Optical phase mining by adjustable spatial differentiator

Tengfeng Zhu,Junyi Huang,Zhichao Ruan

DOI: https://doi.org/10.1117/1.AP.2.1.016001

2019-11-03

Abstract:Phase is a fundamental resource for optical imaging but cannot be directly observed with intensity measurements. The existing methods to quantify a phase distribution rely on complex devices and structures. Here we experimentally demonstrate a phase mining method based on so-called adjustable spatial differentiation, just generally by analyzing the polarization in light reflection on a single planar dielectric interface. With introducing an adjustable bias, we create a virtual light source to render the measured images with a shadow-cast effect. We further successfully recover the phase distribution of a transparent object from the virtual shadowed images. Without any dependence on resonance or material dispersion, this method directly stems from the intrinsic properties of light and can be generally extended to a board frequency range.

Optics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is: how to quantify the phase distribution of light from intensity measurements by means of simple methods without the need for complex equipment. Specifically, the phase plays a crucial role in optical imaging, but it cannot be directly observed through intensity measurements. Existing methods for quantifying the phase distribution rely on complex equipment and structures, leading to difficulties in alignment and adjustment.

To solve this problem, the author proposes a phase - mining method based on an adjustable spatial differentiator. This method is achieved only by analyzing the polarization of the light reflected from a single - plane dielectric interface, and an adjustable bias is introduced to create a virtual light source, thereby producing a shadow effect. This method can directly recover the phase distribution of transparent objects from virtual shadow images without relying on resonance or material dispersion, and is therefore applicable to a wide frequency range.

### Key Point Summary:
1. **Problem Background**:
- Phase information is crucial in biological microscopic imaging and X - ray imaging, but it cannot be directly obtained through intensity measurements.
- Existing methods rely on complex equipment, such as Nomarski prisms, etc., and are difficult to align and adjust.

2. **Solution**:
- A phase - mining method based on an adjustable spatial differentiator is proposed, which is achieved by analyzing the polarization of the light reflected from a single - plane dielectric interface.
- An adjustable bias is introduced to generate a virtual light source and produce a shadow effect, thereby quantifying the phase distribution.

3. **Advantages**:
- It is simple and flexible, and does not require complex modulation equipment.
- It is applicable to a wide frequency range and has the ability of high - throughput real - time image processing.

4. **Experimental Verification**:
- The spatial differential effects in different directions are demonstrated through experiments, and the phase distribution of transparent objects is successfully recovered.
- The experimental results show that this method can accurately recover phase information with a small error.

### Formula Presentation:
- The expression of the spatial - frequency - spectrum transfer function \( H \) is:
\[
H = r_p\sin\gamma_1\sin\gamma_2 + r_s\cos\gamma_1\cos\gamma_2+\delta (r_p + r_s)\sin(\gamma_2-\gamma_1)\frac{2k_y}{k_0}
\]
where \( r_p \) and \( r_s \) are the Fresnel reflection coefficients of p - polarized and s - polarized plane waves respectively, \( \gamma_1 \) and \( \gamma_2 \) are the angles of the polarizers, \( k_y \) is the spatial frequency, and \( k_0 \) is the wave vector number in vacuum.

- Under cross - polarization conditions, the spatial - frequency - spectrum transfer function is simplified to:
\[
H =-(C_1k_x + C_2k_y)
\]
where,
\[
C_1 =-\left(\sin\gamma_1\sin\gamma_2\frac{\partial r_p}{\partial\theta}+\cos\gamma_1\cos\gamma_2\frac{\partial r_s}{\partial\theta}\right)/k_0
\]
\[
C_2=\delta (r_{p0} + r_{s0})\sin(\gamma_1-\gamma_2)/2
\]

- The expression of the output field \( S_{out}(x, y) \) is:
\[
S_{out}(x, y)=i|\vec{l}|\frac{\partial S_{in}}{\partial l}=Ae^{-i\psi}\frac{\partial\psi}{\partial l}
\]
where \( A = |\vec{l}| \), \( \vec{l}=(C_1, C_2) \).

Through these formulas and experimental verification, this paper shows how to utilize...