|国家预印本平台
首页|DeepPD: Joint Phase and Object Estimation from Phase Diversity with Neural Calibration of a Deformable Mirror

DeepPD: Joint Phase and Object Estimation from Phase Diversity with Neural Calibration of a Deformable Mirror

DeepPD: Joint Phase and Object Estimation from Phase Diversity with Neural Calibration of a Deformable Mirror

来源:Arxiv_logoArxiv
英文摘要

Sample-induced aberrations and optical imperfections limit the resolution of fluorescence microscopy. Phase diversity is a powerful technique that leverages complementary phase information in sequentially acquired images with deliberately introduced aberrations--the phase diversities--to enable phase and object reconstruction and restore diffraction-limited resolution. These phase diversities are typically introduced into the optical path via a deformable mirror. Existing phase-diversity-based methods are limited to Zernike modes, require large numbers of diversity images, or depend on accurate mirror calibration--which are all suboptimal. We present DeepPD, a deep learning-based framework that combines neural representations of the object and wavefront with a learned model of the deformable mirror to jointly estimate both object and phase from only five images. DeepPD improves robustness and reconstruction quality over previous approaches, even under severe aberrations. We demonstrate its performance on calibration targets and biological samples, including immunolabeled myosin in fixed PtK2 cells.

Magdalena C. Schneider、Courtney Johnson、Cedric Allier、Larissa Heinrich、Diane Adjavon、Joren Husic、Patrick La Rivière、Stephan Saalfeld、Hari Shroff

细胞生物学计算技术、计算机技术

Magdalena C. Schneider,Courtney Johnson,Cedric Allier,Larissa Heinrich,Diane Adjavon,Joren Husic,Patrick La Rivière,Stephan Saalfeld,Hari Shroff.DeepPD: Joint Phase and Object Estimation from Phase Diversity with Neural Calibration of a Deformable Mirror[EB/OL].(2025-04-18)[2025-05-21].https://arxiv.org/abs/2504.14157.点此复制

评论