Identifying the cause of aberration in general optical systems is difficult and time-consuming because it requires specific measurement and analysis. Therefore, we suggest a novel method based on deep learning to find out misalignment numerically from measured raw images at near-focus, that do not require specific measurements, without the time and effort of analysis. Taking advantage of deep learning, which numerically extracts features from images, our model takes a set of distorted images as input and outputs parameters indicating misalignment. We develop two deep learning models to predict the misalignment of optical systems, a parabolic mirror and a telescope, using a dataset generated through simulation. In spite of real measurement images have noise, the trained model for a parabolic mirror can predict misalignment parameter. Near-focus images suggested by the model exhibit the similar trend in PSF size and stretch direction to the measurement images. To elevate our methods to a practical level, we adjust the telescope in accordance with the model’s predictions. This adjustment results in improved symmetry of the images in the front-back focus direction.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.