Semiconductor Defect Detection Using Multi-Algorithm Defect Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current semiconductor fabrication processes face challenges in accurately and efficiently detecting defects on wafers due to the high precision and uniformity demands of ultra large-scale integration, requiring improved automated examination methods.
Innovation Solution
A system utilizing multiple algorithms to process images of semiconductor specimens with different acquisition parameters, reducing defect sets through unsupervised and supervised methods, and generating a unified defect map for thorough review, with potential retraining based on data from a fleet of examination tools.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple algorithms are used to process defect detection, then detection accuracy is improved, but processing time and system complexity increase
Solution Approach 1:
The defect detection system is divided into multiple independent algorithms, each specializing in detecting specific types of defects. This segmentation allows each algorithm to be optimized for its specific function while collectively providing comprehensive defect detection coverage, resolving the contradiction between accuracy and complexity by organizing complexity in a modular, manageable way
Solution Approach 2:
The system applies multiple algorithms beyond what a single algorithm could provide, using an excessive number of processing steps to ensure thorough defect detection. This partial or excessive action approach accepts increased processing overhead in exchange for significantly improved detection accuracy, particularly for rare or subtle defects that single algorithms might miss
2Measurement precision
If multiple algorithms are used to process defect detection, then detection accuracy is improved, but processing speed decreases
Solution Approach 1:
Algorithms are executed in a predetermined sequence where earlier algorithms perform preliminary filtering and identification of obvious defects. This preliminary action allows subsequent algorithms to focus only on remaining ambiguous cases, reducing their processing burden and maintaining overall processing speed while still achieving high accuracy through the combined effort of multiple algorithms
3Productivity
If automated examination is implemented, then productivity is improved, but measurement precision may deteriorate due to difficulty in detecting submicron features
Solution Approach 1:
The system introduces multiple specialized algorithms as intermediaries between the automated inspection system and the final defect identification. These algorithms act as mediators that enhance the automated system's ability to detect submicron features by applying specialized processing techniques tailored to different defect types, thereby maintaining high productivity while improving measurement precision for challenging submicron defects
Data Source
AI summary
There are provided systems and methods comprising obtaining a first set of candidate defects of a semiconductor specimen, acquired by an inspection tool associated with a first set of acquisition parameters, using one or more first algorithms to generate, based on the first set of candidate defects, a first reduced set of candidate defects, obtaining a second set of candidate defects of the specimen, acquired by the inspection tool associated with a second set of acquisition parameters, using the one or more first algorithms to generate, based on the second set of candidate defects, a second reduced set of candidate defects, comprising less candidate defects than the second set of candidate defects, and feeding the first reduced set of candidate defects and the second reduced set of candidate defects to a second algorithm, to generate a unified set of candidate defects.