. The image processing unit is configured to receive the inspected image, and is configured to analyze the inspected image by using at least one deep learning algorithm in order to determine whether there is any defect image shown in a region of interest in the inspected image. When there is at least one defect image shown in the region of interest in the inspected image, the inspection device is further configured to magnify the region of interest in the inspected image to generate a magnified inspected image for identification of defects.