Low-pass filtering compensation in common-path digital holographic microscopy
File(s)
Author(s)
Type
Journal Article
Abstract
A low-pass filtering compensation (LPFC) method is proposed to compensate for phase aberrations in point diffraction-based common-path digital holographic microscopy. This method estimates the phase aberration from the object hologram by Fourier transform and low-pass spatial filtering. The estimated phase aberration is subtracted from the object phase image to achieve single-hologram phase compensation. The accuracy and capability of LPFC for phase compensation were demonstrated by experiments on a Ronchi grating and a human blood smear. LPFC provides phase compensation for both smooth objects and objects containing abrupt edges, in the special case of a system with relatively high-frequency objects and low-frequency slight phase aberrations. LPFC operates without the need for fitting procedures, iterative steps, or prior knowledge of the optical parameters, which substantially simplifies the process of phase compensation in quantitative phase imaging.
Digital holographic microscopy (DHM) has been developed for a wide range of applications in the examination of cell pathophysiology,1,2 semiconductors,3 and 2D materials.4 Common-path DHM combines common-path geometry and off-axis holography, and hence, it provides subnanometer level optical phase delay (OPD) imaging with high temporal stability and the acquisition speed is limited only by the detector.5–8 Point diffraction-based common-path DHM uses a spatial filter to generate holograms with uniform reference fields, resulting in a compact system with a full field of view (FOV).9–13 In these setups, the zeroth-order beam is low-pass filtered by a pinhole in the Fourier plane of the spatial filtering lens, which is generally assumed to be a uniform field at the surface of the image sensor.11,13 However, due to the use of a microscope objective (MO) and the complex spatial filter, or a non-optimal imaging system, the zeroth-order beam can be distorted, which introduces phase aberrations to the original off-axis holograms. An automatic aberration compensation is desirable to extract the sample phase images.
Various approaches have been proposed to estimate the phase aberrations in DHM. Commonly used double-exposure compensation (DEC) relies on the manual double-exposure operation,14 allowing a calibration of the DHM setup from the second specimen-free hologram. This method requires that the wavefronts of the specimen-free hologram and the specimen hologram are parallel to each other strictly. Zernike polynomial fitting (ZPF) calculates the phase aberration through numerical processing such as computational fitting procedures, which requires prior knowledge about the optical parameters, or iterative procedures15–17 Methods based on a deep learning convolutional neural network (CNN),18 phase variation minimization,19 sparse optimization,20 and synthetic difference21 use complex algorithms to estimate residual aberrations. The self-overlapping approach has been applied without fitting procedures, but it limits the FOV due to the overlapping operations.22 The concept of self-reference conjugated hologram (self-RCH) was introduced for the phase compensation of smooth objects in a modified Mach–Zehnder interferometer.23 However, its accuracy is still to be demonstrated quantitatively.
In this Letter, a low-pass filtering compensation (LPFC) method is proposed to compensate for phase aberrations of both smooth objects and objects with abrupt edges in point diffraction-based common-path DHM, which generates holograms with relatively high-frequency objects and low-frequency slight phase aberrations (the phase contribution due to the aberrations of the whole hologram is less than 10 rad). The capability and accuracy of LPFC are quantitatively proved by the compensation results of a Ronchi grating with abrupt edges and a human blood smear. LPFC requires no fitting procedures, iterative steps, or prior knowledge of the optical parameters, which substantially simplifies the process of phase compensation in quantitative phase imaging.
The schematic of the experimental setup is shown in Fig. 1. A He–Ne laser beam (632 nm, 0.8 mW, HNLS008R-EC, Thorlabs) was expanded and collimated for plane illumination. M1, M2, BE1, and C1 were used to provide illumination in transmission mode. BS1, BE2, C2, and BS2 were for illumination in reflection mode. An infinity-corrected MO (NA = 0.25, Plan N, Olympus) was used to produce a magnified image of the sample. TL was used to collimate the light from MO. G (40 c/mm, Applied Image Inc.) was placed at the image plane to separate the magnified image field into multiple orders. L (f = 100 mm) was used for imaging and spatial filtering simultaneously. The multiple order image fields were isolated in the Fourier plane (the back focal plane) of L, where SF was placed. SF allowed for passing the entire zeroth diffraction order beam, which was used as the sample field. The first order was physically low-pass filtered by a pinhole, which was used as the reference field. All the other diffraction orders were blocked. A CMOS camera (1280 × 1024 pixels, monochrome sensor, DCC1545M, Thorlabs) was used as the detector.
Digital holographic microscopy (DHM) has been developed for a wide range of applications in the examination of cell pathophysiology,1,2 semiconductors,3 and 2D materials.4 Common-path DHM combines common-path geometry and off-axis holography, and hence, it provides subnanometer level optical phase delay (OPD) imaging with high temporal stability and the acquisition speed is limited only by the detector.5–8 Point diffraction-based common-path DHM uses a spatial filter to generate holograms with uniform reference fields, resulting in a compact system with a full field of view (FOV).9–13 In these setups, the zeroth-order beam is low-pass filtered by a pinhole in the Fourier plane of the spatial filtering lens, which is generally assumed to be a uniform field at the surface of the image sensor.11,13 However, due to the use of a microscope objective (MO) and the complex spatial filter, or a non-optimal imaging system, the zeroth-order beam can be distorted, which introduces phase aberrations to the original off-axis holograms. An automatic aberration compensation is desirable to extract the sample phase images.
Various approaches have been proposed to estimate the phase aberrations in DHM. Commonly used double-exposure compensation (DEC) relies on the manual double-exposure operation,14 allowing a calibration of the DHM setup from the second specimen-free hologram. This method requires that the wavefronts of the specimen-free hologram and the specimen hologram are parallel to each other strictly. Zernike polynomial fitting (ZPF) calculates the phase aberration through numerical processing such as computational fitting procedures, which requires prior knowledge about the optical parameters, or iterative procedures15–17 Methods based on a deep learning convolutional neural network (CNN),18 phase variation minimization,19 sparse optimization,20 and synthetic difference21 use complex algorithms to estimate residual aberrations. The self-overlapping approach has been applied without fitting procedures, but it limits the FOV due to the overlapping operations.22 The concept of self-reference conjugated hologram (self-RCH) was introduced for the phase compensation of smooth objects in a modified Mach–Zehnder interferometer.23 However, its accuracy is still to be demonstrated quantitatively.
In this Letter, a low-pass filtering compensation (LPFC) method is proposed to compensate for phase aberrations of both smooth objects and objects with abrupt edges in point diffraction-based common-path DHM, which generates holograms with relatively high-frequency objects and low-frequency slight phase aberrations (the phase contribution due to the aberrations of the whole hologram is less than 10 rad). The capability and accuracy of LPFC are quantitatively proved by the compensation results of a Ronchi grating with abrupt edges and a human blood smear. LPFC requires no fitting procedures, iterative steps, or prior knowledge of the optical parameters, which substantially simplifies the process of phase compensation in quantitative phase imaging.
The schematic of the experimental setup is shown in Fig. 1. A He–Ne laser beam (632 nm, 0.8 mW, HNLS008R-EC, Thorlabs) was expanded and collimated for plane illumination. M1, M2, BE1, and C1 were used to provide illumination in transmission mode. BS1, BE2, C2, and BS2 were for illumination in reflection mode. An infinity-corrected MO (NA = 0.25, Plan N, Olympus) was used to produce a magnified image of the sample. TL was used to collimate the light from MO. G (40 c/mm, Applied Image Inc.) was placed at the image plane to separate the magnified image field into multiple orders. L (f = 100 mm) was used for imaging and spatial filtering simultaneously. The multiple order image fields were isolated in the Fourier plane (the back focal plane) of L, where SF was placed. SF allowed for passing the entire zeroth diffraction order beam, which was used as the sample field. The first order was physically low-pass filtered by a pinhole, which was used as the reference field. All the other diffraction orders were blocked. A CMOS camera (1280 × 1024 pixels, monochrome sensor, DCC1545M, Thorlabs) was used as the detector.
Date Issued
2020-09-21
Date Acceptance
2020-09-12
Citation
Applied Physics Letters, 2020, 117 (12)
ISSN
0003-6951
Publisher
American Institute of Physics
Journal / Book Title
Applied Physics Letters
Volume
117
Issue
12
Copyright Statement
© 2020 The Author(s). This article may be downloaded for personal use only. Any other use requires prior permission of the author and the American Institute of Physics. The following article appeared in: Appl. Phys. Lett. 117, 121105 (2020); doi: 10.1063/5.0019209 and may be found at 10.1063/5.0019209
Identifier
http://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=WOS:000576357000001&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=1ba7043ffcc86c417c072aa74d649202
Subjects
Science & Technology
Physical Sciences
Physics, Applied
Physics
QUANTITATIVE PHASE MICROSCOPY
ABERRATION COMPENSATION
Publication Status
Published
Article Number
ARTN 121105