AI Semiconductor Image Measurement Method
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Accurately measuring semiconductor images during the manufacturing process is challenging due to the shrinking line width of integrated circuits, which affects the yield of subsequent process steps, and existing methods lack the speed and accuracy required by the semiconductor industry.
Innovation Solution
An artificial intelligence (AI) identified measuring method for semiconductor images, utilizing neural network modules such as Convolutional Neural Networks (CNN) or Recurrent Neural Networks (RNN) to identify image types, apply predetermined dimension measuring modes, and extract specific physical parameters from images obtained by microscopes like SEM, TEM, AFM, or X-ray diffractometers, through operations like intensity or differential difference operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional measurement methods are used for semiconductor images, then the measurement process is simple, but the measurement precision and accuracy are insufficient for shrinking line widths
Solution Approach 1:
The patent replaces traditional mechanical/optical measurement systems with an artificial intelligence-based measurement system. The AI system uses neural networks and image processing algorithms to automatically identify and measure semiconductor features, substituting complex physical measurement apparatus with intelligent software-based measurement that achieves higher precision for sub-10nm line widths.
Solution Approach 2:
The patent changes the measurement parameters and approaches based on the identified type of semiconductor image. The AI system automatically selects appropriate measurement parameters (such as edge detection thresholds, measurement directions, and analysis methods) according to the specific image characteristics, enabling accurate measurement across different semiconductor structures and process stages.
2Productivity
If manual or traditional automated measurement methods are used, then the system is easier to operate, but the productivity and speed of measurement are insufficient
Solution Approach 1:
The patent implements a self-service measurement system where the AI automatically performs image analysis, feature identification, and measurement without human intervention. The system autonomously processes semiconductor images, identifies relevant features, selects measurement methods, and generates results, eliminating the need for operators to manually configure measurement parameters or interpret images, thereby dramatically increasing measurement speed.
Solution Approach 2:
The patent performs preliminary classification and identification of image types before executing the actual measurement process. The AI system pre-processes images by categorizing them according to semiconductor structure types and process stages, preparing appropriate measurement strategies in advance, which streamlines the subsequent measurement operations and improves overall productivity.
3Reliability
If comprehensive image analysis is performed to detect all defects, then the detection accuracy improves, but the measurement time and processing complexity increase
Solution Approach 1:
The patent applies local quality analysis by focusing measurement and defect detection efforts on specific regions and features of the semiconductor image that are most critical for quality assessment. The AI system identifies key areas requiring detailed analysis and applies appropriate measurement techniques locally, rather than uniformly processing the entire image, thereby maintaining high detection accuracy while reducing overall processing time.
Solution Approach 2:
The patent segments the image analysis process into distinct stages: initial image classification, feature identification, targeted measurement, and defect detection. The AI system divides the comprehensive analysis into modular segments that can be executed efficiently, performing detailed analysis only on identified features of interest rather than analyzing every pixel uniformly, thus balancing accuracy with speed.
Data Source
AI summary
This inventions provides an artificial intelligence (A.I.) identified measuring method for a semiconductor image, comprising the steps of: providing an original image of a semiconductor; identifying a type and/or a category of the original image by an artificial intelligence; introducing a predetermined dimension measuring mode corresponding to the identified type and/or the identified category to scan the original image to generate a measurement signal of the original image; and extracting a designated object from the original image to generate a specific physical parameter of the original image after operation based on a measurement signal of the designated object and the measurement signal of the original image.