Pattern Image ROI Determination for Critical Dimension Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for measuring critical dimensions of patterns, especially fine or complex patterns in semiconductor manufacturing, face challenges in accurately and rapidly determining the region of interest (ROI), requiring skill and experience.
Innovation Solution
A method and apparatus that automatically determine the ROI by matching real and design pattern images to find the optimum turning points along the pattern contour, setting a region of interest within a predetermined range, and measuring the critical dimension within this region, improving precision and reproducibility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If conventional manual methods are used to determine ROI, then measurement can be performed, but the process requires skill and experience and is time-consuming
Solution Approach 1:
The system automatically determines the ROI by processing the pattern image itself through image processing algorithms, without requiring manual intervention or expert skill. The algorithm independently identifies the region of interest based on pattern characteristics, enabling the system to serve itself rather than requiring human expertise for each measurement.
Solution Approach 2:
The system performs preliminary image processing and analysis to automatically identify and determine the ROI before the actual measurement is performed. By pre-processing the image to extract relevant features and determine the region of interest in advance, the system eliminates the need for manual ROI selection and reduces measurement time.
2Measurement precision
If manual ROI determination is used, then flexibility is maintained, but measurement precision and reproducibility are compromised
Solution Approach 1:
The patent replaces manual mechanical selection of ROI with automated image processing algorithms. The system uses computational methods to automatically identify and measure the region of interest, substituting human judgment and manual operations with automated digital processing, thereby improving precision and reproducibility while maintaining ease of operation through automated functionality.
3Productivity
If conventional measurement methods are used, then simple operation is maintained, but measurement speed and efficiency are reduced
Solution Approach 1:
The system creates a digital copy or representation of the pattern image and processes this digital replica to automatically determine the ROI and perform measurements. By working with a digital copy rather than physically manipulating the actual pattern, the system enables rapid automated measurement while keeping the physical measurement system relatively simple.
Data Source
AI summary
In a method of measuring a critical dimension of a pattern, a pattern image is obtained from an object pattern. A design pattern of the object pattern and the pattern image are matched to determine a detection region on the pattern image. An optimum turning point of the pattern contour is determined in the detection region and a ROI (region of interest) is set within a predetermined range from the optimum turning point. A critical dimension of the pattern is measured in the ROI.