In this paper, we propose a robust and accurate estimation method for the distance required for digital holography (DH) reconstruction using convolutional neural networks (CNN) in off-axis DH (off-axis DH). This method applies adaptive spectral pooling to reflect distance-related optical characteristics and minimize information loss during the training phase. Simulations and experiments have confirmed that the proposed method is more robust and accurate than search-based or CNN-based distance estimation methods.
|Journal||Japanese journal of applied physics|
|Publication status||Published - 2023 Jan 1|
Bibliographical notePublisher Copyright:
© 2022 The Japan Society of Applied Physics.
All Science Journal Classification (ASJC) codes
- Physics and Astronomy(all)