Semiconductor Hole Bottom Contour Detection via Pixel Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image segmentation techniques for semiconductor specimens struggle to accurately determine the shape of hole bottoms due to the wide range of possible shapes, especially in High Aspect Ratio (HAR) Vias, which complicates defect inspection and measurement processes.
Innovation Solution
A system and method utilizing processing and memory circuitry (PMC) to segment inspection images of semiconductor specimens without relying on shape characterizing models. This involves classifying images into clusters of pixels with discernable values and determining the contour of the hole bottom based on these pixel values, facilitating accurate measurements of hole characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If shape characterizing models are used for image segmentation, then the segmentation process can be simplified, but the accuracy of determining hole bottom shapes deteriorates due to the wide range of possible shapes
Solution Approach 1:
The patent applies image segmentation by clustering pixels with similar intensity values into distinct groups. This divides the complex task of hole bottom shape determination into manageable segments (clusters), where each cluster represents a specific intensity range. The method segments the image without requiring pre-defined shape models, thereby maintaining accuracy while managing complexity through systematic pixel classification.
Solution Approach 2:
The patent changes the approach from using geometric shape parameters (models) to using pixel intensity parameters for segmentation. By transitioning from shape-based characterization to intensity-based clustering, the method adapts to the wide variety of hole bottom shapes without requiring complex shape models, thus resolving the contradiction between process simplicity and measurement accuracy.
2Productivity
If automated examination systems are implemented, then productivity and yield are improved, but the complexity of processing and analyzing images increases
Solution Approach 1:
The patent implements self-service automation where the examination system performs defect detection, classification, and measurement automatically without human intervention. The system uses automated image segmentation and analysis algorithms that process images independently, enabling high throughput while managing complexity through algorithmic automation rather than manual analysis.
Solution Approach 2:
The patent replaces manual examination methods with automated computational systems. Instead of human operators visually inspecting images, the system uses computer-based image processing algorithms, clustering methods, and automated measurement tools to perform examination tasks, thereby increasing productivity while managing complexity through software-based solutions.
3Measurement precision
If multiple examination steps are performed with different settings, then measurement accuracy is improved, but examination time and productivity are reduced
Solution Approach 1:
The patent performs preliminary image processing steps such as noise filtering, contrast enhancement, and initial segmentation before detailed defect analysis. By preparing the image data in advance with appropriate preprocessing operations, the system reduces the time required for subsequent detailed examination steps while maintaining or improving measurement accuracy through progressive refinement.
Solution Approach 2:
The patent applies a tiered examination approach where not all images undergo the full sequence of examination steps. Routine images may receive partial examination with fewer steps, while suspicious or critical images receive more comprehensive analysis. This selective application of examination steps reduces overall examination time while maintaining accuracy for critical cases.
Data Source
AI summary
A system for examining a semiconductor specimen that includes a plurality of layers at respective different depths, and a plurality of holes. Each hole has a top portion at the surface of the specimen, and a bottom portion accommodated in one of the layers. The system includes a processing and memory circuitry (PMC) configured to provide an inspection image indicative of the holes, and process a hole image in the inspection image, without using a shape characterizing model. The processing includes segmenting the inspection image and determining data indicative of a contour of the top portion of the hole, and further segmenting the inspection image and determining data indicative of a contour of a shape enclosed within the contour of the top of the hole.