Dimension Measurement Apparatus for Dense Pattern Inspection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional inspection methods fail to accurately measure dimensions between reference and evaluation patterns due to increased pattern density and deformation, leading to incorrect correspondences between points, especially at low magnification and in dense pattern areas, which can result in errors in defect detection.
Innovation Solution
A dimension measuring apparatus and computer program that generates correspondence information between points on reference and evaluation patterns, determines consistency, corrects inconsistent correspondences, and judges segment membership to improve correspondence accuracy, thereby reducing errors in inspection results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If image capture is performed at low magnification to suppress dimension measurement time, then productivity is improved, but measurement precision deteriorates due to unclear white bands in dense pattern areas
Solution Approach 1:
The patent applies preliminary action by performing contour extraction and correspondence determination using distance information before final dimension measurement. The distance transform image is generated in advance to facilitate accurate point correspondence even in dense patterns, preparing the data structure needed for precise measurement at low magnification.
Solution Approach 2:
The patent introduces distance transform images as an intermediary representation between the original low-magnification image and the final measurement results. This intermediary data structure encodes distance information that helps resolve ambiguities in contour extraction and point correspondence, enabling accurate measurements despite the low magnification conditions.
2Productivity
If pattern density increases to improve productivity, then more patterns can be inspected, but reliability deteriorates due to increased frequency of incorrect correspondences
Solution Approach 1:
The patent implements feedback by using distance information from the distance transform image to verify and correct point correspondences. The distance metrics provide a feedback mechanism that validates whether proposed correspondences are geometrically consistent, reducing incorrect matches in dense pattern regions.
Solution Approach 2:
The patent changes the parameter representation by transforming spatial coordinates into distance transform space. This parameter transformation allows the system to work with distance values instead of raw coordinates, providing a more robust representation that maintains correspondence accuracy even when pattern density increases.
3Ease of operation
If distance-based correspondence method is used to simplify the inspection process, then ease of operation is improved, but measurement precision deteriorates due to mistakes in corresponding points in dense patterns
Solution Approach 1:
The patent substitutes the simple distance-based mechanical matching approach with a more sophisticated system that incorporates distance transform images and consistency verification. This replacement maintains the automated nature of the process while improving precision by using richer geometric information and validation mechanisms.
Data Source
AI summary
A dimension measuring apparatus for measuring a dimension between a first data contour which is an evaluation reference of a pattern to be evaluated and a second data contour which is the pattern to be evaluated generates first correspondence information between a point on the first data contour and a point on the second data contour, determines consistency of a correspondence included in the first correspondence information, corrects an inconsistent correspondence, and generates second correspondence information, when associating a point on the first contour data and a point on the second contour data with each other.