Film Thickness Estimation Correcting Substrate Color Influence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing film thickness estimation models do not accurately account for the influence of underlying substrates, leading to variations in estimation results due to the color changes caused by the underlying films, which can affect the accuracy of film thickness measurements.
Innovation Solution
A shape characteristic value estimation apparatus that includes an underlying influence model creation part to correct the estimation results by analyzing the difference between the estimation result of the target film and the actual shape characteristic value of the underlying substrate, using image information from both the processed and underlying substrates to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If an estimation model is applied to estimate the shape characteristic value of the target film using image information, then the measurement efficiency is improved, but the measurement precision deteriorates due to color changes caused by the underlying substrate
Solution Approach 1:
The patent segments the image information into two distinct components: image information from the target film surface and image information from the underlying substrate. By separating these influences, the system can independently analyze and correct for substrate effects, thereby maintaining high measurement efficiency while improving precision through targeted correction of the estimation results
Solution Approach 2:
The patent implements a feedback mechanism where the estimation results are corrected based on the analyzed influence of the underlying substrate. The system uses the image information from the underlying substrate to generate correction values that are fed back to adjust the initial estimation, thereby eliminating the precision degradation caused by substrate color changes while maintaining the efficiency of non-contact measurement
2Device complexity
If the estimation model uses only the processed image information, then the device complexity is reduced, but the measurement precision deteriorates due to unaccounted underlying substrate influence
Solution Approach 1:
The patent applies preliminary action by acquiring and analyzing image information from the underlying substrate before performing the final estimation. The system pre-processes the underlying substrate image information to determine its influence on the target film appearance, and uses this pre-analyzed information to correct the estimation results, thereby improving precision without significantly increasing device complexity
Solution Approach 2:
The patent introduces an intermediary correction mechanism that mediates between the simple estimation model and the complex reality of substrate influence. Rather than complicating the estimation model itself, the system uses an intermediary correction step that accounts for substrate effects, thereby maintaining model simplicity while achieving high measurement precision
Data Source
AI summary
An apparatus includes: a processed-image acquisition part for acquiring a processed image including image information of a surface of a substrate having a target film formed thereon; an estimation part for estimating a shape characteristic value of the target film, by applying an estimation model for estimating the shape characteristic value of the target film to the processed image; an underlying influence model creation part for creating an underlying influence model associated with a correlation between a difference between an estimation result of the shape characteristic value of the target film and a shape characteristic value of the target film acquired without using the estimation model, and information related to a color of a surface of an underlying substrate before the target film is formed, and an estimated-result correction part for correcting the estimation result of the shape characteristic value of the target film based on the underlying influence model.