JP2023090248A - Phase image generation device, phase image generation method, and defocus distance determination method - Google Patents

Abstract

To provide a technique capable of generating a phase image by a defocus distance corresponding to a size of an object.SOLUTION: A phase image generation device 1 comprises an illumination section 3, an imaging optical system 4, a detector 5, an acquisition section 9, and a control section 7. The illumination section 3 irradiates an object 8 with illumination light. The imaging optical system 4 includes an objective lens 41 opposed to the object 8. The detector 5 detects light incident from the object 8 through the imaging optical system 4. The acquisition section 9 acquires a size parameter for a size of the object 8. The control section 7 detects the light from the object 8 in at least two positions apart from each other in an optical axis direction by a defocus distance determined on the basis of the size parameter using the detector 5 and generates a phase image for the object 8 by a solution method of an intensity transport equation using the defocus distance and intensity distribution of the light detected by the detector.SELECTED DRAWING: Figure 3

Description

The present disclosure relates to a phase image generation device, a phase image generation method, and a defocus distance determination method.

Observing transparent objects is difficult. This is because the uniformity of the intensity distribution of the light that passes through the object is high, and it is difficult to visually recognize the difference. On the other hand, when light passes through an object, its phase distribution changes according to the refractive index and thickness of the object, so the phase distribution of light from the object reflects the shape of the object. Therefore, a method of observing an object by phase distribution has been conventionally proposed.

As a specific method of obtaining the phase distribution, a method of calculating the phase distribution of light using an intensity transport equation has been proposed (for example, Patent Document 1 and Non-Patent Document 1). In these documents, an object is irradiated with illumination light, and an image of the object is captured by a camera through an objective lens. Then, a captured image showing the observation plane of the object and a plurality of captured images showing a plurality of displaced planes shifted from the observation plane on the optical axis are obtained, and a method of solving the intensity transport equation using these captured images. generates a phase image on the viewing plane.

Japanese Patent Application Laid-Open No. 2021-39119

Tomohiro Shirai, "Phase Information Hidden in Intensity Distribution", Optics, Vol.38, pp.496-502, 2009

In Patent Document 1, the distance between the observation surface and the displacement surface (hereinafter referred to as the defocus distance) is determined based on the wavelength of the illumination light, the refractive index between the objective lens and the object, and the numerical aperture of the objective lens. have set.

However, with the technique of Patent Document 1, the defocus distance does not change when observing objects of different sizes. Therefore, depending on the size of the object, the defocus distance may deviate from the appropriate range. In this case, there is a problem that the accuracy of the phase image is lowered.

Therefore, an object of the present disclosure is to provide a technique capable of generating a phase image with a defocus distance according to the size of the target object.

A first aspect is a phase image generating device that generates a phase image using an intensity transport equation, and includes an illumination unit that irradiates an object with illumination light, and an imaging optical system that includes an objective lens that faces the object. a detector that detects light incident from the object through the imaging optical system; an acquisition unit that acquires a size parameter about the size of the object; and a value determined based on the size parameter. Light from the object at at least two positions separated from each other in the optical axis direction by a focus distance is detected using the detector, and the defocus distance and the intensity distribution of the light detected by the detector are detected. a controller for generating a phase image of the object by solving an intensity transport equation using .

A second aspect is the phase image generating device according to the first aspect, wherein the acquisition unit includes, as the size parameters, a first size parameter indicating a first size and a second size larger than the first size. A second size parameter indicating a size is obtained, the control unit determines an upper limit value of a recommended range of the defocus distance based on the first size parameter, and determines the defocus distance based on the second size parameter. is determined, and the defocus distance is set to a value within the recommended range.

A third aspect is the phase image generating device according to the second aspect, wherein the control unit sets the defocus distance used for generating the phase image to the sum of the lower limit value and the upper limit value. Decide on half value.

A fourth aspect is the phase image generating device according to the second or third aspect, further comprising a storage unit that stores correspondence information about the size and the defocus distance, wherein the control unit includes the The upper limit is determined based on the first size parameter and the correspondence information, and the lower limit is determined based on the second size parameter and the correspondence information.

A fifth aspect is the phase image generating device according to the fourth aspect, wherein the correspondence information includes the defocus distance and , and an index indicating the calculation accuracy of the phase image.

A sixth aspect is the phase image generating device according to the fifth aspect, wherein the index of calculation accuracy includes information on the maximum phase of the calculated phase distribution on the virtual measurement plane calculated by simulation, and in the simulation setting a model phase distribution on the virtual measurement plane; calculating a calculated intensity distribution on a virtual misalignment plane shifted in the optical axis direction from the virtual measurement plane based on the model phase distribution; The calculated phase distribution is calculated by solving the intensity transport equation using the calculated intensity distribution.

A seventh aspect is the phase image generating device according to the second or third aspect, wherein the control unit includes a first model phase distribution which is a model phase distribution on a virtual measurement plane corresponding to the first size parameter and a first step of setting a second model phase distribution that is the model phase distribution corresponding to the second size parameter; a second step of calculating a first calculated intensity distribution on the virtual misalignment plane based on the first model phase distribution, and calculating a second calculated intensity distribution on the virtual misalignment plane based on the second model phase distribution; , calculating a first calculated phase distribution in the virtual measurement plane by solving an intensity transport equation using the defocus distance and the first calculated intensity distribution, and using the defocus distance and the second calculated intensity distribution; A third step of calculating a second calculated phase distribution on the virtual measurement plane by solving the intensity transport equation, and repeating the second step and the third step while changing the candidate value to obtain the candidate value A fourth step of obtaining the first calculated phase distribution and the second calculated phase distribution for each, and determining the lower limit value of the recommended range based on the first calculated phase distribution for each candidate value, and determining the lower limit value of the recommended range for each candidate value and a fifth step of determining an upper limit value of the recommended range based on the second calculated phase distribution of.

An eighth aspect is the phase image generating device according to the sixth or seventh aspect, wherein the model phase distribution has a sinusoidal shape.

A ninth aspect is the phase image generation device according to any one of the first to eighth aspects, wherein the acquisition unit detects the intensity distribution of the object based on the intensity distribution detected by the detector. get the size.

A tenth aspect is a phase image generation method for generating a phase image using an intensity transport equation, comprising the steps of determining a defocus distance based on a size parameter for the size of an object; At least two positions separated from each other by the defocus distance in the optical axis direction using a detector that irradiates light and detects the light that is incident through an imaging optical system that includes an objective lens facing the object. Light from the object is detected using the detector, and by solving an intensity transport equation using the defocus distance and the intensity distribution of light detected by the detector, and generating a phase image.

An eleventh aspect is a method for determining a defocus distance used in solving an intensity transport equation for generating a phase image, comprising the steps of setting a model phase distribution in a virtual measurement plane based on the size of the object. calculating a calculated intensity distribution on a virtual misalignment plane shifted in the optical axis direction from the virtual measurement plane while changing the defocus distance candidate value based on the model phase distribution, and using the calculated intensity distribution; obtaining a calculated phase distribution on the virtual measurement plane by solving the intensity transport equation; and the defocus distance used for solving the intensity transport equation for generating the phase image based on the calculated phase distribution. and determining.

According to the first and tenth aspects, a phase image can be generated with a defocus distance according to the size of the object.

According to the second aspect, it is possible to generate a phase image of an object having a size equal to or larger than the first size and equal to or smaller than the second size with high accuracy.

According to the third aspect, the phase distribution can be calculated more reliably and with high accuracy.

According to the fourth aspect, it is possible to determine the recommended range of the defocus distance with simple processing. Therefore, the load on the controller can be reduced.

According to the fifth aspect, even if the required calculation accuracy changes, it is possible to determine an appropriate defocus distance accordingly.

According to the sixth aspect, since the index of calculation accuracy is obtained by simulation, correspondence information can be easily generated.

According to the seventh aspect, there is no need to create correspondence information in advance.

According to the eighth aspect, the calculated phase distribution can be calculated with a simple calculation method.

According to the eleventh aspect, it helps to set an appropriate defocus distance.

It is a figure which shows an example of a target object roughly. FIG. 4 is a diagram schematically showing an example of a measurement plane and a displacement plane; It is a figure which shows roughly an example of a structure of a phase image generation apparatus. It is a block diagram which shows roughly an example of the hardware constitutions of a control part. 4 is a flowchart showing an example of the flow of simulation; It is a figure which shows an example of model phase distribution. It is a figure which shows an example of model phase distribution. 7 is a graph showing an example of a correspondence relationship between a defocus distance and an index of calculation accuracy; 7 is a graph showing an example of a correspondence relationship between a defocus distance and an index of calculation accuracy; 1 is a functional block diagram schematically showing a first example of a configuration for determining a defocus distance; FIG. It is a flowchart which shows an example of operation|movement of a phase image generation apparatus. FIG. 7 is a functional block diagram schematically showing a second example of a configuration for determining a defocus distance; 5 is a flow chart showing an example of the operation of a defocus distance determination unit; FIG. 11 is a functional block diagram schematically showing a third embodiment of a configuration for determining a defocus distance; 4 is a diagram schematically showing an example of the configuration of an acquisition unit; FIG.

Embodiments will be described below with reference to the accompanying drawings. Note that the components described in this embodiment are merely examples, and the scope of the present disclosure is not intended to be limited to them. In the drawings, for ease of understanding, the dimensions or number of each part may be exaggerated or simplified as necessary.

When expressions indicating relative or absolute positional relationships are used (e.g., "in one direction", "along one direction", "parallel", "perpendicular", "centered", "concentric", "coaxial", etc.), the expressions are , unless otherwise specified, not only the positional relationship is strictly expressed, but also the relatively displaced state in terms of angle or distance within the range of tolerance or equivalent function. When expressions indicating equality (e.g., “same”, “equal”, “homogeneous”, etc.) are used, unless otherwise specified, the expressions not only express quantitatively strictly equality, but also tolerances or It shall also represent the situation where there is a difference in obtaining the same degree of function. When an expression indicating a shape (e.g., "square" or "cylindrical") is used, the expression is not only a geometrically exact representation of the shape, but also a similar effect, unless otherwise specified. In the range in which is obtained, for example, a shape having unevenness or chamfering is also represented. When the expression "comprising", "comprises", "comprises", "includes" or "has" is used, the expression is not an exclusive expression excluding the presence of other elements. When the expression "at least one of A, B and C" is used, the expression includes A only, B only, C only, any two of A, B and C, and A, B and all of C.

Also, in the drawing, an xyz orthogonal coordinate system is shown as appropriate. Hereinafter, one direction along the z direction will be called the +z direction, and the opposite direction will be called the -z direction. The same is true for the x and y directions.

<Phase image generation method>
First, an example of a method of generating a phase image using an intensity transport equation will be described. FIG. 1 is a diagram schematically showing an example of an object 8 to be observed. The object 8 is transparent, for example, and contains cells such as cultured cells as a specific example. In the example of FIG. 1, the object 8 has a curved convex shape with its central portion raised in the -z direction.

As illustrated in FIG. 1, the illumination light L1 is incident on the object 8 along the +z direction. In the example of FIG. 1, the wavefront of the illumination light L1 before entering the object 8 is perpendicular to the z-axis. That is, the illumination light L1 has the same phase at each position on the xy plane in front of the object 8 . In addition, in FIG. 1, the wavefront of the illumination light L1 is indicated by a broken line. Also, here, the intensity (amplitude) of the illumination light L1 at each position on the xy plane in front of the object 8 is assumed to be the same.

If the object 8 has a high transmittance with respect to the illumination light L1, the amplitude of the illumination light L1 is less likely to change due to the transmission of the object 8, as illustrated in FIG. On the other hand, since the refractive index distribution of the object 8 is different from the surrounding refractive index, the wavefront (phase) of the illumination light L1 changes relatively greatly according to the shape of the object 8 . Such an object in which the rate of change in the amplitude of the illumination light L1 is relatively small and the rate of change in phase is relatively large is also called a phase object.

As described above, since the phase distribution of the illumination light L1 reflects the shape of the object 8, if the phase distribution can be obtained, the transparent object 8 can be observed from the phase image showing the phase distribution. .

Here, a measurement plane S0 including a cross section within the object 8 is introduced. The measurement plane S0 is a virtual plane perpendicular to the z-axis (optical axis) and an observation plane. The phase distribution φ _z (x, y) of the illumination light L1 on the measurement plane S0 is represented by the following intensity transport equation. This intensity transport equation is derived by paraxial approximation of the Helmholtz equation.

Here, λ indicates the wavelength of the illumination light L1, ∇ indicates a two-dimensional gradient operator, and _I0 indicates the intensity of the illumination light L1 on the measurement plane S0. Here, the intensity of the illumination light L1 at each position on the measurement surface S0 is the same as that of the intensity _I0 .

The right side of Equation (1) includes the differential of the intensity distribution I _z (x, y) with respect to the propagation direction (that is, +z direction) of the illumination light L1. This differentiation can be approximated using intensity distributions on a plurality of displaced planes S1 shifted in the z-direction from the measurement plane S0. FIG. 2 is a diagram schematically showing an example of the measurement plane S0 and the displacement plane S1. The differentiation on the right side of Equation (1) is approximated by, for example, the following equation.

Here, I _z−Δz (x, y) denotes the intensity distribution of the illumination light L1 on the displacement plane S11 shifted in the −z direction by a distance (hereinafter referred to as defocus distance) Δz from the measurement plane S0, I _z+Δz (x, y) represents the intensity distribution of the illumination light L1 on the displacement plane S12 shifted in the +z direction by the defocus distance Δz from the measurement plane S0. Note that the misalignment surface S1 can also be called a defocus surface.

As the approximation of differentiation, it is not always necessary to employ equation (2). For example, known approximation formulas such as forward difference approximation, backward difference approximation and polynomial approximation can be employed. The number of misaligned surfaces S1 required for approximation is not limited to two, and an appropriate number may be adopted according to the approximation formula. In short, differentiation should be obtained based on the intensity distribution of light at at least two positions separated from each other in the optical axis direction. In the following, for the sake of simplification, an embodiment using Equation (2) will be described.

Transforming equation (1) by a method based on Fourier transform leads to the following equation.

Here, k indicates a wavenumber (2π/λ), FT[...] indicates a Fourier operator, and IFT[...] indicates an inverse Fourier operator. Also, μ and ν denote spatial frequency components along the x and y directions, respectively, and α denotes a regularization parameter. The regularization parameter α is preset, for example.

As can be seen from equations (2) and (3), obtaining the intensity distributions I _z (x, y) (=I ₀ ), I _{z -Δz} (x, y), I _{z +Δz} (x, y) can be obtained, the phase distribution φ _z (x, y) on the measurement plane S0 can be obtained. In other words, a phase image showing the phase distribution φ _z (x, y) on the measurement plane S0 can be generated.

<Overall Configuration of Phase Image Generating Device>
Next, an example of the configuration of the phase image generation device 1 that generates a phase image will be described. FIG. 3 is a diagram schematically showing an example of the configuration of the phase image generating device 1. As shown in FIG. The phase image generation device 1 is a device that generates a phase image of an object 8 using an intensity transport equation. This phase image generating device 1 can also be called a quantitative phase microscope. A phase image may also be referred to as a quantitative phase image.

In the example of FIG. 3, the phase image generating device 1 includes a holding section 2, an illumination section 3, an imaging optical system 4, a detector 5 and a control section . Below, an outline of each configuration will be described first, and then an example of details of each configuration will be described.

The holding part 2 holds the object 8 . If the object 8 contains cells such as cultured cells, the object 8 is dropped onto, for example, a portable plate (eg, well plate). The user places the plate on the holder 2 . The holding unit 2 is, for example, a transparent stage that transmits the illumination light L1, and the plate is placed on the stage. Object 8 may contain a plurality of cells of different sizes.

The illumination unit 3 irradiates the object 8 with illumination light L1. The imaging optical system 4 forms an image of the object 8 on the detection surface of the detector 5 . The detector 5 has a light receiving element on its detection surface and detects light incident on the detection surface. Detector 5 includes, for example, a camera. The detector 5 outputs the intensity distribution of the detected light (that is, the captured image) to the controller 7 .

The detector 5 can detect intensity distributions I _z (x, y), I _z−Δz (x, y), and I _z+Δz (x, y) on the measurement plane S0 and the displacement plane S1. In the example of FIG. 3, a driving section 6 is provided. The drive unit 6 moves the holding unit 2 along the z-direction to sequentially form images on the measurement surface S0 and the displacement surface S1 on the detection surface of the detector 5 . Thereby, the detector 5 can sequentially detect the intensity distributions I _z (x, y), I _{z -Δz} (x, y), and I _{z +Δz} (x, y). Based on the intensity distributions I _z (x, y), I _{z -Δz} (x, y), and I _{z +Δz} (x, y) input from the detector 5, the control unit 7 calculates equations (2) and ( From 3), a phase image representing the phase distribution φ _z (x, y) on the measurement plane S0 is generated.

<Lighting section>
The illumination unit 3 irradiates the object 8 with the illumination light L1. The illumination unit 3 irradiates the object 8 with illumination light L1 whose wavefront is a plane perpendicular to the traveling direction (z direction), for example. The illumination light L1 is, for example, visible light. In the example of FIG. 3 , the illumination section 3 includes a light source 31 and an illumination optical system 32 .

A light source 31 emits illumination light to an illumination optical system 32 . The light source 31 includes, but is not limited to, coherent light sources such as pulsed lasers and continuous wave lasers. When the light source 31 is a coherent light source, it emits spatially coherent illumination light.

The illumination light passes through the illumination optical system 32 and propagates to the object 8 as illumination light L1. The illumination optical system 32, for example, optically adjusts at least one of the wavelength band and wavefront of the illumination light. In the example of FIG. 3, illumination optical system 32 includes bandpass filter 321 , collector lens 322 , field stop 323 , aperture stop 324 and condenser lens 325 . Illumination light emitted from the light source 31 passes through a collector lens 322, a field stop 323, a bandpass filter 321, an aperture stop 324 and a condenser lens 325 in this order, and then illuminates the object 8 as illumination light L1.

The bandpass filter 321 is a bandpass filter having a pass band of a predetermined narrow wavelength band. A filter such as a dielectric multilayer filter, filter glass, or liquid crystal tunable filter is used for the band-pass filter 321, for example. Note that the bandpass filter 321 is unnecessary when the wavelength band of the illumination light is narrow.

The collector lens 322 is a lens that adjusts the direction of illumination light emitted from the light source 31 . A field stop 323 adjusts the illumination area on the holding part 2 . An aperture stop 324 adjusts the numerical aperture of the illumination optical system 32 . The condenser lens 325 refracts the illumination light so that the wavefront is perpendicular to the optical axis.

The illumination light L1 that irradiates the object 8 is not necessarily limited to visible light, and may be, for example, invisible light such as ultraviolet rays and infrared rays. Further, the wavefront of the illumination light L1 does not necessarily have to be perpendicular to the optical axis, and the shape of the wavefront may be known. For example, the wavefront of the illumination light L1 may be spherical.

<Imaging optical system>
An imaging optical system 4 images the object 8 onto the detection plane of the detector 5 . An imaging optical system 4 is provided between the object 8 and the detector 5 . In the example of FIG. 3, the imaging optical system 4 includes an objective lens 41, an aperture stop 42 and an imaging lens 43. FIG. The objective lens 41 is a lens for projecting an image by enlarging the object 8 to be observed. The objective lens 41 is provided facing the object 8 so that its front focal point is located on the object 8 . An aperture stop 42 adjusts the numerical aperture of the imaging optical system 4 . The imaging lens 43 forms an image of the object 8 on the detection surface of the detector 5 . That is, the imaging lens 43 is provided so that its rear focal point is positioned on the detection surface of the detector 5 .

Light transmitted through the object 8 travels through the objective lens 41 , the aperture stop 42 and the imaging lens 43 and enters the detection surface of the detector 5 .

<Detector>
The detector 5 detects light incident through the imaging optical system 4 and outputs the detection result to the controller 7 . The detector 5 is, for example, a so-called camera including an image sensor such as a CCD (Charge Coupled Device) image sensor and a CMOS (Complementary Metal Oxide Semiconductor) image sensor. The detector 5 generates a captured image showing a two-dimensional intensity distribution of light incident on the detection surface.

The drive unit 6, for example, moves at least one of the holding unit 2 and the objective lens 41 (the holding unit 2 in the example of FIG. 3) along the z direction (for example, the optical axis direction of the objective lens 41). The drive unit 6 includes, for example, a ball screw mechanism and a motor.

The drive unit 6 positions the front focal point of the objective lens 41 on the measurement surface S0 of the object 8 by moving the holding unit 2 along the optical axis direction. At this time, the detector 5 can detect the intensity distribution I _z (x, y) on the measurement plane S0. Further, the drive unit 6 moves the holding unit 2 along the optical axis direction to position the front focal point of the objective lens 41 on the displacement surface S11. At this time, the detector 5 can detect the intensity distribution I _z-Δz (x, y) on the displacement surface S11. Further, the drive unit 6 moves the holding unit 2 along the optical axis direction to position the front focal point of the objective lens 41 on the displacement surface S12. At this time, the detector 5 can detect the intensity distribution I _z+Δz (x, y) on the displacement surface S12. The detector 5 outputs intensity distributions I _z (x, y), I _z−Δz (x, y), and I _z+Δz (x, y) to the controller 7 .

When the image of the measurement surface S0 is formed on the detection surface of the detector 5, the detection surface of the detector 5 can be regarded as the measurement surface S0. For example, it may be moved along the optical axis direction of the imaging lens 43 . This also makes it possible to obtain the intensity distributions of the measurement surface S0 and the displacement surface S1.

<Control unit 7>
The control unit 7 can output control signals to the light source 31, the detector 5 and the drive unit 6 to control them.

Further, as illustrated in FIG. 3, the control section 7 includes a defocus distance determination section 7a and a phase distribution calculation section 7b. The defocus distance determination unit 7a determines the defocus distance Δz based on the size of the object 8, as will be described in detail later. The phase distribution calculator 7b calculates the defocus distance Δz determined by the defocus distance determiner 7a and the intensity distributions I _z (x, y), I _z−Δz (x, y), The phase distribution _{φ z (x, y) on the measurement plane S0 is calculated by solving the intensity transport equations (2) and (3) based on I z} +Δz (x, y). In other words, the controller 7 generates a phase image showing the phase distribution φ _z (x, y).

FIG. 4 is a block diagram schematically showing an example of the hardware configuration of the controller 7. As shown in FIG. The control unit 7 is an electronic circuit and has, for example, a data processing unit 701 and a storage unit 702 . In the specific example of FIG. 4, the data processing section 701 and the storage section 702 are interconnected via the bus 70 . The data processing unit 701 may be an arithmetic processing device such as a CPU (Central Processor Unit). The storage unit 702 may have a non-temporary storage unit (eg, ROM (Read Only Memory) or hard disk) 703 and a temporary storage unit (eg, RAM (Random Access Memory)) 704 . The non-temporary storage unit 703 may store, for example, a program that defines processing to be executed by the control unit 7 . By the data processing unit 701 executing this program, the control unit 7 can execute the processing specified in the program. Of course, part or all of the processing executed by the control unit 7 may be executed by hardware such as a dedicated logic circuit.

In the example of FIG. 3 , a display section 71 is electrically connected to the control section 7 . The display unit 71 is a display such as a liquid crystal display. The display unit 71 receives the phase image from the control unit 7 and displays the phase image. The user can observe the object 8 by viewing the phase image displayed by the display unit 71 .

In the example of FIG. 3, an input section 72 is electrically connected to the control section 7 . The input unit 72 is a device that receives user input, and includes at least one of a keyboard, mouse, touch panel, and microphone, for example. The input unit 72 may accept input of size parameters for the size of the object 8 . The user inputs a size parameter about the size of the object 8 to the input section 72 , and the input section 72 outputs the size parameter to the control section 7 . The defocus distance determination unit 7a of the control unit 7 determines the defocus distance Δz based on this size parameter, as will be detailed later.

<Calculation Accuracy of Defocus Distance Δz and Phase Distribution>
Before explaining the method of determining the defocus distance Δz based on the size of the object 8, first, the relationship between the size of the object 8, the defocus distance Δz, and the approximation accuracy of Equation (2) will be explained. In the present embodiment, the correspondence relationship will be clarified by a simulation which will be described later.

FIG. 5 is a flow chart showing an example of the flow of simulation performed by the simulator. This simulator may be implemented by the control unit 7 or may be implemented by another arithmetic circuit.

Here, a virtual object that is a model of the object 8 is introduced, and the phase distribution on the virtual measurement plane of the light that passes through the virtual object is set as the model phase distribution φ _mz (x, y) (step ST1). . The virtual measurement plane is a virtual observation plane used in the simulation, and is a virtual plane perpendicular to the z-axis (optical axis). 6 and 7 are diagrams schematically showing an example of the model phase distribution φ _mz (x, y). 6(a) and 7(a) show plan views of the model phase distribution φ _mz (x, y), and FIG. 6(b) shows the model phase distribution on line AA in FIG. 6(a). φ _mz (x, y), and FIG. 7(b) shows the model phase distribution φ _mz (x, y) on line AA in FIG. 7(a). AA line is a line corresponding to one cycle. Note that in FIG. 6B, an example of a calculated phase distribution φ _cz (x, y), which will be described later, is indicated by a dashed line.

In the examples of FIGS. 6B and 7B, the model phase distribution φ _mz (x, y) has a sinusoidal shape with the position in the x direction as a variable. That is, here, the shape of the object 8 is modeled using only the fundamental frequency component. The spatial frequency of this model phase distribution φ _mz (x,y) in the x direction reflects the size of each virtual object. A single virtual object is assumed in FIG. 6, and a state in which a plurality of virtual objects are arranged in the x direction is assumed in FIG. The higher the spatial frequency, the smaller the virtual object. As a specific example, the x-direction size of the virtual object in FIG. 6 is 800 μm, and the x-direction size of the virtual object in FIG. 7 is 40 μm. Also, here, the maximum phase φ _{m_max} of the model phase distribution φ _mz (x, y) is set to 2π.

Here, simulations are performed for virtual objects of different sizes. Specifically, candidate values f1 and f2 are set for the spatial frequency f of the model phase distribution φ _mz (x, y), and the candidate values f1 and f2 are sequentially subjected to the simulation described later. The candidate value f1 is the spatial frequency when the size of the virtual object is a smaller first size (for example, 40 μm), and the candidate value f2 is the second size (for example, 40 μm) larger than the first size of the virtual object. for example, 800 μm).

First, the simulator sets the model phase distribution φ _mz (x, y) at the spatial frequency of the candidate value f1 (step ST1). This model phase distribution φ _mz (x, y) is, for example, the model phase distribution φ _mz (x, y) in FIG. 7, and the size of the virtual object is, for example, 40 μm.

Next, the simulator sets the defocus distance Δz (step ST2). More specifically, the simulator adopts the smallest candidate value Δz1 among a plurality of candidate values Δz1, . . . , Δzn assumed as the defocus distance Δz. For example, 1 μm can be used as the candidate value Δz1, and 300 μm, for example, can be used as the maximum candidate value Δzn.

Next, the simulator calculates the intensity distribution (hereinafter referred to as the calculated intensity distribution) on the virtual misalignment plane described below by a known light propagation calculation method based on the model phase distribution φ _mz (x, y). (Step ST3). The virtual misalignment plane is a virtual plane shifted in the z direction from the virtual measurement plane by the defocus distance Δz set in step ST2. For example, the simulator calculates the calculated intensity distribution I cz−Δz (x, y) on the virtual misalignment plane shifted in the −z direction from the virtual measurement surface by the defocus distance Δz, and the calculated intensity distribution I _cz−Δz (x, y) on the virtual misalignment plane shifted in the +z direction. Calculate the distribution Icz _+Δz (x,y).

As the light propagation calculation method, for example, diffraction calculation methods such as Fresnel diffraction calculation and angular spectrum method are employed. A uniform intensity distribution was adopted as the virtual intensity distribution (hereinafter referred to as a model intensity distribution) _Imz (x, y) on the virtual measurement plane. In other words, the intensity at each position on the virtual measurement plane was assumed to be the same. Also, the wavelength of the illumination light L1 was adopted as the wavelength used in the simulation. Also, in the light propagation calculation, the light propagates through the air.

_{Next , the} simulator calculates the phase A distribution (hereinafter referred to as a calculated phase distribution) φ _cz (x, y) is calculated (step ST4).

More specifically, the simulator first converts the model intensity distribution I _mz (x, y) and the calculated intensity distributions I _cz−Δz (x, y), I _cz+Δz (x, y) to the number of bits of the detector 5. Quantize with The number of bits of the detector 5 here is, for example, the number of bits of each pixel value of the captured image of the detector 5, and as a specific example, it is 8 bits.

Note that the simulator generates random noise imitating noise such as aberration of the optical system of the phase image generation device 1 and thermal noise of the electrical system by using the model intensity distribution I _mz (x, y) before quantization and the calculated intensity distribution I _cz−Δz (x, y) and I _cz+Δz (x, y) may be multiplied.

Then, based on the model intensity distribution I _mz (x, y) after quantization and the calculated intensity distributions I _cz−Δz (x, y) and I _cz+Δz (x, y), the simulator uses equation (2) and equation A calculated phase distribution φ _cz (x, y) is calculated by solving the intensity transport equation of (3).

As described above, the simulator sets the model phase distribution φ _mz (x, y) and the defocus distance Δz, and uses the model phase distribution φ _mz (x, y) and the defocus distance Δz to calculate the phase distribution φ _cz Calculate (x, y). The more similar the calculated phase distribution φ _cz (x, y) to the model phase distribution φ _mz (x, y), the more appropriately the phase distribution is reproduced, and the higher the calculation accuracy. If the calculation accuracy is high, it can be said that the defocus distance Δz used to calculate the calculated phase distribution φ _cz (x, y) is appropriate.

Therefore, the simulator calculates an index indicating calculation accuracy (step ST5). The indicator will be detailed later.

Next, the simulator determines whether or not the index of calculation accuracy has been calculated for all the candidate values Δz1, . . . , Δzn of the defocus distance Δz (step ST6). If the calculation of indices for all the candidate values Δz1, . Execute.

When the index calculations for all the candidate values Δz1, . FIG. 8 is a graph showing an example of a correspondence relationship between the defocus distance Δz and an index of calculation accuracy. The index of calculation accuracy will be described in detail later.

Next, the simulator judges whether or not the above correspondence relationship has been calculated for all the candidate values f1 and f2 of the spatial frequency f (step ST7). When the calculation of the correspondence relationship for all the candidate values f1 and f2 has not yet been completed, the simulator sets the model phase distribution φ _mz (x, y) at the next candidate value f2 of the spatial frequency f in step ST1. This model phase distribution φ _mz (x, y) is, for example, the model phase distribution φ _mz (x, y) in FIG. 6, and the size of the virtual object is, for example, 800 μm. Then, the simulator sequentially executes steps ST2 to ST6 again.

When the calculation of the correspondence relationship for all the candidate values f1 and f2 is completed, the simulator ends the processing. Thereby, the correspondence relationship between the defocus distance Δz and the index of the calculation accuracy can be obtained for each of the candidate values f1 and f2 of the spatial frequency f, that is, for each size of the virtual object. FIG. 9 is a graph showing an example of a correspondence relationship between the defocus distance Δz and an index of calculation accuracy.

Here, an example of the index of calculation accuracy will be described. As the index, referring to FIG. 6, the maximum phase φ _{c_max} in a predetermined region within the calculated phase distribution φ _cz (x, y) can be used. In FIG. 6, the calculated phase distributions φ _cz (x, y) corresponding to different defocus distances Δz are schematically indicated by dashed lines. For example, the AA line on the virtual measurement plane can be used as the predetermined area. In the examples of FIGS. 6 and 7, the AA line corresponds to one cycle line of the phase distribution on the virtual measurement plane. Note that the predetermined area may be the entire area of the phase distribution. Also, since the minimum phase in the phase distribution is substantially 0, it can be said that the index indicates the difference (phase range) between the maximum phase and the minimum phase.

FIG. 8 shows the correspondence (hereinafter referred to as the first correspondence) between the defocus distance Δz and the maximum phase _{φc_max} when the virtual object has a relatively small size of 40 μm. As can be understood from FIG. 8, when the virtual object is small, the maximum phase φ _{c_max} is close to its true value, the maximum phase φ _{m_max} , within a short range of the defocus distance Δz. On the other hand, within a range where the defocus distance Δz is long, the maximum phase φ _{c_max} decreases as the defocus distance Δz increases, and gradually moves away from the maximum phase φ _{m_max} . That is, the approximation accuracy decreases as the defocus distance Δz increases.

The reason why such a first correspondence is established is considered as follows. In the first place, when the defocus distance Δz is long, the approximation of equation (2) becomes rough, and the approximation accuracy decreases. This decrease in accuracy becomes more pronounced as the spatial frequency of the intensity distribution Iz(x, y) increases. When the thickness of the virtual object is the same, the smaller the size of the virtual object in plan view, the smaller the curvature of the wavefront of light (see also the convex wavefront in FIG. 1). tend to become Therefore, as the virtual object becomes smaller, the decrease in the approximation accuracy of Equation (2) becomes more pronounced when the defocus distance Δz is set longer.

In FIG. 9, solid lines indicate the correspondence (hereinafter referred to as the second correspondence) when the virtual object is relatively large at 800 μm. As can be understood from FIG. 9, when the virtual object is large, the maximum phase _{φc_max} varies significantly with changes in the defocus distance Δz within a short range of the defocus distance Δz. On the other hand, within a range where the defocus distance Δz is long, the variation width of the maximum phase φ _{c_max} is small, and the maximum phase φ _{c_max} is close to its true value, the maximum phase φ _{m_max} .

The reason why such correspondence is established is considered as follows. That is, when the size of the virtual object in plan view is large, the amount of change in the right side of Equation (2) is small because the low frequency components are large. Therefore, noise caused by various factors such as aberration of the optical system, thermal noise of the electrical system, and quantization by the detector 5 becomes relatively large with respect to the amount of change on the right side of the equation (2). influence becomes greater. Then, when the interval between the misaligned surfaces S11 and S12 (=twice the defocus distance Δz) becomes narrower, the difference between the intensity distributions I _{z + Δz} (x, y) and I _{z - Δz} (x, y) becomes smaller. Therefore, it is considered that the influence of noise becomes remarkable when the defocus distance Δz is set short. That is, when the size of the virtual object is large, the shorter the defocus distance Δz is set, the lower the noise immunity.

As described above, the correspondence relationship between the defocus distance Δz and the index of calculation accuracy greatly depends on the size of the virtual object. Therefore, the phase image generation device 1 according to the present embodiment determines the defocus distance Δz according to the size of the object 8 to be measured (observed). A specific example will be described below.

<Concept of determination of defocus distance>
As can be understood from FIG. 8, when the object 8 is small, the calculation accuracy tends to decrease as the defocus distance Δz increases. Therefore, the upper limit of the recommended range R1 of the defocus distance Δz should be determined based on the assumed minimum size of the object 8 . On the other hand, as can be understood from FIG. 9, when the object 8 is large, the shorter the defocus distance Δz, the more the calculation accuracy fluctuates with respect to changes in the defocus distance Δz. It is not preferable to adopt a value within such a range where the fluctuation width of the calculation accuracy is large as the defocus distance Δz. Therefore, the lower limit value of the recommended range R1 of the defocus distance Δz should be determined based on the maximum size assumed as the size of the target object 8 . If the defocus distance Δz is set to a value within this recommended range R1, the phase image of the object 8 having an expected size can be detected with higher accuracy.

<correspondence information>
FIG. 10 is a functional block diagram schematically showing a first embodiment of a configuration for determining the defocus distance Δz. In the example of FIG. 10, an input section 72 and a storage section 73 are connected to the defocus distance determining section 7a.

The input unit 72 inputs a first size parameter indicating a first size (for example, minimum size) of the object 8 and a second size parameter indicating a second size (for example, maximum size) larger than the first size of the object 8. accepts the input of The first size parameter and the second size parameter may be the size of the object 8 itself, or may be spatial frequencies. For example, when the object 8 includes a plurality of unit objects (eg, cells), the user inputs the first size assumed to be the smallest among the sizes of the unit objects to the input unit 72 as a first size parameter, A second size assumed to be the largest is input to the input unit 72 as a second size parameter. The input unit 72 outputs the input first size parameter and second size parameter to the defocus distance determination unit 7a.

The storage unit 73 is a non-temporary storage unit such as a memory, and stores the correspondence relationship information D. FIG. The correspondence information D includes information about the size of the object 8 and the defocus distance Δz. As a more specific example, the correspondence information D includes the first correspondence between the defocus distance Δz for each first size and the index of calculation accuracy, and the defocus distance Δz for each second size and the index of calculation accuracy. contains a second correspondence with That is, the correspondence information D includes a plurality of first correspondences for each first size similar to FIG. 8, and a plurality of second correspondences for each second size similar to the solid line graph in FIG. This correspondence information D is created by the simulation described above.

The defocus distance determination unit 7 a determines the defocus distance Δz based on the first size parameter and the second size parameter from the input unit 72 and the correspondence information D stored in the storage unit 73 . As a specific example, first, the defocus distance determining unit 7a identifies from the correspondence information D the first correspondence (see FIG. 8) corresponding to the first size parameter. The defocus distance determination unit 7a determines the upper limit value of the first range R11 of the defocus distance Δz described below as the upper limit value of the recommended range R1 of the defocus distance Δz. The first range R11 is the defocus distance Δz at which the error Δφ between the maximum phase φ _{m_max} (that is, the true value) and the maximum phase φ _{c_max} is stably equal to or less than the prescribed threshold value Ref1 in the first correspondence relationship. Range. The term "stable" as used herein means that the error .DELTA..phi. is always less than or equal to the threshold value Ref1 within the range of change even if the defocus distance .DELTA.z changes slightly.

Next, the defocus distance determining unit 7a identifies a second correspondence (see FIG. 9) corresponding to the second size parameter from the correspondence information D, and stabilizes the error Δφ in the second correspondence. The lower limit of the second range R12 of the defocus distance Δz that is equal to or less than the threshold Ref1 is determined as the lower limit of the recommended range R1.

Next, the defocus distance determination unit 7a determines the defocus distance Δz to be a value within the recommended range R1. As a more specific example, the defocus distance determination unit 7a determines the defocus distance Δz to be the median value of the recommended range R1 (that is, the half value of the sum of the lower limit value and the upper limit value).

Note that when the first correspondence corresponding to the input first size parameter is not included in the correspondence information D, the defocus distance determination unit 7a appropriately uses an interpolation method such as linear interpolation to perform the input A first correspondence corresponding to the obtained first size parameter may be calculated. The same applies to the second correspondence.

<Operation of phase image generating device>
FIG. 11 is a flowchart showing an example of the operation of the phase image generating device 1. FIG. First, the size of the object 8 is obtained (step ST10). As a specific example, the user inputs the assumed minimum size (first size) of the target object 8 to the input unit 72 as a first size parameter, and the assumed maximum size (second size) of the target object 8 is input to the input unit 72 as a second size parameter. The input section 72 outputs the first size parameter and the second size parameter to the control section 7 . The display unit 71 may display information prompting the input of the first size (minimum size) and the second size (maximum size) on the size parameter input screen.

Next, the defocus distance determination unit 7a determines the defocus distance Δz based on the first size parameter, the second size parameter and the correspondence information D as described above (step ST11).

Next, the control unit 7 acquires intensity distributions I _z (x, y), I _z−Δz (x, y), and I _z+Δz (x, y) on the measurement surface S0 and the displacement surfaces S11 and S12 ( step ST12). As a specific example, while the illumination unit 3 is emitting the illumination light L1, the driving unit 6 moves the holding unit 2 along the optical axis direction to position the front focal point of the objective lens 41 on the measurement surface S0. . As a result, an image of the measurement surface S0 is formed on the detection surface of the detector 5. FIG. The detector 5 detects the intensity distribution I _z (x, y) on the detection plane and outputs it to the controller 7 . Similarly, the driving section 6 moves the holding section 2 along the -z direction by the defocus distance Δz determined in step ST11. As a result, an image of the displacement surface S11 is formed on the detection surface of the detector 5. FIG. The detector 5 detects the intensity distribution I _z-Δz (x, y) on the detection plane and outputs it to the controller 7 . Similarly, the intensity distribution I _z+Δz (x, y) on the misaligned surface S12 is input to the controller 7 .

Next _, the phase distribution calculator 7b calculates the phase distribution _{φ z ( x} , y) are calculated (step ST13). In other words, the phase distribution calculator 7b generates a phase image.

The control unit 7 appropriately outputs the phase image to the display unit 71 and causes the display unit 71 to display the phase image. This allows the user to observe the object 8 .

As described above, the defocus distance determination unit 7a determines the defocus distance Δz according to the size of the object 8. FIG. The detector 5 detects the intensity distribution of the measurement surface S0 and the intensity distribution of the misaligned surface S1 defined by the defocus distance Δz, and the phase distribution calculator 7b calculates the defocus distance Δz and these intensity distributions. is used to calculate the phase distribution φ _z (x, y). Therefore, the phase distribution φ _z (x, y) can be calculated with higher accuracy according to the size of the object 8 . In other words, a phase image can be generated with high accuracy.

Moreover, in the above example, the upper limit of the recommended range R1 is determined based on the first size (for example, the minimum size) of the object 8, and the recommended range R1 is determined based on the second size (for example, the maximum size) of the object 8. A lower limit value is determined, and the defocus distance Δz is determined to be a value within the recommended range R1. Therefore, the phase image generation device 1 can more reliably generate a phase image of the object 8 having the first size or more and the second size or less with high accuracy. In addition, even if the object 8 includes a plurality of unit objects (for example, cells) in a range of the first size or more and the second size or less, the phase image generation device 1 may A phase distribution can be calculated with high accuracy.

Further, in the above example, the correspondence information D stored in advance in the storage unit 73 is used. Therefore, the phase image generation device 1 does not need to perform a simulation for calculating the correspondence when generating the phase image of the object 8 . Therefore, the processing load on the control unit 7 can be reduced.

In the above example, the correspondence information D includes the correspondence between the defocus distance Δz and the index of calculation accuracy. Therefore, even if the calculation accuracy requirement changes, the defocus distance determining unit 7a sets the threshold value Ref1 for the calculation accuracy according to the required accuracy, so that an appropriate defocus distance Δz corresponding to the change can be obtained. can be determined.

Further, since the correspondence information D is generated by simulation, the correspondence information D can be generated more easily than by experiment.

<Regarding model phase distribution>
In the above example, in the simulation for generating the correspondence information D, a sine wave with only fundamental frequency components was adopted as the model phase distribution φ _mz (x, y) (see also FIGS. 6 and 7). According to this, the correspondence information D can be generated by simpler calculation.

Also, in the examples of FIGS. 6 and 7, the y-direction distribution of the model phase distribution φ _mz (x, y) is set to be uniform. That is, the object 8 is modeled with a sine wave of only the fundamental frequency component in one direction (here, x direction). This also makes it possible to generate the correspondence information D with a simpler calculation method.

Also, in the above example, the maximum phase φ _{m_max} of the model phase distribution φ _mz (x, y) is set to 2π. That is, the thickness of the virtual object is assumed to be large.

By the way, the larger the maximum phase _{φm_max} is set, the larger the error Δφ between the maximum phase _{φm_max} and the maximum phase _{φc_max} calculated by simulation. Therefore, it is preferable to set the maximum phase _{φm_max} to 2π so as to maximize the error Δφ and determine the defocus distance Δz so that the error Δφ is equal to or less than the threshold value Ref1. According to this, by using the determined defocus distance Δz, the error Δφ can be made equal to or less than the threshold value Ref1 even when the maximum phase _{φm_max} is equal to or less than the set value. That is, it is possible to detect the phase image of the object 8 having a thickness equal to or less than the assumed thickness with high accuracy.

<Index of calculation accuracy>
In the above example, the maximum phase φ _{c_max} corresponding to the thickness of the virtual object is used as an index of calculation accuracy. Since the present embodiment aims to generate a phase image of the object 8 having thickness with high accuracy, the maximum phase _{φc_max} is suitable as an index of calculation accuracy. Moreover, since the calculation of the maximum phase _{φc_max} is simple, an index of calculation accuracy can be obtained more easily.

<Correspondence information D>
In the above example, the correspondence information D includes the correspondence between the defocus distance Δz and the index of calculation accuracy. However, it is not necessarily limited to this. As can be understood from FIG. 9, if the threshold value Ref1 is set, the upper limit value of the appropriate recommended range R1 corresponding to the first size and the lower limit value of the appropriate recommended range R1 corresponding to the second size are determined in advance. It is possible to Therefore, the correspondence information D may include the correspondence between the first size and the upper limit of the defocus distance Δz, and the correspondence between the second size and the lower limit of the defocus distance Δz (Table 1 reference).

In this case, the defocus distance determination unit 7a does not have to perform the comparison processing between the maximum phase _{φc_max} and the threshold value Ref1. A defocus distance Δz can be determined. That is, the processing load on the control unit 7 can be reduced.

<Calculation>
Although the correspondence information D is stored in the storage unit 73 in the above example, it is not necessarily limited to this. When it is not necessary to reduce the processing load of the control unit 7, the control unit 7 may perform a simulation according to the size parameter to determine the defocus distance Δz.

FIG. 12 is a functional block diagram schematically showing a second embodiment of the configuration for determining the defocus distance Δz. In the example of FIG. 12, the defocus distance determination unit 7a includes a model setting unit 71a, a defocus distance setting unit 72a, an intensity distribution calculation unit 73a, a phase distribution calculation unit 74a, and an accuracy calculation unit 75a. The defocus distance determination unit 7a calculates the first correspondence (see FIG. 8) and the second correspondence (FIG. 9) by simulation based on the first size parameter and the second size parameter from the input unit 72. Then, the defocus distance Δz is determined based on the first correspondence and the second correspondence.

FIG. 13 is a flow chart showing an example of the operation of the defocus distance determining section 7a. First, the model setting unit 71a sets the model phase distribution φ _mz (x, y) based on the size parameter from the input unit 72 (step ST21). Specifically, first, the model setting unit 71 a sets the model phase distribution φ _mz (x, y) based on the first size parameter from the input unit 72 . The model intensity distribution I _mz (x, y) is set with a uniform distribution, for example.

Next, the defocus distance setting unit 72a sets a candidate value for the defocus distance Δz (step ST22), as in step ST2. Here, the defocus distance setting unit 72a sets the candidate value of the defocus distance Δz to, for example, the candidate value Δz1.

Next, similarly to step ST3, the intensity distribution calculator 73a converts the calculated intensity distributions I _cz−Δz (x, y) and I _cz+Δz (x, y) on the virtual misalignment plane into the model phase distribution φ _mz (x, y) (step ST23).

Next, the phase distribution calculator 74a calculates the calculated phase distribution φ _cz (x, y) on the virtual measurement plane, similarly to step ST4 (step ST24). Specifically, the phase distribution calculator 74a calculates the set candidate value of the defocus distance Δz, the model intensity distribution I _mz (x, y), and the calculated intensity distributions I _cz−Δz (x, y), I _cz+Δz ( x, y), the calculated phase distribution φ _cz (x, y) is calculated by solving the intensity transport equation.

Next, the defocus distance determination unit 7a determines whether the calculated phase distribution φ _cz (x, y) has been calculated for all the candidate values Δz1, . . . , Δzn (step ST25). , Δzn are not yet completed, the defocus distance determining unit 7a sets the defocus distance Δz to the next candidate value in step ST22, and repeats step ST23 and step ST23. ST24 is executed in order.

In short, the defocus distance determination unit 7a changes the candidate value of the defocus distance Δz from step ST22 to step ST25, and determines the calculated intensity distribution I _cz−Δz on the virtual displacement plane shifted in the optical axis direction from the virtual measurement plane. (x, y), I _cz+Δz (x, y), and the calculated phase distribution φ _cz (x, y) by solving the intensity transport equation are calculated in order.

When the calculations for all the candidate values Δz1, _. step ST26). If the calculation has not yet been completed, the model setting section 71a sets the model phase distribution φ _mz (x, y) based on the second size parameter in step ST21. Then, the defocus distance determining section 7a again sequentially executes steps ST22 to ST25.

When the calculation of the model phase distribution φ _mz (x, y) of the second size parameter is completed, the defocus distance determination unit 7a calculates the defocus distance based on the calculated multiple phase distributions φ _cz (x, y). Δz is determined (step ST27). Specifically, as described above, the defocus distance determination unit 7a determines the upper limit value of the recommended range R1 based on the first correspondence relationship of the first size parameter, and determines the upper limit value of the recommended range R1 based on the second correspondence relationship of the second size parameter. , the lower limit of the recommended range R1 is determined, and the defocus distance Δz is set to a value within the recommended range R1.

As described above, according to the second embodiment as well, a phase image can be generated with an appropriate defocus distance Δz according to the size of the object 8 . Therefore, the phase image generation device 1 can generate a phase image with high accuracy. Moreover, according to the second embodiment, there is no need to create the correspondence information D in advance.

Here, steps ST21 to ST27 are associated with the first to fifth steps in the column of means for solving problems.

Step ST21 is executed twice according to the first size parameter and the second size parameter to set the model phase distribution φ _mz (x, y) for the first size parameter and the second size parameter, respectively. Therefore, in step ST21, the first model phase distribution, which is the model phase distribution φ _mz (x, y) on the virtual measurement plane corresponding to the first size parameter, and the model phase distribution φ _mz ( x, y) corresponds to the first step of setting the second model phase distribution.

Step ST22 is executed for the first size parameter and the second size parameter to calculate the calculated intensity distributions I _cz−Δz (x, y) and I _cz+Δz (x, y) respectively. Therefore, in step ST23, the first calculated intensity distribution on the virtual misalignment plane shifted in the optical axis direction from the virtual measurement plane by the candidate value of the defocus distance Δz is calculated based on the first model phase distribution, and the virtual misalignment plane corresponds to the second step of calculating the second calculated intensity distribution in based on the second model phase distribution.

Step ST23 is executed for the first size parameter and the second size parameter to calculate the calculated phase distribution φ _cz (x, y) for each. Therefore, in step ST23, by solving the intensity transport equation using the defocus distance Δz and the first calculated intensity distribution, the first calculated phase distribution on the virtual measurement plane is calculated, and the defocus distance Δz and the second calculated intensity distribution are calculated. This corresponds to the third step of calculating the second calculated phase distribution on the virtual measurement plane by solving the intensity transport equation using .

Steps ST22 to ST24 are repeatedly executed for each of the first size parameter and the second size parameter to calculate the calculated phase distribution φ _cz (x, y) for each candidate value of the defocus distance Δz. Therefore, steps ST22 to ST24 are a fourth step of obtaining the first calculated phase distribution and the second calculated phase distribution for each candidate value by performing the second step and the third step while changing the candidate value of the defocus distance Δz. corresponds to

A step ST27 is a fifth step of determining the lower limit value of the recommended range R1 based on the first calculated phase distribution for each candidate value and determining the upper limit value of the recommended range R1 based on the second calculated phase distribution for each candidate value. corresponds to

<Defocus decision by user>
In the above example, the defocus distance determination unit 7a automatically determines the defocus distance Δz, but this is not necessarily the case. For example, the defocus distance determining unit 7a may notify the user of the first correspondence and the second correspondence between the defocus distance Δz and the index of the calculation accuracy, and leave the final decision to the user.

FIG. 14 is a functional block diagram schematically showing a third embodiment of the configuration for determining the defocus distance Δz. The defocus distance determination unit 7a includes a display instruction unit 76a. The defocus distance determining unit 7a determines, for example, It is generated based on the correspondence information D, and the display instructing section 76a causes the display section 71 to display the first correspondence and the second correspondence.

Displaying the first correspondence and the second correspondence on the display unit 71 allows the user to determine a preferable numerical range for the defocus distance Δz. Then, the user inputs the value of the defocus distance Δz to the input unit 72 . The input section 72 outputs the input information (value) to the control section 7 . The defocus distance determination unit 7a determines the input information (value) input from the input unit 72 as the defocus distance Δz.

Alternatively, the defocus distance determination unit 7a may determine the recommended range R1 of the defocus distance Δz, and the display instruction unit 76a may cause the display unit 71 to display the recommended range R1. By displaying the recommended range R1 on the display unit 71, the user can more appropriately recognize the preferable recommended range R1 for the defocus distance Δz. The user then inputs a value within the recommended range R1 into the input unit 72 . The input section 72 outputs the input information (value) to the control section 7 . The defocus distance determination unit 7a determines the input information (value) input from the input unit 72 as the defocus distance Δz.

As described above, according to the third embodiment, the user can determine the defocus distance Δz, so usability can be improved.

<Acquisition unit>
In the example above, the user entered the size parameter of the object 8 into the input unit 72 . In this case, the input unit 72 is an example of the acquisition unit 9 that acquires the size parameter regarding the size of the object 8 . However, the acquisition unit 9 is not necessarily limited to this.

FIG. 15 is a diagram schematically showing an example of the acquisition unit 9. As shown in FIG. In the example of FIG. 15, the acquisition unit 9 is composed of the detector 5 and the control unit 7. FIG. The detector 5 acquires a captured image (intensity distribution) of the object 8 as described above, and outputs the captured image to the control unit 7 . Although the intensity distribution in the captured image is relatively uniform, slight intensity differences may occur. Although it is difficult to observe an accurate image of the object 8 from such a captured image, an assumed size of the object 8 can be obtained.

In the example of FIG. 15, the controller 7 includes a size calculator 7c. The size calculator 7c calculates the size of the object 8 included in the captured image by image processing. Although the image processing is not particularly limited, for example, processing such as binarization processing, labeling processing, edge enhancement processing, and template matching may be employed as appropriate. For example, the size calculator 7c extracts continuous edges from the image after edge enhancement processing, specifies the contour of the object 8, and calculates its size. Specifically, the size calculator 7c preferably calculates the smallest size among the sizes of a plurality of unit objects (for example, cells) as a first size parameter, and calculates the largest size as a second size parameter.

The defocus distance determination unit 7a determines the defocus distance Δz as described above based on the first size parameter and the second size parameter calculated by the size calculation unit 7c.

According to this, the user does not need to input the size of the target object 8, so the user's trouble can be reduced.

As described above, the phase image generation device 1, the phase image generation method, and the defocus distance determination method have been described in detail. not a thing It is understood that numerous variations not illustrated can be envisioned without departing from the scope of this disclosure. Each configuration described in each of the above embodiments and modifications can be appropriately combined or omitted as long as they do not contradict each other.

In the above specific example, the defocus distance Δz is determined based on the first size parameter and the second size parameter, but this is not necessarily the case. If the object 8 is a single object, the acquisition unit 9 may acquire one size parameter as the size of the object 8 . Then, the defocus distance determination unit 7a may determine the defocus distance Δz based on the correspondence relationship corresponding to the size parameter. For example, when the size of the object 8 is 800 μm, the defocus distance determination unit 7a determines the defocus distance Δz to be approximately 30 μm or more based on the correspondence relationship of the solid line graph in FIG. 9, for example.

In the above-described specific example, the defocus distance Δz is determined so that the error Δφ becomes small by comparing the calculation accuracy index (eg maximum phase φ _{c_max} ) with the true value (eg maximum phase φ _{m_max )} . However, the error Δφ also depends on parameters other than the defocus distance Δz. If this parameter is set inappropriately, the error Δφ may not be less than or equal to the threshold value Ref1 even if the defocus distance Δz is changed.

Therefore, the maximum phase _{φc_max} may be normalized and the normalized value may be compared with a preset threshold value. As a specific example, each maximum phase _{φc_max} is normalized by an average value within a range in which the maximum phase _{φc_max} changes in the vicinity of a constant value even when the defocus distance Δz changes. Then, the defocus distance Δz may be determined so that the standard value is within a predetermined range including 1 (for example, a range of 0.9 or more and 1.1 or less). According to this, the defocus distance Δz can be appropriately determined while avoiding the influence of other parameters.

Also, in the above example, the detector 5 detects light that has passed through the object 8, but this is not necessarily the case. The illumination unit 3 and the detector 5 may be provided on the same side of the object 8 , and the detector 5 may detect the light reflected and scattered by the surface of the object 8 .

Also, in the above example, the size parameter about the size of the object 8 is given to the defocus distance determination unit 7a of the control unit 7, but this is not necessarily the case. A worker may determine the defocus distance Δz according to the size of the object 8 in advance by the above method, and input the value of the defocus distance Δz to the input unit 72 .

1 phase image generation device 3 illumination unit 4 imaging optical system 41 objective lens 5 detector 7 control unit 71 display unit 72 input unit 73 storage unit 8 object 9 acquisition unit D correspondence information L1 illumination light ST10 to ST13 process (step )
ST21 First step (step)
ST23 Second process (step)
ST24 Third process (step)
ST27 Fifth process (step)

Claims (11)

A phase image generator that generates a phase image using an intensity transport equation,
an illumination unit that irradiates an object with illumination light;
an imaging optical system including an objective lens facing the object;
a detector that detects light incident from the object through the imaging optical system;
an acquisition unit that acquires a size parameter for the size of the object;
using the detector to detect light from the object at at least two positions separated from each other in the optical axis direction by a defocus distance determined based on the size parameter, the defocus distance and the detection; and a controller for generating a phase image of the object by solving an intensity transport equation using the light intensity distribution detected by the detector. The phase image generating device according to claim 1,
The acquisition unit acquires, as the size parameters, a first size parameter indicating a first size and a second size parameter indicating a second size larger than the first size,
The control unit determines an upper limit value of the recommended range of the defocus distance based on the first size parameter, determines a lower limit value of the recommended range of the defocus distance based on the second size parameter, A phase image generating device, wherein the defocus distance is determined to be a value within the recommended range. The phase image generating device according to claim 2,
The phase image generating device, wherein the control unit determines the defocus distance used for generating the phase image to be a half value of the sum of the lower limit value and the upper limit value. The phase image generating device according to claim 2 or 3,
Further comprising a storage unit that stores correspondence information about the size and the defocus distance,
The phase image generation device, wherein the control unit determines the upper limit based on the first size parameter and the correspondence information, and determines the lower limit based on the second size parameter and the correspondence information. The phase image generating device according to claim 4,
The phase image generation device, wherein the correspondence relationship information includes a correspondence relationship between the defocus distance for each of a plurality of sizes including the first size and the second size and an index indicating calculation accuracy of the phase image. . The phase image generating device according to claim 5,
The index of calculation accuracy includes information on the maximum phase of the calculated phase distribution on the virtual measurement plane calculated by simulation,
In the simulation, a model phase distribution on the virtual measurement plane is set, a calculated intensity distribution on a virtual misalignment plane shifted in the optical axis direction from the virtual measurement plane is calculated based on the model phase distribution, and the defocus A phase image generating device for calculating the calculated phase distribution by solving an intensity transport equation using the distance and the calculated intensity distribution. The phase image generating device according to claim 2 or 3,
The control unit
A first step of setting a first model phase distribution, which is a model phase distribution on a virtual measurement plane corresponding to the first size parameter, and a second model phase distribution, which is the model phase distribution corresponding to the second size parameter. and,
calculating a first calculated intensity distribution on a virtual misalignment plane shifted in the optical axis direction from the virtual measurement plane by the candidate value of the defocus distance based on the first model phase distribution; 2 a second step of calculating a calculated intensity distribution based on the second model phase distribution;
By solving the intensity transport equation using the defocus distance and the first calculated intensity distribution, the first calculated phase distribution in the virtual measurement plane is calculated, and the defocus distance and the second calculated intensity distribution are used. a third step of calculating a second calculated phase distribution in the virtual measurement plane by solving an intensity transport equation;
a fourth step of obtaining the first calculated phase distribution and the second calculated phase distribution for each candidate value by repeatedly performing the second step and the third step while changing the candidate value;
A fifth step of determining a lower limit value of the recommended range based on the first calculated phase distribution for each candidate value and determining an upper limit value of the recommended range based on the second calculated phase distribution for each candidate value. and a phase image generator. The phase image generating device according to claim 6 or claim 7,
The phase image generator, wherein the model phase distribution has a sinusoidal shape. The phase image generation device according to any one of claims 1 to 8,
The phase image generation device, wherein the acquisition unit acquires the size of the object based on the intensity distribution detected by the detector. A phase image generation method for generating a phase image using an intensity transport equation, comprising:
determining a defocus distance based on a size parameter for the size of the object;
Using a detector that irradiates the object with illumination light and detects light incident through an imaging optical system including an objective lens facing the object, separated from each other at the defocus distance in the optical axis direction detecting light from the object at at least two positions using the detector, and solving an intensity transport equation using the defocus distance and the intensity distribution of the light detected by the detector, and generating a phase image of said object. A method for determining a defocus distance for use in solving an intensity transport equation for generating a phase image, comprising:
setting a model phase distribution in a virtual measurement plane based on the size of the object;
While changing the candidate value of the defocus distance, a calculated intensity distribution on a virtual misalignment plane shifted in the optical axis direction from the virtual measurement plane is calculated based on the model phase distribution, and the calculated intensity distribution is used. obtaining a calculated phase distribution in the virtual measurement plane by solving an intensity transport equation;
determining the defocus distance to be used in solving the intensity transfer equation to generate the phase image based on the calculated phase distribution.

Images