Aerial Image Determination via Computational Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing metrology systems face challenges in determining production aerial images under demanding illumination and imaging conditions, as they require complex and difficult-to-fabricate measurement setups.
Innovation Solution
The method involves using a different measurement illumination setting than the production setting, allowing for simpler realization within the optical measurement system, thereby relaxing the requirements on the system. This enables simulation of production aerial images that would be difficult or impossible to achieve directly in the measurement setup.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the measurement illumination setting is configured to match the production illumination setting, then the measurement precision is improved, but the device complexity increases and ease of manufacture deteriorates
Solution Approach 1:
The patent creates a computational model (PSF) that copies the production system's imaging characteristics without physically replicating its complex illumination optics. The model is trained using measurement data from a simplified illumination setting, then applied to simulate aerial images for the actual production illumination setting, thereby achieving accurate measurements without complex hardware
Solution Approach 2:
The patent changes the illumination parameters in the measurement setup from the complex production setting to a simpler alternative setting that is easier to implement. By adjusting this parameter and using computational modeling to bridge the difference, the system achieves the desired measurement capability with reduced hardware complexity
2Manufacturing precision
If the production illumination setting uses complex configurations (e.g., elliptic edge contour, freeform illumination), then the manufacturing precision is improved, but the ease of manufacture deteriorates
Solution Approach 1:
Instead of physically fabricating complex setting stops with elliptic or freeform illumination patterns, the patent creates a computational model that copies the imaging behavior of the production system. The model is trained with data from simple illumination settings and then used to simulate the effect of complex production illumination settings, eliminating the need to manufacture difficult-to-fabricate optical elements
Solution Approach 2:
The patent introduces a computational model (PSF - Point Spread Function) as an intermediary between the simple measurement setup and the complex production system. This mediator allows the system to translate measurements taken with simple illumination into accurate predictions for complex production illumination settings, bridging the gap without requiring physical replication of complex optics
Data Source
AI summary
In order to determine a production aerial image of an object to be measured as a result of an illumination and imaging with illumination and imaging conditions of an optical production system, firstly a measurement aerial image of the object to be measured is captured. This is carried out with illumination and imaging conditions of an optical measurement system, which conditions include a predefined measurement illumination setting. Data of the measurement aerial image are generated during the capturing. An object structure of the object to be measured is reconstructed from the data of the captured measurement aerial image by use of a reconstruction algorithm. Data of the reconstructed object structure are generated during the reconstructing. A production aerial image is simulated from the data of the reconstructed object structure with the illumination and imaging conditions of the optical production system. Said conditions include a production illumination setting, which is different than the measurement illumination setting. This results in a determining method in which the requirements made of an optical measurement system used in the determining method, even under demanding illumination and imaging conditions of the optical production system, are relaxed.