Observation device and observation method

Abstract

To provide an observation device capable of generating a thickness distribution image with superior quantitativity in phase contrast imaging using a light source outputting light having a bandwidth.SOLUTION: An observation device 1 comprises a light source 10, an imaging part 12, a thickness distribution image generation part 13, and a thickness distribution image correction part 14. The imaging part 12 receive X rays, output from the light source 10 and passed through an observation object 20, on an imaging surface to acquire a phase contrast image. The thickness distribution image generation part 13 generates a thickness distribution image based upon the phase contrast image. The thickness distribution image correction part 14 corrects the thickness distribution image by repeating calculation of an intensity distribution of light based upon a spectrum and the thickness distribution image using the thickness distribution image generated by the thickness distribution image generation part 13 as an initial image, calculation of an update amount based upon the calculated intensity distribution and phase contrast image, and update of the thickness distribution image based upon the calculated update amount.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to an observation device and an observation method. [Background technology]

[0002] Phase contrast imaging is one of the techniques for observing an object. In phase contrast imaging, a phase contrast image can be obtained by irradiating the object with light output from a light source and receiving the light that has passed through the object with an imaging unit. In addition, a thickness distribution image (phase image) of the object can be generated based on this phase contrast image.

[0003] The light source used here is a point light source (or one that can be substantially regarded as a point light source). The observation object is made of a material that can transmit at least a part of the light output from the light source. In general, the observation object is made of a known single material. The phase contrast image acquired by the imaging unit represents the spatial change in the intensity of the transmitted light of the observation object, and is a two-dimensional image (edge-emphasized image) in which the intensity is high at the boundary (edge) where the thickness of the observation object changes. As a method for generating a thickness distribution image from a phase contrast image, the method described in Non-Patent Document 1 is known. This method assumes that the light output from the light source is a single wavelength in a narrow band (monochromatic case). This method is hereinafter referred to as the "monochromatic Paganin method."

[0004] Phase contrast imaging is applicable not only when the light output from the light source is a single wavelength in a narrow band (monochromatic case) but also when the light output from the light source has a certain band (multicolored case), and can be applied, for example, when an X-ray tube is used as a light source. When performing phase contrast imaging using a light source that outputs light (multicolored light) having a certain band, not limited to X-rays, a phase contrast image can be acquired, and a thickness distribution image can be generated based on this phase contrast image. In this case, the method described in Non-Patent Document 2 is known as a method for generating a thickness distribution image from a phase contrast image. The method described in Non-Patent Document 2 is an extension of the monochromatic Paganin method described in Non-Patent Document 1 to a multicolor method. This method extended to a multicolor method is hereinafter referred to as the "multicolor Paganin method." In addition to the method described in Non-Patent Document 2, several other multicolor Paganin methods have been proposed. [Prior art documents] [Non-patent literature]

[0005] [Non-Patent Document 1] Paganin, D., Mayo, SC, Gureyev, TE, Miller, PR & Wilkins, SW, J. Microsc. 206, 33-40(2002). [Non-Patent Document 2] Wenz, J., Schleede, S.,Khrennikov, K. et al., Nat. Commun. 6, 7568 (2015). Summary of the Invention [Problem to be solved by the invention]

[0006] Phase contrast imaging is considered to be more sensitive and less susceptible to noise than absorption imaging. However, when phase contrast imaging is performed using a light source that outputs light with a certain band (multi-color light), the quantitativeness of the thickness distribution image generated based on the phase contrast image is low.

[0007] The present invention has been made to solve the above problems, and has an object to provide an observation device and an observation method that are capable of generating a thickness distribution image with excellent quantitativeness when performing phase contrast imaging using a light source that outputs light having a band. [Means for solving the problem]

[0008] A first aspect of the observation device of the present invention includes a light source that outputs light having a band and irradiates it onto an observation object, an imaging unit that receives the light output from the light source and passed through the observation object to obtain a phase contrast image of the observation object, a thickness distribution image generating unit that generates a thickness distribution image of the observation object based on the phase contrast image, and a thickness distribution image correcting unit that corrects the thickness distribution image based on the spectrum of the light output from the light source. The thickness distribution image correcting unit repeats the following steps: using the thickness distribution image generated by the thickness distribution image generating unit as an initial image, calculating the intensity distribution of light after passing through the observation object based on the spectrum and the thickness distribution image, calculating an update amount based on the calculated intensity distribution and the phase contrast image, and updating the thickness distribution image based on the calculated update amount.

[0009] In a second aspect of the observation device of the present invention, in addition to the first aspect, the thickness distribution image generating unit normalizes the phase contrast image based on the background image acquired by the imaging unit, and generates a thickness distribution image based on this normalized phase contrast image.

[0010] In a third aspect of the observation device of the present invention, in addition to the first or second aspect, the thickness distribution image generating unit generates a thickness distribution image based on the phase contrast image by the Paganin method, which assumes that the light irradiating the observation object is monochromatic.

[0011] In a fourth aspect of the observation device of the present invention, in addition to the first or second aspect, the thickness distribution image generating unit generates a thickness distribution image based on the phase contrast image by a method that extends the Paganin method to the case where the light irradiating the observation object has a band.

[0012] A fifth aspect of the observation device of the present invention comprises, in addition to any one of the first to fourth aspects, a spectrum measuring section that measures the spectrum of light output from the light source.

[0013] In a sixth aspect of the observation device of the present invention, in addition to the fifth aspect, the imaging section and the spectrum measuring section are common.

[0014] A first aspect of the observation method of the present invention includes an imaging step of irradiating an observation object with light having a band output from a light source and receiving the light that has passed through the observation object by an imaging unit to obtain a phase contrast image of the observation object, a thickness distribution image generating step of generating a thickness distribution image of the observation object based on the phase contrast image, and a thickness distribution image correcting step of correcting the thickness distribution image based on the spectrum of the light output from the light source. In the thickness distribution image correcting step, the thickness distribution image generated in the thickness distribution image generating step is used as an initial image, and calculation of the intensity distribution of light after passing through the observation object based on the spectrum and the thickness distribution image, calculation of an update amount based on the calculated intensity distribution and the phase contrast image, and updating of the thickness distribution image based on the calculated update amount are repeated.

[0015] In a second aspect of the observation method of the present invention, in addition to the first aspect, in the thickness distribution image generating step, the phase contrast image is normalized based on the background image acquired by the imaging unit, and a thickness distribution image is generated based on this normalized phase contrast image.

[0016] In a third aspect of the observation method of the present invention, in addition to the first or second aspect, in the thickness distribution image generating step, a thickness distribution image is generated based on a phase contrast image by the Paganin method, which assumes that the light irradiating the object to be observed is monochromatic.

[0017] In a fourth aspect of the observation method of the present invention, in addition to the first or second aspect, in the thickness distribution image generating step, a thickness distribution image is generated based on a phase contrast image by a method that extends the Paganin method to the case where the light irradiating the object to be observed has a band.

[0018] A fifth aspect of the observation method of the present invention includes, in addition to any one of the first to fourth aspects, a spectrum measuring step of measuring the spectrum of light output from the light source by a spectrum measuring section.

[0019] In a sixth aspect of the observation method of the present invention, in addition to the fifth aspect, the imaging section and the spectrum measuring section are common. Effect of the Invention

[0020] According to the present invention, when phase contrast imaging is performed using a light source that outputs light having a band, a thickness distribution image with excellent quantitativeness can be generated. [Brief description of the drawings]

[0021] [Figure 1] FIG. 1 is a diagram showing the configuration of an observation device 1. As shown in FIG. [Diagram 2] FIG. 2 is a flow chart of the observation method. [Diagram 3] FIG. 3 is a diagram showing an example of a phase contrast image of the observation object acquired by the imaging section 12. As shown in FIG. [Figure 4] FIG. 4 is a diagram showing an example of a thickness distribution image of the observation object generated by the thickness distribution image generating unit 13 based on the phase contrast image shown in FIG. [Diagram 5]FIG. 5 is a diagram showing an example of the intensity distribution on a straight line passing through the center position of a bead in a thickness distribution image obtained when the tube voltage and tube current of the X-ray source are set to various values. [Figure 6] FIG. 6 is a diagram showing the intensity distribution in a thickness distribution image of an object to be observed, which is generated based on a phase contrast image by simulation. [Figure 7] FIG. 7 is a diagram showing an X-ray spectrum when the X-rays irradiated onto the observation object are broadband. [Figure 8] FIG. 8 is a flowchart of the thickness distribution image correcting step S4. [Figure 9] FIG. 9 is a flowchart of the light intensity distribution calculation step S12. [Figure 10] FIG. 10 is a diagram showing the intensity distribution and the like in a thickness distribution image of an object to be observed that is generated based on a phase contrast image by simulation. [Figure 11] FIG. 11 is a diagram showing the influence of the measurement accuracy of the spectrum of light output from a light source on the intensity distribution in a corrected thickness distribution image. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0022] Hereinafter, the embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the description of the drawings, the same elements are given the same reference numerals, and duplicated descriptions are omitted. The present invention is not limited to these examples, but is indicated by the claims, and is intended to include all modifications within the meaning and scope equivalent to the claims.

[0023] FIG. 1 is a diagram showing the configuration of an observation device 1. The observation device 1 includes a light source 10, a spectrum measuring unit 11, an imaging unit 12, a thickness distribution image generating unit 13, and a thickness distribution image correcting unit 14. FIG. 2 is a flowchart of an observation method. The observation method includes a spectrum measuring step S1 performed by the spectrum measuring unit 11, an imaging step S2 performed by the imaging unit 12, a thickness distribution image generating step S3 performed by the thickness distribution image generating unit 13, and a thickness distribution image correcting step S4 performed by the thickness distribution image correcting unit 14. The observation device 1 and the observation method can obtain a phase contrast image of an observation object 20 arranged between the light source 10 and the imaging unit 12, and generate a thickness distribution image (phase image) of the observation object 20 based on the phase contrast image. The observation object 20 is made of a known single material that can transmit at least a part of the light output from the light source 10.

[0024] The light source 10 is a point light source (or can be substantially regarded as a point light source). The light source 10 may be a microfocus X-ray source. In the following description, the light source 10 is assumed to be a microfocus X-ray source. When performing phase contrast imaging, the light source 10 outputs X-rays, which are divergent light, and irradiates the observation object 20. When performing spectrum measurement, the light source 10 causes the X-rays to enter the spectrum measurement unit 11 without passing through the observation object 20.

[0025] The spectrum measuring unit 11 measures the energy spectrum of the X-rays output from the light source 10 (spectrum measuring step S1). The X-ray spectrum measured by the spectrum measuring unit 11 is used when the thickness distribution image is corrected by the thickness distribution image correcting unit 14. The X-ray spectrum may also be used when the thickness distribution image generating unit 13 generates a thickness distribution image.

[0026] The spectrum measurement step S1 does not need to be performed first, but may be performed before any process that requires information on the X-ray spectrum. When the spectrum of the X-rays output from the light source 10 is known from a previous measurement, or when the X-ray spectrum corresponding to the tube voltage and tube current set in the light source 10 is known, the spectrum measurement unit 11 does not need to be provided, and the spectrum measurement step S1 does not need to be performed every time the imaging unit 12 acquires a phase contrast image.

[0027] The imaging unit 12 receives, on the imaging surface, the X-rays output from the light source 10 and transmitted through the observation object 20, and obtains a phase contrast image I of the observation object 20 (imaging step S2). In addition, the imaging unit 12 receives, on the imaging surface, the X-rays output from the light source 10 in a state in which the observation object 20 is not placed, and obtains a background image I in Note that phase contrast imaging using X-rays in a configuration such as that shown in FIG. 1 is sometimes called propagation-based phase contrast X-ray imaging or in-line phase contrast X-ray imaging.

[0028] As the spectrum measuring unit 11, for example, a semiconductor detector such as Si or CdTe, a photon-counting X-ray imaging sensor, or the like can be used. A photon-counting X-ray imaging sensor can also be used as the imaging unit 12, and the spectrum can be measured by using the imaging unit 12 to reduce light during spectrum measurement and counting photons. In these cases, the spectrum measuring unit 11 and the imaging unit 12 can be a common unit, and it is preferable because spectrum correction is not required. If the spectrum measuring unit 11 and the imaging unit 12 use devices made of the same elements, it is preferable because sensitivity correction is not required.

[0029] The thickness distribution image generating unit 13 generates a thickness distribution image of the observed object 20 based on the phase contrast image acquired by the imaging unit 12 (thickness distribution image generating step S3). The thickness distribution image correcting unit 14 corrects the thickness distribution image generated by the thickness distribution image generating unit 13 (thickness distribution image correcting step S4). Details of these processes will be described later.

[0030] The thickness distribution image generating unit 13 and the thickness distribution image correcting unit 14 may be a computer equipped with a processing unit (e.g., a CPU, etc.) that performs calculation processing, a memory unit (e.g., a hard disk drive, a ROM, a RAM, etc.) that stores programs and various data required for the calculation processing, a display unit (e.g., a liquid crystal display, etc.) that displays calculation conditions and calculation results, etc., and an input unit (e.g., a keyboard, a mouse, etc.) that accepts various inputs.

[0031] FIG. 3 is a diagram showing an example of a phase contrast image of an observation object acquired by the imaging unit 12. (a) shows a phase contrast image, and (b) shows the intensity distribution on a line passing through the center position of a bead in the phase contrast image. FIG. 4 is a diagram showing an example of a thickness distribution image of an observation object generated by the thickness distribution image generating unit 13 based on the phase contrast image shown in FIG. 3. (a) shows the thickness distribution image, and (b) shows the intensity distribution on a line passing through the center position of a bead in the thickness distribution image. The observation object used here is a spherical polystyrene bead with a diameter of 300 μm (the same applies to the following figures).

[0032] When generating a thickness distribution image (FIG. 4) based on the phase contrast image (FIG. 3), the monochromatic Paganin method (the following formula (1)) described in Non-Patent Document 1 was used. The parameters appearing in the following formula (1) are as follows. An xyz Cartesian coordinate system is set as shown in FIG. 1. The x-axis and y-axis are parallel to the imaging plane of the imaging unit 12. The z-axis is parallel to the direction from the light source 10 toward the imaging unit 12. r is a variable representing the two-dimensional position on the xy plane at the z-direction position where the observation object is placed. M is the magnification when acquiring the phase contrast image. The distance between the light source 10 and the observation object 20 is defined as the z axis.1 Let z be the distance between the observation object 20 and the imaging surface of the imaging unit 12. 2 Then, the magnification M is expressed by the following formula (2): I(Mr) is a phase contrast image acquired by the imaging unit 12. I in (Mr) is a background image acquired by the imaging unit 12 when the observation object 20 is not placed, and is used to normalize the phase contrast image I(Mr). T(r) is a thickness distribution image. k is the magnitude of the component parallel to the imaging plane of the imaging unit 12 among the wave vectors of light on the imaging plane. μ is the linear attenuation coefficient of the observation object 20. 1-δ is the refractive index of the observation object 20, and δ is generally 10 -6 F is the Fourier transform, and F -1 is the inverse Fourier transform.

[0033]

number

[0034]

number

[0035] Even if the phase contrast image contains noise as shown in FIG. 3, the effect of noise is reduced in the thickness distribution image as shown in FIG. 4. In this way, phase contrast imaging is resistant to noise and has high sensitivity. When the light output from the light source is a single wavelength in a narrow band, the quantitativeness of the thickness distribution image generated by the above formula (1) is excellent. However, when the light output from the light source has a certain band, the quantitativeness of the thickness distribution image generated by the above formula (1) is low.

[0036] FIG. 5 is a diagram showing an example of the intensity distribution on a straight line passing through the center position of the bead in a thickness distribution image obtained when the tube voltage and tube current of the X-ray source are set to various values. As shown in this figure, when the tube voltage of the X-ray source is different, the thickness distribution image also differs greatly. This is considered to be due to the following reason. When the tube voltage of the X-ray source is different, the spectrum of the X-ray output from the X-ray source is different. In addition, when the energy or wavelength of the X-ray is different, the values ​​of μ and δ of the observation object are different. However, in the above formula (1), the thickness distribution image is generated using the values ​​of μ and δ at a certain specific energy of the X-ray. For this reason, when phase contrast imaging is performed using a light source that outputs light having a certain band such as an X-ray source, the quantitativeness of the thickness distribution image generated based on the phase contrast image by the above formula (1) is low.

[0037] In order to solve such problems, it is conceivable to generate a thickness distribution image based on a phase contrast image using a method (multi-color Paganin method) that is an extension of the Paganin method intended to be applicable when the light irradiating the observation object has a band, instead of the monochromatic Paganin method of the above formula (1). The multi-color Paganin method described in Non-Patent Document 2 is known, and several other multi-color Paganin methods have also been proposed. The multi-color Paganin method described in Non-Patent Document 2 uses μ, which is a weighted average of μ and δ, which depend on the X-ray energy, in the above formula (1), by the X-ray spectrum. mean , δ mean This has been replaced with.

[0038] The following formula (3) is a formula for generating a thickness distribution image by the multi-color Paganin method devised by the present inventor. mean is the weighted average of μ by the X-ray spectrum S(λ) and is expressed by the following formula (4). mean is the weighted average of δ with the X-ray spectrum S(λ), and is expressed by the following formula (5), where λ is the wavelength. Formulas (4) and (5) are expressed as integral calculations, but can be found by summation calculations in numerical calculations.

[0039]

number

[0040]

number

[0041]

number

[0042] Fig. 6 is a diagram showing the intensity distribution in a thickness distribution image of an object to be observed, which is generated based on a phase contrast image by simulation. This diagram shows the intensity distribution on a straight line passing through the center position of the object to be observed (beads) in each of thickness distribution image A generated by the monochromatic Paganin method with X-rays of a single wavelength irradiated to the object to be observed, thickness distribution image B generated by the monochromatic Paganin method with X-rays of a wide band (multicolor) irradiated to the object to be observed, and thickness distribution image C generated by the multicolor Paganin method with X-rays of a wide band (multicolor) irradiated to the object to be observed. Fig. 7 is a diagram showing the X-ray spectrum when X-rays of a wide band (multicolor) are irradiated to the object to be observed.

[0043] The thickness distribution image A was generated by the monochromatic Paganin method (the above formula (1)) using the X-rays with a single wavelength (energy 8.4 keV) irradiated to the observation object and the μ and δ values ​​for this X-ray energy. The quantitativeness of this thickness distribution image A is excellent.

[0044] The thickness distribution image B was generated by the monochromatic Paganin method (above formula (1)) using a wide band of X-rays irradiated to the observation object, and the values ​​of μ and δ at an X-ray energy of 8.4 keV in that band. Compared to the thickness distribution image A, the thickness distribution image B shows a larger thickness overall, and is less quantitative. Furthermore, the thickness distribution image B shows that there is thickness even in the area just outside the observation object (area where the observation object does not exist).

[0045] The thickness distribution image C is obtained by irradiating the object with broadband X-rays, and weighting μ and δ with the X-ray spectrum (Fig. 7). mean , δ mean The thickness distribution image C was generated by the multi-color Paganin method (above formula (3)) using the values ​​of each of the above. Compared to thickness distribution image A, thickness distribution image C has a smaller overall thickness and is less quantitative. However, compared to thickness distribution image B, thickness distribution image C has improved quantitativeness in the thin-thickness region.

[0046] Thus, when the X-rays irradiated to the observation object are broadband (multicolor), the quantitativeness of the generated thickness distribution image is low whether the monochromatic Paganin method or the multicolor Paganin method is used. Therefore, in this embodiment, after the thickness distribution image generating step S3 in which the thickness distribution image generating unit 13 generates a thickness distribution image based on the phase contrast image, the thickness distribution image correcting unit 14 corrects the thickness distribution image in a thickness distribution image correction step S4, thereby improving the quantitativeness of the thickness distribution image.

[0047] 8 is a flowchart of the thickness distribution image correcting step S4. The thickness distribution image correcting step S4 includes an initialization step S11, a light intensity distribution calculation step S12, an update amount calculation step S13, an evaluation value calculation step S14, a determination step S15, a thickness distribution image updating step S16, a determination step S17, and a non-negative constraint step S18.

[0048] In the initialization step S11, the thickness distribution image T generated in the thickness distribution image generating step S3 is 0 The thickness distribution image T(r) is set as an initial image of the thickness distribution image T(r) in the subsequent repeated processing of steps S12 to S18. 0 is the background image I in It is preferable that the thickness distribution image T is generated based on the phase contrast image I after normalization by the above equation. 0 may have been generated using either the monochromatic or polychromatic Paganin method.

[0049] In the light intensity distribution calculation step S12, a light intensity distribution I' on the imaging surface of the imaging unit 12 is calculated based on the X-ray spectrum S(λ) and the thickness distribution image T(r). Specifically, when X-rays having the X-ray spectrum S(λ) output from the light source 10 are irradiated onto an observation object having a thickness distribution T(r), an intensity distribution I' of the X-rays that reach the imaging surface of the imaging unit 12 after passing through the observation object is calculated. The light intensity distribution calculation step S12 will be described in detail later.

[0050] In the update amount calculation step S13, an update amount ΔT(r) for the thickness distribution image T(r) is calculated based on the light intensity distribution I′ calculated in the light intensity distribution calculation step S12 and the phase contrast image I. The update amount ΔT(r) is expressed by the following formula (6).

[0051]

number

[0052] In evaluation value calculation step S14, an evaluation value ε is calculated to determine whether or not to end the repeated processing of steps S12 to S18. The evaluation value ε may be any index indicating whether or not the absolute value of the update amount ΔT(r) calculated in update amount calculation step S13 has become sufficiently small (i.e., whether or not the thickness distribution image T(r) has sufficiently converged) or whether or not the absolute value of the difference between the light intensity distribution I' and the phase contrast image I has become sufficiently small. The evaluation value ε may be expressed by, for example, the following formula (7) or (8).

[0053]

number

[0054]

number

[0055] In the judgment step S15, it is judged whether the evaluation value ε calculated in the evaluation value calculation step S14 is greater than the allowable value α. If the evaluation value ε is equal to or less than the allowable value α, the process of the thickness distribution image correction step S4 is terminated. If the evaluation value ε is greater than the allowable value α, the process proceeds to the thickness distribution image update step S16.

[0056] In the thickness distribution image updating step S16, the thickness distribution image T(r) is updated using the update amount ΔT(r) calculated in the update amount calculating step S13 (the following formula (9)).

[0057]

number

[0058] In the determination step S17, it is determined whether or not each pixel value of the thickness distribution image T(r) after being updated in the thickness distribution image update step S16 is equal to or greater than 0. If a pixel value of the thickness distribution image T(r) is negative, the pixel value is set to 0 in the non-negative constraint step S18, and the process returns to the light intensity distribution calculation step S12.

[0059] 9 is a flowchart of the light intensity distribution calculation step S12. The light intensity distribution calculation step S12 is a step of calculating the light intensity distribution on the imaging surface of the imaging unit 12 based on the X-ray spectrum S and the thickness distribution image, and includes an initialization step S21, a transmitted light complex amplitude calculation step S22, a light propagation calculation step S23, a light intensity distribution update step S24, a wavelength update step S25, and a determination step S26. Here, the energy spectrum S(E) of the X-ray output from the light source 10 is calculated based on the energy E 1 ~E N Let's say that each of them is required.

[0060] In the initialization step S21, the light intensity distribution I' on the imaging surface of the imaging unit 12 is initialized. That is, each pixel value of the light intensity distribution I' is set to 0. Also, in the initialization step S21, the energy E 1 ~E N The nth energy E nThe value of the index n representing this is set to 1.

[0061] In the transmitted light complex amplitude calculation step S22, the energy E n The complex amplitude distribution ψ immediately after the light passes through the object being observed. n Calculate (r). Energy E n Light intensity distribution I immediately after light passes through the object being observed n (r) is expressed by the following equation (10). Energy E n The amount of phase change φ caused by the light passing through the object being observed n (r) is expressed by the following equation (11). Energy E n The complex amplitude distribution ψ immediately after the light passes through the object being observed. n (r) is expressed by the following formula (12), where e is the elementary charge, c is the speed of light in a vacuum, and h is Planck's constant.

[0062]

number

[0063]

number

[0064]

number

[0065] In the light propagation calculation step S23, the energy E immediately after passing through the observation object calculated in the transmitted light complex amplitude calculation step S22 is calculated. n The complex amplitude distribution of light ψ n Based on (r), a light propagation calculation is performed to obtain the energy E n The complex amplitude distribution of light Ψ n The Fresnel diffraction calculation method, the angular spectrum method, etc. are used for this light propagation calculation.

[0066] In the light intensity distribution update step S24, the energy E n The complex amplitude distribution of light Ψ n The square of the absolute value of is cumulatively added to the light intensity distribution I' to update the light intensity distribution I' on the imaging plane of the imaging unit 12 (Equation (13) below).

[0067]

number

[0068] In wavelength update step S25, the value of n is incremented by 1, and the next energy is selected. In decision step S26, it is determined whether n is equal to or less than N. If n is greater than N, the process of light intensity distribution calculation step S12 is terminated. If n is equal to or less than N, the process returns to transmitted light complex amplitude calculation step S22.

[0069] The light intensity distribution I′ at the time when the process of the light intensity distribution calculation step S12 is completed is calculated by the X-ray spectrum S(E n When multi-wavelength X-rays having a thickness distribution T(r) are irradiated onto an object to be observed having a thickness distribution T(r), the intensity distribution of the X-rays that reach the imaging surface of the imaging unit 12 after passing through the object to be observed is I'.

[0070] FIG. 10 shows the intensity distribution and the like in a thickness distribution image of an object to be observed, which is generated based on a phase contrast image by simulation. (a) shows the intensity distribution on a straight line passing through the center position of the bead in the thickness distribution of the object to be observed (a spherical polystyrene bead with a diameter of 300 μm) used in the simulation. (b) shows the intensity distribution on a straight line passing through the center position of the bead in a thickness distribution image generated by the monochromatic Paganin method (formula (1) above) using the values ​​of μ and δ for the X-ray energy with a single wavelength (energy 8.4 keV) irradiated to the object to be observed (same as thickness distribution image A in FIG. 6). (c) shows the intensity distribution on a straight line passing through the center position of the bead in a thickness distribution image generated by the polychromatic Paganin method (formula (3) above) with a wide band X-ray irradiated to the object to be observed and then corrected in the thickness distribution image correction step S4.

[0071] As can be seen from these figures, the corrected thickness distribution image (c) has better quantitative performance than the thickness distribution images B and C in Figure 6. The corrected thickness distribution image (c) also has better quantitative performance than the thickness distribution image (b) that is not corrected as a single wavelength, and the quantitative performance is particularly improved in the area just outside the observation object (area where the observation object does not exist). The corrected thickness distribution image (c) matches extremely well with the actual thickness distribution (a).

[0072] Fig. 11 is a diagram showing the influence of the measurement accuracy of the spectrum of the light output from the light source on the intensity distribution in the thickness distribution image after correction. The degrees of agreement (correlation coefficient) between the measured spectrum and the true spectrum were set to values ​​of 1.0, 0.96, 0.78, and 0.55. Even when the degree of agreement was as low as 0.55, it was confirmed that the thickness distribution image after correction had superior quantitativeness compared to thickness distribution images B and C in Fig. 6.

[0073] Furthermore, when the effect of noise was examined, it was confirmed that the effect of noise on the quantitativeness of the corrected thickness distribution image was small. In other words, the characteristic of phase contrast imaging that it is robust against noise is maintained even when the thickness distribution image is corrected in the thickness distribution image correction step S4.

[0074] As described above, according to this embodiment, when phase contrast imaging is performed using a light source that outputs light having a band, a thickness distribution image with excellent quantitativeness can be generated. [Explanation of symbols]

[0075] 1... observation device, 10... light source, 11... spectrum measurement unit, 12... imaging unit, 13... thickness distribution image generation unit, 14... thickness distribution image correction unit, 20... observation object.

Claims

1. a light source that outputs light having a band and irradiates the object to be observed; an imaging unit that receives light output from the light source and transmitted through the observation object to obtain a phase contrast image of the observation object; a thickness distribution image generating unit that generates a thickness distribution image of the observation object based on the phase contrast image; a thickness distribution image correcting unit that corrects the thickness distribution image based on a spectrum of light output from the light source; Equipped with The thickness distribution image correction unit The thickness distribution image generated by the thickness distribution image generating unit is used as an initial image, repeating the steps of calculating an intensity distribution of light after passing through the object based on the spectrum and the thickness distribution image, calculating an update amount based on the calculated intensity distribution and the phase contrast image, and updating the thickness distribution image based on the calculated update amount; Observation equipment.

2. the thickness distribution image generating unit normalizes the phase contrast image based on the background image acquired by the imaging unit, and generates the thickness distribution image based on the normalized phase contrast image. The observation device according to claim 1.

3. the thickness distribution image generating unit generates the thickness distribution image based on the phase contrast image by a Paganin method assuming that the irradiating light to the observation object is monochromatic; The observation device according to claim 1.

4. the thickness distribution image generating unit generates the thickness distribution image based on the phase contrast image by a method obtained by expanding the Paganin method to a case where the irradiating light to the observation object has a band; The observation device according to claim 1.

5. Further comprising a spectrum measurement unit for measuring the spectrum of the light output from the light source. The observation device according to claim 1.

6. The imaging unit and the spectrum measuring unit are common to each other. The observation device according to claim 5.

7. an imaging step of irradiating an observation object with light having a band output from a light source and receiving the light that has passed through the observation object with an imaging unit to obtain a phase contrast image of the observation object; a thickness distribution image generating step of generating a thickness distribution image of the observation object based on the phase contrast image; a thickness distribution image correcting step of correcting the thickness distribution image based on a spectrum of light output from the light source; Equipped with In the thickness distribution image correcting step, The thickness distribution image generated in the thickness distribution image generating step is used as an initial image, repeating the steps of calculating an intensity distribution of light after passing through the object based on the spectrum and the thickness distribution image, calculating an update amount based on the calculated intensity distribution and the phase contrast image, and updating the thickness distribution image based on the calculated update amount; Observation method.

8. In the thickness distribution image generating step, the phase contrast image is normalized based on a background image acquired by the imaging unit, and the thickness distribution image is generated based on the normalized phase contrast image. The observation method according to claim 7.

9. In the thickness distribution image generating step, the thickness distribution image is generated based on the phase contrast image by a Paganin method assuming that the light irradiating the observation object is monochromatic. The observation method according to claim 7.

10. In the thickness distribution image generating step, the thickness distribution image is generated based on the phase contrast image by a method obtained by expanding the Paganin method to a case where the irradiating light to the observation object has a band. The observation method according to claim 7.

11. The method further includes a spectrum measuring step of measuring a spectrum of light output from the light source by a spectrum measuring unit. The observation method according to claim 7.

12. The imaging unit and the spectrum measuring unit are common to each other. The observation method according to claim 11.

Images