Sub-area Image Comparison for Micro-fabrication Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current defect inspection methods for micro-fabrication objects, such as wafers and MEMs devices, face limitations in sensitivity and accuracy due to process variations and edge effects when comparing image information of repetitive structural elements, leading to reduced detection of defects, especially in non-repetitive regions.
Innovation Solution
A method and system that compare image information of sub-areas within an object, where each sub-area includes an array of repetitive structural elements surrounded by identical non-repetitive regions, allowing for a more robust evaluation by comparing proximate sub-areas, thereby reducing the impact of process variations and edge effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If cell to cell comparison is used for defect inspection, then sensitivity to small defects is improved, but sensitivity to process variations and edge effects increases
Solution Approach 1:
The wafer is divided into multiple sub-areas, each containing repetitive structural elements. By segmenting the comparison into sub-area level rather than individual cell level, the method reduces the impact of edge effects and process variations while maintaining defect detection sensitivity. Each sub-area comparison averages out local variations, providing more reliable defect detection.
2Reliability
If die to die comparison is used for non-repetitive regions, then robustness to process variations is improved, but sensitivity to defects is reduced
Solution Approach 1:
The method merges the advantages of both cell-to-cell and die-to-die comparison by comparing sub-areas that contain both repetitive structural elements and non-repetitive regions. This combination approach maintains robustness to process variations while preserving defect detection sensitivity, as the sub-area comparison captures both types of regions simultaneously.
3Measurement precision
If image acquisition parameters are adjusted to improve sensitivity to small defects, then defect detection capability is improved, but sensitivity to edge effects and process variations increases
Solution Approach 1:
By segmenting the wafer into multiple sub-areas and performing comparisons at the sub-area level rather than individual cell level, the method reduces the impact of edge effects and process variations that are amplified when adjusting image acquisition parameters for high sensitivity. The segmentation approach averages out these harmful factors across multiple elements within each sub-area.
Data Source
AI summary
A method, system and a computer program product for evaluating a object; the method includes: (i) obtaining an image of an area of the object; wherein the area comprises multiple arrays of repetitive structural elements that are at least partially surrounded by at least one group of non-repetitive regions; wherein non-repetitive regions that belong to a single group of non-repetitive regions are ideally identical to each other; wherein the non-repetitive regions are arranged in a repetitive manner; and(ii) providing an evaluation result in response to a comparison between image information of a first sub-area to image information of a second sub-area that is proximate to the first sub-area; wherein the first sub-area comprises a first array of repetitive structural elements and a first non-repetitive region; wherein the second subarea comprises a second array of repetitive structural elements and a second non-repetitive region.