Calibration system and calibration method for calibration system

Abstract

The present disclosure relates to a calibration system, and a calibration method for a calibration system, with which it is possible to easily and highly accurately calibrate a spectroscopic measurement device. Light from a light source transmitted through a mask, on which a band-pass filter that transmits light in a plurality of different wavelength bands from among the light from the light source is disposed, is spectrally measured. The PSF for each wavelength used for calibration is calculated through signal processing based on a result of spectroscopic measurement by a spectroscopic measurement device. The band-pass filter that transmits light in a plurality of different wavelength bands and is disposed on the mask is arranged such that projection takes place in a state in which the respective PSFs for each of the different wavelengths do not overlap in the result of spectroscopic measurement. The present disclosure can be applied to a calibration system for a spectroscopic measurement device.

Description

CALIBRATION SYSTEM AND CALIBRATION METHOD FOR CALIBRATION SYSTEM - Patent application

[0001] The present disclosure relates to a calibration system and a calibration method for the calibration system, and more particularly to a calibration system and a calibration method for the calibration system that enable easy and highly accurate calibration of a snapshot-type spectroscopic measurement device.

[0002] A snapshot type spectroscopic measurement device has been proposed (see Patent Document 1).

[0003] A snapshot-type spectroscopic measurement device needs to be calibrated by acquiring a PSF (Point Spread Function) for each wavelength characteristic.

[0004] The method for obtaining the PSF for this calibration is generally to use an illumination source and multiple bandpass filters with different wavelength characteristics, and to obtain the PSF by repeatedly taking spectral images with a snapshot spectroscopic camera while switching between the bandpass filters.

[0005] However, if calibration is performed using this general method of acquiring the PSF, spectral imaging is required for the number of band-pass filters with multiple wavelength characteristics that have been prepared, which results in a lot of time and effort required for spectral imaging.

[0006] Therefore, a technology has been proposed in which lighting devices with different wavelength characteristics for each spatial position are arranged within the field of view of a snapshot-type spectroscopic measurement device, and when all lighting devices are turned on simultaneously, PSFs of multiple wavelength characteristics are obtained by signal processing from the results of a single spectroscopic image taken with the snapshot-type spectroscopic measurement device, thereby reducing the time and effort required for calibration (see Patent Document 2).

[0007] International Publication No. WO 2019 / 216213 International Publication No. WO 2022 / 196351

[0008] However, while the technology of Patent Document 2 makes it possible to acquire PSFs of all wavelength characteristics with a single image capture, PSFs of illumination with different wavelength characteristics may overlap and be projected onto the image sensor, which complicates signal processing and may cause errors in the acquired PSF due to the complex signal processing.

[0009] The present disclosure has been made in view of the above circumstances, and in particular, aims to easily and highly accurately calibrate a snapshot type spectroscopic measurement device.

[0010] A calibration system according to one aspect of the present disclosure is a calibration system for realizing calibration of a spectroscopic measurement device, and includes: a mask in which bandpass filters that transmit light in a plurality of different wavelength bands from light emitted by a light source are arranged; a spectroscopic measurement device that performs spectroscopic measurement of the light from the light source that has passed through the mask; and a signal processing unit that calculates a PSF for each wavelength used in the calibration by signal processing based on the spectroscopic measurement results of the spectroscopic measurement device, wherein the bandpass filters that transmit light in a plurality of different wavelength bands and are arranged on the mask are arranged so that the PSFs for the different wavelengths do not overlap in the spectroscopic measurement results.

[0011] A calibration method for a calibration system according to one aspect of the present disclosure is a calibration method for a calibration system that realizes calibration of a spectroscopic measurement device that includes a mask in which bandpass filters that transmit light in a plurality of different wavelength bands from light emitted by a light source are arranged, and a spectroscopic measurement device that performs spectroscopic measurement of light from the light source that has passed through the mask, the calibration method including performing signal processing based on spectroscopic measurement results of the spectroscopic measurement device to calculate a PSF for each wavelength used in the calibration, and the bandpass filters that transmit light in a plurality of different wavelength bands and are arranged on the mask are arranged such that the PSFs for the different wavelengths do not overlap in the spectroscopic measurement results.

[0012] In one aspect of the present disclosure, a mask is provided in which bandpass filters that transmit light in a plurality of different wavelength bands from a light source are arranged, and a spectroscopic measurement device is provided that performs spectroscopic measurement of the light from the light source that has passed through the mask, and a PSF for each wavelength used for the calibration is calculated by signal processing based on the spectroscopic measurement results of the spectroscopic measurement device, and the bandpass filters that transmit light in a plurality of different wavelength bands and are arranged on the mask are arranged so that the PSFs for the different wavelengths do not overlap in the spectroscopic measurement results.

[0013] FIG. 14 is a diagram illustrating a spectroscopic measurement device. FIG. 15 is a diagram illustrating calibration of the spectroscopic measurement device. FIG. 16 is a diagram illustrating calibration of the spectroscopic measurement device. FIG. 17 is a diagram illustrating an overview of calibration according to the present disclosure. FIG. 18 is a diagram illustrating an overview of calibration according to the present disclosure. FIG. 19 is a diagram illustrating an example configuration of a first embodiment of a calibration system according to the present disclosure. FIG. 19 is a diagram illustrating a mask generation method. FIG. 20 is a diagram illustrating a mask generation method. FIG. 21 is a flowchart illustrating PSF generation processing in the first embodiment. FIG. 22 is a flowchart illustrating mask design processing in the first embodiment. FIG. 23 is a diagram illustrating a mask pattern according to a second embodiment. FIG. 24 is a diagram illustrating multiple PSFs obtained using the mask of FIG. 13 and a PSF generated by interpolation from multiple PSFs. FIG. 25 is a diagram illustrating an example configuration of a second embodiment of a calibration system according to the present disclosure. FIG. 26 is a flowchart illustrating PSF generation processing in the second embodiment. FIG. 27 is a flowchart illustrating mask design processing in the second embodiment. FIG. 28 is a diagram illustrating an example configuration of a general-purpose computer.

[0014] Preferred embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functional configurations are designated by the same reference numerals, and redundant description will be omitted.

[0015] Hereinafter, embodiments of the present technology will be described in the following order: 1. Overview of the present disclosure 2. First embodiment 3. Second embodiment 4. Example of execution by software

[0016] <<1. Overview of the Present Disclosure>> The present disclosure provides for easy and highly accurate calibration of a snapshot spectroscopic camera. Before describing the overview of the present disclosure, the configuration of a snapshot spectroscopic camera will be described.

[0017] FIG. 1 shows an example of the configuration of a snapshot type spectroscopic measurement device.

[0018] The snapshot type spectroscopic measurement device 11 in FIG. 1 is composed of an objective lens 31, a slit (field stop) 32, a collimator lens 33, a diffractive optical element (DoE) , an imaging lens , and an image sensor .

[0019] The objective lens 31 focuses light from the measurement target onto a slit 32. The slit (field stop) 32 transmits the light focused by the objective lens 31 to a collimating lens 33. The collimating lens 33 converts the light that has passed through the slit 32 into parallel light, which then passes through a diffractive optical element (DoE) 34. The diffractive optical element 34 applies a diffraction phenomenon to the parallel light, thereby dispersing the light into wavelength units and transmitting the light through an imaging lens 35. The imaging lens 35 projects the spectral results Ps of the light in wavelength units onto the imaging surface of an image sensor (area sensor) 36.

[0020] With this configuration, the snapshot-type spectroscopic measurement device 11 records light of different wavelength components from different points on the measurement object in different elements (pixels) on the imaging surface of the image sensor (area sensor) 36 .

[0021] Superimposed light of various wavelengths from various positions on the measurement object is recorded as a spectral result Ps in each element (pixel) of the image sensor (area sensor) 36. Signal processing is then performed on this spectral result Ps to restore the positions (x, y) of the measurement object and the wavelengths of light corresponding to each position (λ: spectrum information), and a three-dimensional data cube 51 consisting of the spatial direction (XY) and wavelength direction (λ) of the measurement object is generated.

[0022] For details of this snapshot type spectroscopic measurement device 11, please refer to Patent Document 1.

[0023] In the spectroscopic measurement device 11 as described above, it is necessary to perform calibration by acquiring a PSF (Point Spread Function: response function, point spread function) for each wavelength and for each spatial position.

[0024] A method of acquiring the PSF in calibration involves, for example, disposing a light source 71 equipped with a band-pass filter that can be switched to a plurality of bands in front of the spectroscopic measurement device 11 as shown in the upper part of FIG. 2, and then, as shown in the lower part of FIG. 2, switching the band-pass filter to switch the wavelength, thereby obtaining the PSF for each wavelength as Hλ(X, Y).

[0025] However, the task of finding the PSF for each wavelength as Hλ(X, Y) while changing the bandpass filter requires capturing images the number of times equal to the number of bandpass filters, which is a time-consuming task.

[0026] Therefore, as shown in the upper part of FIG. 3, a technology has been proposed in which light sources 81, each equipped with a band-pass filter having a different wavelength characteristic for each position, are placed in front of the spectroscopic measurement device 11 and turned on all at once, and the imaging results obtained in one imaging are decomposed into a PSF: Hλ(X, Y) for each wavelength by signal processing (signal processing for feature decomposition in the figure) (see Patent Document 2).

[0027] However, in this method, the signal processing required to decompose the results of a single imaging session into a PSF for each wavelength: Hλ(X, Y) (wavelength feature amount) is complex, which can easily result in errors in the PSF being obtained, making it difficult to perform appropriate calibration.

[0028] Furthermore, as shown in the lower left of FIG. 4, a light source 91 having band-pass filters #1 to #n with different characteristics arranged in the order indicated by the arrows in the figure may be placed in front of the spectroscopic measurement device 11 as shown in the upper part of FIG. 4 and turned on all at once, and the imaging results obtained in one imaging session may be decomposed into a PSF: Hλ(X, Y) for each wavelength by signal processing (signal processing for feature decomposition in the figure).

[0029] However, in the configuration of Figure 4, the spectroscopic image captured by the spectroscopic measurement device 11 is as shown by the spectroscopic image Psnap_x in the lower right corner of Figure 4, and PSFs of light that has passed through different bandpass filters of first-order diffracted light may overlap and be projected, for example, as shown in the area surrounded by an ellipse.

[0030] For this reason, if an attempt is made to separate the PSFs of first-order diffracted light, which are projected in an overlapping state through different bandpass filters, into PSFs for each wavelength: Hλ(X, Y) through signal processing, appropriate separation will not be possible.

[0031] Therefore, in the present disclosure, a mask with bandpass filters arranged so that light transmitted through different bandpass filters does not overlap on the imaging plane of the spectroscopic measurement device 11 is set, and an image is taken with this mask arranged between the light source and the spectroscopic measurement device 11, and the imaging results are subjected to signal processing to appropriately separate the PSF for each wavelength.

[0032] That is, as shown in the upper part of Figure 5, a white light source 101, a mask 102, and a spectroscopic measurement device 11 are provided, and with the white light source 101 turned on, light transmitted through the mask 102 is spectroscopically imaged, and the PSF for each wavelength is determined from the spectroscopic image results by signal processing.

[0033] Here, FIG. 5 shows an example in which only bandpass filters BPF#a and BPF#b for wavelengths a and b are set on the mask 102.

[0034] Light emitted from the white light source 101 passes through bandpass filters BPF#a and BPF#b provided on the mask 102, and is captured by the spectroscopic measurement device 11 as an image Psnap_r shown in the lower left part of FIG.

[0035] In addition, Figure 5 shows an example in which light corresponding to the colors (black and gray) assigned to the bandpass filters BPF#a and BPF#b provided on the mask 102 is imaged as a projection point of the same color on the image Psnap_r.

[0036] As shown in the left part of FIG. 5, the light transmitted through the bandpass filters BPF#a and BPF#b is projected onto the image Psnap_r without overlapping.

[0037] That is, the patterns of the bandpass filters BPF#a and BPF#b in the mask 102 are arranged so that when the light passing through each bandpass filter BPF#a and BPF#b is focused in both the zeroth-order diffraction direction and the first-order diffraction direction on the image captured by the spectroscopic measurement device 11, the patterns do not overlap with each other.

[0038] Furthermore, by making the size of each bandpass filter BPF#a, BPF#b in the mask 102 as small as possible, the light that passes through the bandpass filters BPF#a, BPF#b is projected onto as few pixels as possible when projected onto the imaging surface of the spectroscopic measurement device 11.

[0039] With this configuration, if the imaging lens 35 of the spectroscopic measurement device 11 has a long focal length and is free of aberrations, each point on the image obtained by the calibration system of the present disclosure can be assigned to a corresponding wavelength band, as shown in Figure 6, and the image data cube after assignment becomes the PSF. If the imaging lens 35 of the spectroscopic measurement device 11 has a long focal length and is free of aberrations, the PSF will be the same regardless of spatial position, since the response function is the same even at different spatial positions for the same wavelength. Therefore, in the calibration system of the present disclosure, it is possible to obtain the PSF for each wavelength and each spatial position with high accuracy using a single imaging result Psnap.

[0040] Furthermore, if the imaging lens 35 of the spectroscopic measurement device 11 has a short focal length or has aberrations, it is necessary to obtain the PSF of the optical system for all spatial positions for each wavelength within the imaging field of view of the spectroscopic measurement device 11, so as shown in Fig. 7, a mask 102 consisting of bandpass filter patterns whose projection points do not overlap on the image sensor 36 of the spectroscopic measurement device 11 is provided with only patterns #1 to #n, and PSF_p1 to PSF_pn obtained by repeatedly capturing images while changing patterns #1 to #n are used to obtain the PSF for each wavelength and for each spatial position by interpolation processing using a PSF interpolation unit Inter. Note that patterns #1 to #n are a sufficiently small number relative to all spatial positions.

[0041] As described above, according to the present disclosure, it is possible to acquire the PSF for each wavelength and for each position with simple signal processing and with a small number of imaging attempts, thereby enabling simple and highly accurate calibration.

[0042] 2. First Embodiment Configuration Example in which the Lens of the Spectroscopic Measurement Apparatus Has a Long Focal Length and No Aberration Next, a first embodiment of a calibration system according to the present disclosure will be described with reference to FIG.

[0043] The calibration system 131 in FIG. 8 is composed of an illumination unit 151 , a mask 152 , a spectroscopic measurement device 153 , a bandpass filter setting unit 154 , a mask design unit 155 , a mask generation unit 156 , a recording unit 157 , and a signal processing unit 158 ​​.

[0044] The illumination 151 is an illumination having a broadband wavelength characteristic such as a halogen lamp, and is a light source that includes at least the wavelength range required for spectroscopic imaging. Therefore, if the wavelength range required for spectroscopic imaging is a specific range, it is sufficient that the illumination 151 includes the wavelength band of that specific range.

[0045] In this case, the specific range of wavelength band may be, for example, a near infrared (NIR) wavelength band, a short wavelength infrared (SWIR) wavelength band, a mid wavelength infrared (MWIR) wavelength band, a long wavelength infrared (LWIR) wavelength band, or a far infrared (FIR) wavelength band.

[0046] The mask 152 corresponds to the mask 102 described with reference to Fig. 5, and has holes formed at different spatial positions and bandpass filters with different wavelength characteristics attached. The pattern in which the bandpass filters are arranged in the mask 152 is such that, when light emitted by the illumination 151 passes through, the projected points do not overlap on the image sensor (configuration corresponding to the image sensor 36 in Fig. 1) in the spectroscopic measurement device 153. In Fig. 8, different waveforms represent different wavelength characteristics, and the holes, represented as white dots on the mask 152, represent the bandpass filters with different wavelength characteristics attached.

[0047] The spectroscopic measurement device 153 has a configuration corresponding to the spectroscopic measurement device 11 described above, and by transmitting light emitted from the illumination 151 through the mask 152, it captures a spectroscopic image in which multiple PSFs do not overlap, as described with reference to the image Psnap in Figure 6, for example, and outputs the image to the signal processing unit 158.

[0048] In the first embodiment, a case will be described in which the lens of the spectroscopic measurement device 153 (a lens corresponding to the imaging lens 35 in FIG. 1) has a long focal length and no aberration. A case in which the lens of the spectroscopic measurement device 153 has a short focal length or has aberration will be described later in a second embodiment.

[0049] The bandpass filter setting unit 154 sets the required number of bandpass filters according to the wavelength band and wavelength resolution input in advance, and supplies the number to the mask design unit 155 .

[0050] More specifically, for example, if the wavelength band required for the spectroscopic measurement device 153 is the visible light region of 450 nm to 780 nm and the wavelength resolution is 10 nm, bandpass filters must be set at intervals of 10 nm in bandwidth within the wavelength range of 450 nm to 780 nm, and therefore the bandpass filter setting unit 154 sets a total of 34 bandpass filters of 450 nm, 460 nm, 470 nm, ..., 770 nm, 780 nm as the required number.

[0051] The mask design unit 155 designs the arrangement of bandpass filters on the mask 152 for each of a plurality of different wavelength bands based on the required number of bandpass filters supplied from the bandpass filter setting unit 154, and outputs the design results to the mask generation unit 156 and the recording unit 157. Note that, hereinafter, design information related to the arrangement of the bandpass filters on the mask 152 will be referred to as mask design values. A specific method for designing the mask 152 will be described later.

[0052] The mask generation unit 156 generates a mask by drilling holes in the mask 152 and attaching a corresponding bandpass filter based on the mask design values ​​supplied from the mask design unit 155, and places the mask between the illumination 151 and the spectroscopic measurement device 153.

[0053] The description will be given on the assumption that the generation of the mask 152 in the mask generation unit 156 is automated, but automation is not essential, and the mask may be manually produced based on mask design values ​​and placed between the illumination 151 and the spectroscopic measurement device 153. A specific method for generating the mask 152 based on the mask design values ​​will be described in detail later.

[0054] The recording unit 157 records the mask design values ​​and supplies the recorded mask design values ​​to the signal processing unit 158 ​​as necessary.

[0055] The signal processing unit 158 ​​acquires the mask design values ​​recorded in the recording unit 157, and performs signal processing based on the mask design values ​​and the spectroscopic imaging results captured by the spectroscopic measurement device 153 to separate and output PSFs for each wavelength and position.

[0056] <Mask Design Method> Next, a description will be given of a method for designing the arrangement of the required number of bandpass filters to be set on the mask 152. Note that, although it is necessary to explain the components that make up the spectroscopic measurement device 153, since the configuration corresponds to that of the spectroscopic measurement device 11 in Fig. 1 , the spectroscopic measurement device 153 will also be explained as having the same configuration as the spectroscopic measurement device 11. That is, the spectroscopic measurement device 153 will be explained as being composed of an objective lens 31, a slit (field stop) 32, a collimator lens 33, a diffractive optical element (DoE) 34, an imaging lens 35, and an image sensor 36, just like the spectroscopic measurement device 11 in Fig. 1 .

[0057] First, when the parameters of the imaging lens 35 of the spectroscopic measurement device 153 are known, the projection position xi on the imaging plane of the image sensor 36 of the spectroscopic measurement device 153 can be identified by the incident position xm (similar to the spatial position on the mask 152 in this case) of the end of the incident light (parallel light) on the diffractive optical element 34 and the diffraction angle θ of the diffractive optical element. Furthermore, since the diffraction angle θ of the diffractive optical element can be expressed as a function g(λ) that depends on the wavelength λ, the incident position xi on the image sensor can be expressed, for example, by the following equation (1) using the spatial position xm on the mask 152 and the wavelength λ. Note that for details of equation (1), see Patent Document 1.

[0058] xi=F(x0,λ) =((f-d1) / f)(d2・tan(g(λ))+xm)+d1・tan(g(λ))...(1)

[0059] Here, xi is the incident position on the image sensor 36 of the spectroscopic measurement device 153, and xm is the incident position on the diffractive optical element 34 of the spectroscopic measurement device 153 (spatial position on the mask 152). Also, f is the focal length of the imaging lens 35, d1 is the distance between the diffractive optical element 34 and the imaging lens 35, d2 is the distance between the imaging lens 35 and the image sensor 36, and g(λ) is the diffraction angle θ expressed as a function of wavelength λ.

[0060] For the sake of simplicity, equation (1) represents only one dimension in the x direction, but in reality it is treated as coordinates (x, y) in two-dimensional space.

[0061] That is, from equation (1), for any wavelength λ, the incident position xi on the image sensor 36 of the spectroscopic measurement device 153 is specified by the incident position xm on the diffractive optical element 34 of the spectroscopic measurement device 153 (spatial position on the mask 152).

[0062] Therefore, it is conceivable to design the mask 152 by setting bandpass filters corresponding to different wavelengths λa and λb as incident positions (spatial positions on the mask 152) xm1 and xm2 (≠xm1) on the diffractive optical element 34 of different spectroscopic measurement devices 153 on the mask 152, and determining the incident positions xia and xib on the image sensor 36.

[0063] That is, as shown in the following equations (2) and (3), when the incident positions (spatial positions on the mask 152) of different wavelengths λa and λb on the diffractive optical element 34 are set as xm1 and xm2 (≠xm1), the incident positions xia and xib on the image sensor 36 are expressed as follows:

[0064] xia=F(x0,λa) =((f-d1) / f)(d2・tan(g(λa))+xm1)+d1・tan(g(λa))...(2)

[0065] xib=F(x0,λb) =((f-d1) / f)(d2・tan(g(λb))+xm2)+d1・tan(g(λb))...(3)

[0066] That is, if the incident positions xia and xib on the image sensor 36 calculated by equations (2) and (3) are always such that xia≠xib, then it is possible to design a mask 152 in which the PSFs do not overlap, as described with reference to Figure 5.

[0067] However, if an attempt is made to mechanically assign the incident position xm and the wavelength λ, the incident positions may partially coincide, such as xia=xib.

[0068] Therefore, in the present disclosure, as described above, the mask design unit 155 designs the positions of the bandpass filters of different wavelength bands on the mask 152 so that light passing through the bandpass filters of different wavelength bands on the mask 152 does not enter at the same position on the image sensor 36.

[0069] More specifically, the mask design unit 155 sets the available area on the mask 152 and the available area on the image sensor 36, and randomly sets the bandpass filter position (xmn, ymn, λn) for the nth wavelength λn from the available area on the mask 152. Note that here, the required number of bandpass filters is N, and n (1≦n≦N) is used as an identifier to identify each bandpass filter, and the positions are set in ascending order of identifier, that is, starting from the n=1 bandpass filter.

[0070] Next, the mask design unit 155 sets the incident position (xinp, yinp) (1≦p≦P) on the image sensor 36 where light that has passed through the position (xmn, ymn, λn) of the nth bandpass filter is incident, and records the association between the bandpass filter position (xmn, ymn, λn) on the mask 152 and the incident position (xinp, yinp) on the image sensor 36 as a mask design value in the recording unit 157.

[0071] Here, P is a code that identifies the incident position obtained by the zeroth-order diffraction and the first-order diffraction, and for example, if there is one incident position obtained by the zeroth-order diffraction and eight incident positions obtained by the first-order diffraction, the total number of incident positions is nine, so P = 9. Furthermore, in this case, taking into consideration that there is a certain degree of spread when the light enters the image sensor 36, it is also possible to set the incident position on the image sensor 36 by adding several pixels in the horizontal and vertical directions from the obtained coordinate position.

[0072] Then, the mask design unit 155 determines and updates the available positions on the mask 152 from the available area on the image sensor 36 based on the wavelength λ(n+1) of the λ(n+1)th bandpass filter whose identifier n is the next (n+1)th.

[0073] That is, from the available positions on the image sensor 36, the available positions on the mask 152 are determined by back-calculating the above-mentioned equation (1) using the wavelength λ(n+1) of the λ(n+1)th bandpass filter whose identifier n is the next (n+1)th.

[0074] In other words, from the available area where there is no problem when projecting onto the image sensor 36, the available area where there is no problem when placing the wavelength λ(n+1) of the λ(n+1)th bandpass filter whose identifier n is the next (n+1)th on the mask 152 is determined.

[0075] This relationship can be expressed by the following equation (4): where the pixel set of the usable area on the image sensor 36 is X' and the wavelength of the (n+1)th bandpass filter is λ(n+1), then the set X of the usable area on the mask can be expressed by the following equation (4):

[0076] X=(X'-d1・tan(g(λ(n+1))))・(f / (f-d1)) -d2・tan(g(λ(n+1)))...(4)

[0077] By repeating the above process for N bandpass filters, it is possible to design N types of bandpass filters on the mask 152 so that the transmitted light does not overlap on the image sensor 36.

[0078] <Mask Generation Method> Next, a method for generating the mask 152 based on the mask design values ​​by the mask generation unit 156 will be described.

[0079] The mask generation unit 156 determines the size of the mask 152 based on the size of the bandpass filter (the holes in which it is formed), the pixel range of the projection points on the image sensor 36 of the zeroth-order diffraction caused by light that has passed through the bandpass filter, and the size of the range in which the projection points on the image sensor 36 of the zeroth-order diffraction are distributed.

[0080] For example, as shown in the left part of Figure 9, consider the case where the vertical and horizontal dimensions of the mask 152 are rm x cm, and the hole 152h in which the bandpass filter is provided is circular with a diameter d, and how to determine the dimension rm x cm of this mask 152.

[0081] In this case, as shown in the right part of Figure 9, if the projection point 36h on the image sensor 36 of the zeroth-order diffracted light of the light that has passed through the band-pass filter is a circular pixel range with a diameter dp (pixels), and the number of vertical and horizontal pixels in the range (a rectangular range surrounding) in which the projection points of the zeroth-order diffraction on the image sensor 36 (the multiple white parts in the figure including the projection point 36h) may be distributed is rp (pixels) x cp (pixels), the mask generation unit 156 determines the vertical and horizontal size rm x cm of the mask 152 by calculating the following equation (5).

[0082] rm=(rp / dp)×d cm=(cp / dp)×d (5)

[0083] The mask generation unit 156 generates the mask 152 by drilling holes according to the size of each bandpass filter at the position of the mask 152 calculated by the above-mentioned equation (5) based on the mask design values ​​supplied by the mask design unit 155 and attaching the corresponding bandpass filter.

[0084] For example, as shown in the left part of Figure 10, if the number of vertical and horizontal pixels in the range (a rectangular range surrounding) in which the projection points of zero-order diffraction (the multiple white parts in the figure including projection point 36h) on image sensor 36 are distributed is rp (pixels) x cp (pixels), and the projection points 36h-1 and 36h-2 of light that has passed through two band-pass filters designed by mask design unit 155 are circles with a diameter dp, then mask 152 will be generated as shown in the right part of Figure 10.

[0085] That is, as shown in the right part of Figure 10, the vertical x horizontal size of the mask 152 is set to rm x cm as calculated by equation (5), and the mask generation unit 156 forms circular holes 152h-1 and 152h-2 of radius d at positions on the mask 152 where the design positions of the two band pass filters correspond to the projection points 36h-1 and 36h-2, and then pastes the corresponding band pass filters BPF1 and BPF2 so as to fill the holes 152h-1 and 152h-2.

[0086] With this configuration, even if the light emitted by the illumination 151 passes through the band-pass filters BPF1 and BPF2 provided in the hole 152h of the mask 152 and is projected onto the image sensor 36, the PSFs for each band generated by passing through each band-pass filter BPF1 and BPF2 do not overlap.

[0087] As a result, the signal processing unit 158 ​​can easily and accurately determine the PSF for each wavelength and for each position by signal processing based on a single spectroscopic image captured by the spectroscopic measurement device 153.

[0088] More specifically, the signal processing unit 158 ​​reads out (xmn, ymn, λn) recorded in the recording unit 157 as the mask design values ​​by the mask design unit 155, in association with the incident position (xinp, yinp) on the image sensor 36.

[0089] Then, based on the mask design values, the signal processing unit 158 ​​searches for the projection point closest to the incident position (xinp, yinp) on the image sensor 36 for the coordinates of each projection point on the image sensor 36, and identifies the PSF of the wavelength λn corresponding to the projection point on the image sensor 36 from the (xmn, ymn, λn) recorded in correspondence with the closest projection point.

[0090] The signal processing unit 158 ​​compares the incident position (xinp, yinp) on the image sensor 36 recorded in the recording unit 157 with the coordinates of each projection point to search for a match, and can specify the wavelength λn from the correspondingly recorded coordinate position and wavelength information (xmn, ymn, λn) of the bandpass filter on the mask 152, thereby specifying the PSF with only a small processing load and making it possible to determine the PSF with high accuracy.

[0091] Furthermore, even if the number of bandpass filters increases, the PSFs do not overlap when designing the mask 152, so it is possible to achieve high-precision calibration with simple processing even with a single spectroscopic imaging.

[0092] <PSF Generation Processing When the Lens of the Spectroscopic Measurement Device Has a Long Focal Length and No Aberration> Next, with reference to the flowchart in FIG. 11 , a PSF generation processing by the calibration system 131 when the lens of the spectroscopic measurement device 153 (a lens corresponding to the imaging lens 35 in FIG. 1 ) has a long focal length and no aberration will be described.

[0093] In step S 11 , the bandpass filter setting unit 154 sets the required number N of bandpass filters in accordance with the required wavelength band and wavelength resolution that have been input in advance, and supplies this to the mask design unit 155 .

[0094] In step S12, the mask design unit 155 executes a mask design process to design the arrangement of bandpass filters for each of a plurality of different wavelength bands on the mask 152, based on the required number N of bandpass filters supplied from the bandpass filter setting unit 154. Details of the mask design process will be described later with reference to the flowchart in FIG.

[0095] In step S13, the mask design unit 155 outputs the arrangement of bandpass filters on the mask 152 for each of the different wavelength bands, which is design information, to the mask generation unit 156 as mask design values ​​by correlating them with the bandpass filters on the mask 152 and their positions on the sensor, and also records them in the recording unit 157.

[0096] In step S14, the mask generation unit 156 determines the size of the mask 152 based on the mask design values ​​supplied by the mask design unit 155, and generates the mask 152 by drilling holes for placing band-pass filters and attaching the band-pass filters to the corresponding positions.

[0097] In step S15, the mask generation unit 156 places the generated mask 152 between the illumination 151 and the spectroscopic measurement device 153 and causes light to be emitted from the illumination 151. At this time, the spectroscopic measurement device 153 captures a spectroscopic image projected by the light emitted from the illumination 151 passing through the mask 152, and outputs the captured image result to the signal processing unit 158.

[0098] In step S16, the signal processing unit 158 ​​acquires the design results recorded in the recording unit 157, and performs signal processing based on the mask design values ​​and the spectroscopic imaging results captured by the spectroscopic measurement device 153 to separate and output PSFs for each wavelength and position.

[0099] As a result of the above processing, the PSFs that have passed through bandpass filters of different wavelength bands do not overlap in the spectral imaging results, making it possible to identify and separate the PSFs of each wavelength band based on the mask design values.

[0100] As a result, if the lens of the spectroscopic measurement device has a long focal length and is free of aberration, it is possible to obtain the PSF for each wavelength with high accuracy through simple signal processing in a single spectroscopic imaging.Furthermore, if the lens of the spectroscopic measurement device has a long focal length and is free of aberration, the PSF is constant regardless of spatial position, and the obtained PSF can be applied to all spatial positions, so it is possible to obtain the PSF for each wavelength and for each spatial position through simple signal processing in a single spectroscopic imaging.

[0101] As a result, it is possible to easily and highly accurately calibrate the spectroscopic measurement device 153.

[0102] <Mask Design Processing When the Spectroscopic Measurement Apparatus Lens Has a Long Focal Length and No Aberration> Next, with reference to the flowchart in FIG. 12, a mask design processing when the spectroscopic measurement apparatus lens has a long focal length and no aberration will be described.

[0103] In step S31, the mask design unit 155 initializes (n=1) an identifier n that identifies the available area on the mask 152, the available area on the image sensor 36, and the bandpass filter, and sets a number n that identifies the required number of bandpass filters.

[0104] In step S32, the mask design unit 155 randomly sets the bandpass filter position (xmn, ymn, λn) for the n-th wavelength λn from the available area on the mask 152.

[0105] In step S33, the mask design unit 155 sets the incident position (xinp, yinp) (1≦p≦P) on the image sensor 36 where light that has passed through the position (xmn, ymn, λn) of the nth bandpass filter is incident.

[0106] In step S34, the mask design unit 155 associates the bandpass filter position (xmn, ymn, λn) on the mask 152 with the incident position (xinp, yinp) on the image sensor 36 and records the association in the recording unit 157 as mask design values.

[0107] In step S35, the mask design unit 155 updates the available areas on the image sensor 36 that have not been set as incident positions onto which light is incident after passing through the band-pass filter.

[0108] In step S36, the mask design unit 155 calculates the available area on the mask 152 by back-calculating equation (1) from the available area on the image sensor 36 based on the wavelength λ(n+1) of the λ(n+1)th bandpass filter whose identifier n is the next (n+1)th, and updates the available area on the mask 152.

[0109] In step S37, the mask design unit 155 increments the identifier n by one to update it.

[0110] In step S38, the mask design unit 155 determines whether the identifier n is N+1, which is greater than the required number N of band-pass filters.

[0111] In step S38, if the identifier n is not N+1, which is greater than the required number N of band-pass filters, the process returns to step S32, and the subsequent processes are repeated until the identifier n exceeds the required number N of band-pass filters.

[0112] Then, in step S38, if it is determined that the identifier n is N+1, which is greater than the required number N of band-pass filters, the process ends.

[0113] That is, by the above processing, an arrangement of holes is designed on the mask 152 to provide a set number of bandpass filters through which the incident light passes and the PSF is projected without overlapping on the image sensor 36 of the spectroscopic measurement device 153 after the incident light passes through the bandpass filters.

[0114] This makes it possible for the spectroscopic measurement device 153 to acquire the PSF for each wavelength and for each position by simply performing spectroscopic imaging once, thereby enabling calibration of the spectroscopic measurement device 153 to be achieved simply and with high accuracy.

[0115] <<3. Second Embodiment>> Next, as a second embodiment, a calibration method when the imaging lens 35 of the spectroscopic measurement device 153 has a short focal length or when there is aberration will be described.

[0116] If the imaging lens 35 of the spectroscopic measurement device 153 has a short focal length or has aberration, the PSF will have spatial instability and will therefore differ for each spatial position and for each wavelength.

[0117] However, if the PSF is to be measured for each spatial position, it is necessary to repeat spectroscopic imaging while changing the spatial position, which increases the burden of spectroscopic imaging.

[0118] Therefore, in the present disclosure, as described with reference to FIG. 7 , the PSF is measured at a plurality of spatial positions for each wavelength band, and then the PSF for each spatial position is obtained by interpolation using the PSF measurement results at the plurality of spatial positions for each wavelength.

[0119] More specifically, as shown in Fig. 13, for example, the mask 152 is divided into spatially equal regions. Fig. 13 shows four types of masks 152#1 to 152#4, each divided into four regions and assigned region numbers 1 to 4 (numbers are given in circles in the figure). Note that the number of divisions of the mask 152 is four, which is merely an example, and the mask 152 may be divided into more than this number. The greater the number of divisions, the higher the accuracy of the PSF for each wavelength and each spatial position that is ultimately obtained. However, each divided region must be spatially equal.

[0120] Then, a wavelength range of the bandpass filter is assigned to each of the regions set by the region number.

[0121] For example, if the wavelength band of the bandpass filters is 450 nm to 780 nm, the resolution is 10 nm, and 34 bandpass filters are required, then 450 nm to 530 nm (9 filters) bandpass filters may be assigned to region number 1, 540 nm to 620 nm (9 filters) bandpass filters may be assigned to region number 2, 630 nm to 710 nm (9 filters) bandpass filters may be assigned to region number 3, and 720 nm to 780 nm (7 filters) bandpass filters may be assigned to region number 4.

[0122] Similarly, when the mask 152 is divided into five or more pieces, the mask 152 is spatially divided into a plurality of equal regions, and a wavelength band is assigned to each region for the bandpass filter.

[0123] 13, masks 152 of four types of patterns are set so that the regions with region numbers 1 to 4 are arranged in different positions, respectively. In Fig. 13, masks 152#1 of a pattern arranged with region number 1243, mask 152#2 of a pattern arranged with region number 4123, mask 152#3 of a pattern arranged with region number 3421, and mask 152#4 of a pattern arranged with region number 2341 are shown clockwise from the top left.

[0124] As shown in FIG. 13, the bandpass filters are allocated in units of bandwidth and divided into regions, so that PSFs of the same (or nearby) wavelength bands are generated at positions as spatially separated as possible among the PSFs obtained from the four different types of masks.

[0125] Furthermore, the masks 152#1 to 152#4 of the four different patterns are sequentially switched and placed in front of the illumination 151, and spectroscopic imaging is performed by the spectroscopic measurement device 153.

[0126] In this case, for example, the zeroth-order diffracted light of the PSF of wavelength λi assigned to region number 1 is acquired as shown by the four PSFλi,1 to PSFλi,4 shown in the upper part of FIG.

[0127] In the upper part of Figure 14, the zeroth-order diffracted light in the wavelength band assigned to area number 1, starting from left, is projected onto the image sensor 36 as PSFλi,1 to PSFλi,4, starting from left, at different positions corresponding to the area with area number 1 on masks 152#1 to 152#4.

[0128] That is, the zero-order diffracted light transmitted through the band-pass filter of wavelength λi is projected onto spatially different positions on the image sensor 36 .

[0129] Therefore, by PSF interpolation processing, which is a known technique using PSFλi,1 to PSFλi,4, the zeroth-order diffracted light at other spatially different positions is obtained as shown in the lower left part of FIG.

[0130] In this way, even if the imaging lens 35 of the spectroscopic measurement device 153 has a short focal length or has aberration, it is possible to obtain the PSF for each wavelength and for each spatial position.

[0131] <Configuration Example When the Lens of the Spectroscopic Measurement Apparatus Has a Short Focal Length or When There is Aberration> Next, a second embodiment of the calibration system of the present disclosure will be described with reference to FIG.

[0132] In the calibration system 131' of FIG. 15, components having the same functions as those in the calibration system 131 of FIG. 8 are denoted by the same reference numerals, and the description thereof will be omitted as appropriate.

[0133] The calibration system 131′ in FIG. 15 differs from the calibration system 131 in FIG. 8 in that, instead of the mask design unit 155, the mask generation unit 156, the recording unit 157, and the signal processing unit 158, a mask design unit 155′, a mask generation unit 156′, a recording unit 157′, and a signal processing unit 158′ are provided, and further, a mask region division unit 201 and a PSF interpolation unit 202 are provided.

[0134] The mask area dividing unit 201 divides the area of ​​the mask 152 into a predetermined number of divisions, spatially equally dividing the area, and assigns area numbers to each area and a band-pass filter to each area. Furthermore, the mask area dividing unit 201 sets different mask patterns for each divided area. That is, as described with reference to FIG. 13 , when the area of ​​the mask 152 is divided into four areas, four mask patterns, masks 152#1 to 152#4, with different patterns arranged in different positions for the divided areas, are set.

[0135] The mask design unit 155', mask generation unit 156', recording unit 157', and signal processing unit 158' have the same basic functions as the mask design unit 155, mask generation unit 156, recording unit 157, and signal processing unit 158, but differ in the processing involved in dividing the mask 152 into multiple regions and generating multiple masks 152 by changing the positions of the divided regions.

[0136] That is, the mask design unit 155′ designs multiple masks in which the divided areas are arranged at different positions corresponding to the number of divisions of the mask 152, for example, as described with reference to FIG. 13, and outputs the mask design values ​​of the multiple masks to the mask generation unit 156′ and records them in the recording unit 157′.

[0137] The mask generation unit 156' generates the masks 152 of the multiple patterns based on the mask design values ​​of the masks 152 of the multiple patterns, as described with reference to Figure 13, and switches between the masks 152 of the multiple patterns to cause the spectroscopic measurement device 153 to capture images.

[0138] The recording unit 157' records the mask design values ​​of the mask 152 of a plurality of patterns and outputs them to the signal processing unit 158' as appropriate.

[0139] The signal processing unit 158′ performs signal processing on the spectroscopic imaging results captured by the spectroscopic measurement device 153 each time the masks 152 of the multiple patterns generated by the mask generation unit 156′ are switched, thereby determining the PSF and supplying it to the PSF interpolation unit 202.

[0140] That is, each time the masks 152 of the multiple patterns are switched, spectroscopic imaging is performed, and for example, a PSF of a predetermined wavelength λi is imaged for each of the masks 152 of the multiple patterns. From each spectroscopic imaging result, multiple PSFs with different spatial positions, as described with reference to FIG. 14, are calculated and output to the PSF interpolation unit 202.

[0141] The PSF interpolation unit 202 performs interpolation processing based on a plurality of PSF results supplied from the signal processing unit 158', and determines and outputs a PSF for each wavelength and for each spatial position.

[0142] For details of the PSF interpolation process performed by the PSF interpolator 202, see Petr Janout, et al., Interpolation methods for the improvement of the point spread function estimation, 2018 28th International Conference Radioelektronika (RADIOELEKTRONIKA).

[0143] <PSF generation process when the lens of the spectroscopic measurement device is short-focus or has aberration> Next, with reference to the flowchart in Figure 16, we will explain the PSF generation process by the calibration system 131' when the lens of the spectroscopic measurement device 153 (the lens corresponding to the imaging lens 35 in Figure 1) is short-focus or has aberration.

[0144] In step S 51 , the bandpass filter setting unit 154 sets the required number N of bandpass filters in accordance with the wavelength band and wavelength resolution input in advance, and supplies the number N to the mask area dividing unit 201 .

[0145] In step S52, the mask area dividing unit 201 spatially divides the area of ​​the mask 152 into a predetermined number of equal parts, allocates the required number of band-pass filters to the number of divisions, and supplies the allocated band-pass filters to the mask design unit 155' together with information about the divided areas. For example, if the number of divisions is 4, and the required number of band-pass filters is N, the mask area dividing unit 201 may allocate N / 4 band-pass filters to each area. At this time, the mask area dividing unit 201 sets patterns in which the divided areas are arranged at different positions for the number of divisions.

[0146] In step S53, the mask design unit 155' acquires information on the divided regions and the number of allocated band-pass filters supplied from the mask region dividing unit 201, and executes a mask design process to design the arrangement of band-pass filters for each of a plurality of different wavelength bands on the mask 152. At this time, the mask design unit 155' designs the mask 152 including a plurality of patterns in which the arrangement of the divided regions is interchanged. The mask design process will be described in detail below with reference to the flowchart of FIG.

[0147] In step S54, the mask design unit 155 associates the bandpass filters on the mask 152 of the plurality of patterns with the positions on the sensor, outputs them as mask design values ​​to the mask generation unit 156', and records them in the recording unit 157'.

[0148] In step S55, the mask generation unit 156' determines the size of the mask 152 based on the mask design values ​​of the multiple patterns supplied by the mask design unit 155, and generates multiple masks 152 of the multiple patterns by drilling holes for installing bandpass filters and attaching the bandpass filters to the corresponding positions.

[0149] In step S56, the mask generation unit 156' sets one of the masks 152 of the unprocessed patterns among the plurality of masks 152 of the plurality of patterns as the mask to be processed.

[0150] In step S57, the mask generation unit 156′ places the mask 152 to be processed between the illumination 151 and the spectroscopic measurement device 153, and causes light to be emitted from the illumination 151. At this time, the spectroscopic measurement device 153 captures a spectroscopic image projected by the light emitted from the illumination 151 passing through the mask 152, and outputs the captured image result to the signal processing unit 158′.

[0151] In step S58, the signal processing unit 158′ acquires the mask design values ​​of the mask to be processed that are recorded in the recording unit 157′, separates them into PSFs for each wavelength by signal processing based on the mask design values ​​and the spectroscopic imaging results captured by the spectroscopic measurement device 153, and outputs the PSFs to the PSF interpolation unit 202.

[0152] In step S59, the mask generation unit 156' determines whether or not there is an unprocessed mask 152 of a pattern among the masks 152 of a plurality of patterns.

[0153] In step S59, if there is an unprocessed pattern mask 152, the process returns to step S56.

[0154] That is, the processes of steps S56 to S59 are repeated until there are no more unprocessed pattern masks. For example, when the masks 152#1 to 152#4 described with reference to Fig. 13 have been generated, there will be four pattern masks 152, and therefore the processes of steps S56 to S59 are repeated four times, and spectroscopic imaging is performed using the masks 152 of each pattern, and the PSFs of each pattern are calculated.

[0155] In step S59, if a PSF is generated in which spectroscopic imaging is performed using the masks 152 of all patterns, and it is determined that there are no unprocessed masks 152 of patterns, the process proceeds to step S60.

[0156] In step S60, the PSF interpolation unit 202 generates a PSF for each wavelength and for each spatial position by interpolation using the PSF obtained from the spectroscopic imaging results using the masks 152 of all patterns.

[0157] By the above processing, even if the lens of the spectroscopic measurement device 153 (a lens corresponding to the imaging lens 35 in FIG. 1) has a short focal length or has aberrations, the PSF can be obtained with a small number of spectroscopic imaging operations and simple signal processing.

[0158] As a result, even if the lens of the spectroscopic measurement device 153 (a lens corresponding to the imaging lens 35 in Figure 1) has a short focal length or has aberrations, calibration can be performed easily and with high accuracy.

[0159] <Mask design processing when the lens of the optical measurement device is short focal length or has aberration> Next, a mask design processing when the lens of the optical measurement device is short focal length or has aberration will be described with reference to the flowchart of Fig. 17. Note that the processing of steps S72 and S74 to S79 ​​in the flowchart of Fig. 17 is the same as the processing of steps S31 and S33 to S38 in Fig. 12, and therefore description thereof will be omitted.

[0160] That is, in the second embodiment, as a result of dividing the area of ​​the mask 152, the process in step S73 differs from the process in step S32.

[0161] That is, in step S73, the mask design unit 155′ randomly sets the bandpass filter position (xmn, ymn, λn) for the nth wavelength λn from within the divided area of ​​the available area on the mask 152 to which the bandpass filter for the nth wavelength λn is assigned.

[0162] Furthermore, since the region division results in the generation of a plurality of masks 152 consisting of a plurality of patterns generated by rearranging the divided regions, one of the unprocessed patterns is selected in step S71, and a mask 152 of the selected pattern is generated in steps S72 to S79. Then, once a mask 152 is generated by the processes of steps S72 to S79, it is determined in step S80 whether or not there is a mask 152 of an unprocessed pattern that has not yet been generated, and if there is an unprocessed pattern, the process returns to step S71.

[0163] That is, the processes of steps S71 to S80 are repeated until there are no more masks with unprocessed patterns, and the design of a plurality of masks 152 with different patterns is repeated.

[0164] Then, in step S80, when it is determined that masks 152 for all patterns have been designed and there are no unprocessed masks 152 for patterns, the process ends.

[0165] By the above processing, a set number of bandpass filters are placed on a plurality of masks 152 each consisting of a pattern according to the number of divisions into which the area of ​​the mask 152 is divided, and an arrangement of holes is designed to provide bandpass filters through which the PSF is projected without overlapping on the image sensor 36 of the spectroscopic measurement device 153 after the incident light has passed through.

[0166] As a result, even if the lens of the spectroscopic measurement device 153 (a lens corresponding to the imaging lens 35 in Figure 1) has a short focal length or has aberrations, it is possible to obtain a PSF for each wavelength and for each position by simply performing spectroscopic imaging in the spectroscopic measurement device 153 a number of times corresponding to the number of divisions of the mask 152.

[0167] As a result, even if the lens of the spectroscopic measurement device 153 (a lens corresponding to the imaging lens 35 in Figure 1) has a short focal length or has aberrations, it is possible to achieve simple and highly accurate calibration of the spectroscopic measurement device 153.

[0168] <<4. Example of Execution by Software>> The above-described series of processes can be executed by hardware, but can also be executed by software. When the series of processes is executed by software, the program that constitutes the software is installed from a recording medium into a computer that is built into dedicated hardware, or into, for example, a general-purpose computer that can execute various functions by installing various programs.

[0169] 18 shows an example of the configuration of a general-purpose computer. This computer has a built-in CPU (Central Processing Unit) 1001. An input / output interface 1005 is connected to the CPU 1001 via a bus 1004. A ROM (Read Only Memory) 1002 and a RAM (Random Access Memory) 1003 are connected to the bus 1004.

[0170] The input / output interface 1005 is connected to an input unit 1006 including input devices such as a keyboard and a mouse through which a user inputs operation commands, an output unit 1007 that outputs a processing operation screen and images of processing results to a display device, a storage unit 1008 including a hard disk drive or the like that stores programs and various data, and a communication unit 1009 including a LAN (Local Area Network) adapter or the like that executes communication processing via a network typified by the Internet. Also connected is a drive 1010 that reads and writes data from / to a removable storage medium 1011 such as a magnetic disk (including a flexible disk), an optical disk (including a CD-ROM (Compact Disc-Read Only Memory) and a DVD (Digital Versatile Disc)), a magneto-optical disk (including an MD (Mini Disc)), or a semiconductor memory.

[0171] The CPU 1001 executes various processes in accordance with a program stored in a ROM 1002 or a program read from a removable storage medium 1011 such as a magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, installed in a storage unit 1008, and loaded from the storage unit 1008 into a RAM 1003. The RAM 1003 also stores data necessary for the CPU 1001 to execute various processes as appropriate.

[0172] In a computer configured as described above, the CPU 1001 performs the above-described series of processes by, for example, loading a program stored in the memory unit 1008 into the RAM 1003 via the input / output interface 1005 and the bus 1004 and executing it.

[0173] The program executed by the computer (CPU 1001) can be provided by being recorded on a removable storage medium 1011 such as a package medium, for example. The program can also be provided via a wired or wireless transmission medium such as a local area network, the Internet, or digital satellite broadcasting.

[0174] In a computer, a program can be installed in the storage unit 1008 via the input / output interface 1005 by inserting a removable storage medium 1011 into the drive 1010. The program can also be received by the communication unit 1009 via a wired or wireless transmission medium and installed in the storage unit 1008. Alternatively, the program can be installed in advance in the ROM 1002 or the storage unit 1008.

[0175] The program executed by the computer may be a program that processes in chronological order according to the order described in this specification, or may be a program that processes in parallel or at the required timing, such as when called.

[0176] 18 realizes the functions of the bandpass filter setting unit 154, the mask design unit 155, and the signal processing unit 158 ​​in FIG. 8, and the functions of the bandpass filter setting unit 154, the mask region dividing unit 201, the mask design unit 155′, the signal processing unit 158′, and the PSF interpolation unit 202 in FIG. 15.

[0177] In this specification, a system refers to a collection of multiple components (devices, modules (components), etc.), regardless of whether all of the components are contained in the same housing. Therefore, multiple devices housed in separate housings and connected via a network, and a single device with multiple modules housed in a single housing, are both systems.

[0178] Furthermore, the embodiments of the present disclosure are not limited to the above-described embodiments, and various modifications are possible within the scope of the gist of the present disclosure.

[0179] For example, the present disclosure can be configured as a cloud computing system in which a single function is shared and processed collaboratively by multiple devices via a network.

[0180] Furthermore, each step described in the above flowchart can be executed by one device, or can be shared and executed by a plurality of devices.

[0181] Furthermore, when one step includes multiple processes, the multiple processes included in that one step can be executed by one device or can be shared and executed by multiple devices.

[0182] The present disclosure can also be configured as follows.

[0183] <1> A calibration system for calibrating a spectroscopic measurement device, comprising: a mask in which bandpass filters that transmit light of a plurality of different wavelength bands from a light source are arranged; a spectroscopic measurement device that performs spectroscopic measurement of the light from the light source that has passed through the mask; and a signal processing unit that calculates a PSF for each wavelength used for the calibration by signal processing based on the spectroscopic measurement result of the spectroscopic measurement device, wherein the bandpass filters that transmit light of a plurality of different wavelength bands and are arranged on the mask are arranged such that PSFs for the different wavelengths do not overlap in the spectroscopic measurement result. <2> The spectroscopic measurement device comprises: a diffractive optical element that disperses the light that has passed through the mask; an imaging lens that forms an image of the light dispersed by the diffractive optical element; and an image sensor that captures the dispersed light imaged by the imaging lens as a spectroscopic image that is the spectroscopic measurement result, wherein the light that has passed through the bandpass filters of different wavelength bands that are arranged on the mask and dispersed is imaged on the image sensor in a state where it does not overlap. <3> The arrangement of the bandpass filters on the mask is determined by: randomly determining the arrangement of bandpass filters of a first wavelength band in usable areas on the mask where bandpass filters of other wavelength bands are not arranged, then specifying a projection position on the image sensor onto which light that has passed through and been dispersed by the bandpass filter of the first wavelength band is projected, the projection position being specified by a calculation based on the positional relationships between the mask, the diffractive optical element, the imaging lens, and the image sensor; specifying a usable area on the mask from usable areas on the image sensor that are areas on the image sensor that have not been specified as projection positions of light that has passed through and been dispersed by the bandpass filters of the other wavelength bands, where a bandpass filter of a second wavelength band that is specified by an inverse calculation of the calculation can be installed; and updating the usable area on the mask to the specified area, repeating this process for the required number of bandpass filters.<4> The calibration system described in <3>, further including a recording unit that, when the arrangement of the bandpass filter on the mask is determined, records, as mask design values, the position on the mask of the bandpass filter whose arrangement has been determined, the wavelength band, and the projection position on the image sensor onto which light transmitted through the bandpass filter whose arrangement has been determined and dispersed is projected, wherein the signal processing unit calculates a PSF for each wavelength used for the calibration based on the mask design value recorded in the recording unit and the spectral image. <5> The calibration system described in <4>, wherein the signal processing unit searches for a projection position on the image sensor to be registered as the mask design value based on the projection position in the spectral image, and identifies the wavelength of the PSF based on information on the wavelength band registered in association with the closest one. <6> The calibration system described in <2>, when the imaging lens has a long focal length and is free of aberration, the signal processing unit calculates a PSF for each wavelength used for the calibration based on one of the spectral images. <7> The calibration system according to <2>, wherein the signal processing unit calculates a PSF for each wavelength used for the calibration based on a plurality of spectroscopic images obtained by spectroscopic imaging using a plurality of masks when the imaging lens has a single focus or when there is an aberration, and further includes an interpolation unit that calculates a PSF for each wavelength and for each spatial position by interpolation using the PSF for each wavelength obtained from the plurality of spectroscopic images. <8> The calibration system according to <7>, further includes an area dividing unit that equally divides an area of ​​the mask into a plurality of divided areas and assigns bandpass filters that transmit light of the plurality of different wavelength bands to each of the plurality of divided areas in units of bandwidth, and the plurality of masks are masks in which the arrangement of the plurality of divided areas is different.<9> The calibration system described in <8>, wherein the PSFs for each wavelength obtained from the plurality of spectral images corresponding to the plurality of masks are PSFs with different spatial positions for each wavelength according to the arrangement of the divided areas in the plurality of masks, and the interpolation unit calculates the PSFs for each wavelength and for each spatial position by interpolation using the PSFs with different spatial positions for each wavelength obtained from the plurality of spectral images corresponding to the plurality of masks. <10> The arrangement of the band-pass filters on the mask is determined by: randomly determining the arrangement of the band-pass filters of the first wavelength band in the divided regions on the mask that are usable regions on the mask where band-pass filters of other wavelength bands are not arranged and to which band-pass filters of a first wavelength band are assigned; then specifying a projection position on the image sensor onto which light that has passed through the band-pass filter of the first wavelength band and been dispersed is projected, the projection position being specified by a calculation based on positional relationships between the mask, the diffractive optical element, the imaging lens, and the image sensor; specifying a usable region on the mask from an image sensor usable region that is a region on the image sensor that has not been specified as a projection position of light that has passed through the band-pass filter of the other wavelength band and been dispersed, where a band-pass filter of a second wavelength band that is specified by an inverse calculation of the calculation can be installed; and updating the usable region on the mask to the specified region, repeating this process for the required number of band-pass filters. <11> The calibration system according to <10>, further including a recording unit that, when the arrangement of the band-pass filters on the mask is determined, records, as mask design values ​​for each of the plurality of masks, the positions on the mask of the band-pass filters whose arrangement has been determined, the wavelength bands, and the projection positions on the image sensor onto which light that has passed through the band-pass filters whose arrangement has been determined and dispersed is projected, and the signal processing unit calculates, for each of the plurality of masks, a PSF for each wavelength used for the calibration based on the mask design values ​​recorded in the recording unit and the spectral images.<12> The calibration system described in <11>, wherein the signal processing unit searches for a projection position on the image sensor that is registered as a mask design value based on the projection position in the spectral image for each of the plurality of masks, and identifies the wavelength of the PSF based on information on a wavelength band that is registered in correspondence with the closest one. <13> The calibration system described in <1>, wherein the spectroscopic measurement device is a snapshot-type spectroscopic measurement device. <14> A calibration method for a calibration system that achieves calibration of a spectroscopic measurement device including: a mask in which bandpass filters that transmit light of a plurality of different wavelength bands from a light source are arranged; and a spectroscopic measurement device that performs spectroscopic measurement of the light from the light source that has passed through the mask, the calibration method including performing signal processing based on spectroscopic measurement results of the spectroscopic measurement device to calculate a PSF for each wavelength used in the calibration, and the bandpass filters that transmit light of a plurality of different wavelength bands that are arranged on the mask are arranged so that PSFs for different wavelengths do not overlap in the spectroscopic measurement results.

[0184] REFERENCE SIGNS LIST 11 Spectroscopic measurement device, 31 Objective lens, 32 Slit (field stop), 33 Collimator lens, 34 Diffractive optical element (DoE), 35 Imaging lens, 36 Image sensor, 131, 131' Calibration system, 151 Illumination, 152 Mask, 153 Spectroscopic measurement device, 154 Bandpass filter setting unit, 155, 155' Mask design unit, 156, 156' Mask generation unit, 157, 157' Recording unit, 158, 158' Signal processing unit, 201 Mask area division unit, 202 PSF interpolation unit

Claims

1. A calibration system for realizing calibration of a spectroscopic measurement device, comprising: a mask in which bandpass filters that transmit light of a plurality of different wavelength bands from light emitted by a light source are arranged; a spectroscopic measurement device that performs spectroscopic measurement of the light from the light source that has passed through the mask; and a signal processing unit that calculates a PSF for each wavelength used for the calibration by signal processing based on the spectroscopic measurement results of the spectroscopic measurement device, wherein the bandpass filters that transmit light of a plurality of different wavelength bands and are arranged on the mask are arranged so that the PSFs for each different wavelength do not overlap in the spectroscopic measurement results.

2. The calibration system according to claim 1, wherein the spectroscopic measurement device comprises: a diffractive optical element that disperses light that has passed through the mask; an imaging lens that forms an image of the light dispersed by the diffractive optical element; and an image sensor that captures the dispersed light imaged by the imaging lens as a spectroscopic image that is the spectroscopic measurement result; and the dispersed light that passes through each of the bandpass filters of different wavelength bands arranged on the mask is imaged on the image sensor without overlapping.

3. The calibration system according to claim 2, wherein the arrangement of the band-pass filters on the mask is determined by: randomly determining the arrangement of band-pass filters of a first wavelength band in usable areas on the mask where band-pass filters of other wavelength bands are not arranged; then specifying a projection position on the image sensor onto which light that has passed through and been dispersed by the band-pass filter of the first wavelength band is projected, the projection position being specified by a calculation based on the positional relationships between the mask, the diffractive optical element, the imaging lens, and the image sensor; specifying a usable area on the mask where a band-pass filter of a second wavelength band, specified by an inverse calculation of the calculation, can be installed from usable areas on the image sensor that are areas on the image sensor that have not been specified as projection positions of light that has passed through and been dispersed by the band-pass filters of the other wavelength bands; and updating the usable area on the mask to the specified area, repeating this process for the required number of band-pass filters.

4. The calibration system according to claim 3, further comprising a recording unit that, when the arrangement of the band-pass filter on the mask is determined, records, as mask design values, the position on the mask of the band-pass filter whose arrangement has been determined, the wavelength band, and the projection position on the image sensor onto which light that has passed through the band-pass filter whose arrangement has been determined and dispersed is projected, and the signal processing unit calculates a PSF for each wavelength used in the calibration based on the mask design values ​​recorded in the recording unit and the spectral image.

5. The calibration system according to claim 4, wherein the signal processing unit searches for a projection position on the image sensor that is registered as the mask design value based on the projection position in the spectral image, and identifies the wavelength of the PSF based on information on the wavelength band that is registered in association with the closest one.

6. The calibration system according to claim 2, wherein when the imaging lens has a long focal length and is free of aberration, the signal processing unit calculates a PSF for each wavelength used for the calibration based on one of the spectral images.

7. The calibration system according to claim 2, wherein, when the imaging lens has a single focus or has aberration, the signal processing unit calculates a PSF for each wavelength used in the calibration based on a plurality of the spectroscopic images obtained by spectroscopic imaging using a plurality of masks, and further includes an interpolation unit that calculates a PSF for each wavelength and for each spatial position by interpolation using the PSFs for each wavelength obtained from the plurality of spectroscopic images.

8. The calibration system according to claim 7, further comprising an area dividing section that divides the area of ​​the mask evenly into a plurality of divided areas and assigns bandpass filters that transmit light of the plurality of different wavelength bands to each of the plurality of divided areas in units of bandwidth, and the plurality of masks are masks in which the arrangement of the plurality of divided areas differs.

9. The calibration system according to claim 8, wherein the PSFs for each wavelength obtained from the plurality of spectral images corresponding to the plurality of masks are PSFs with different spatial positions for each wavelength according to the arrangement of the divided areas in the plurality of masks, and the interpolation unit calculates the PSFs for each wavelength and for each spatial position by interpolation using the PSFs with different spatial positions for each wavelength obtained from the plurality of spectral images corresponding to the plurality of masks.

10. The calibration system according to claim 8, wherein the arrangement of the band-pass filters on the mask is determined by: randomly determining the arrangement of the band-pass filters of the first wavelength band in the divided regions on the mask that are usable regions on the mask where band-pass filters of other wavelength bands are not arranged and to which band-pass filters of a first wavelength band are assigned; then specifying a projection position on the image sensor onto which light that has passed through the band-pass filter of the first wavelength band and been dispersed is projected, the projection position being specified by a calculation based on the positional relationships between the mask, the diffractive optical element, the imaging lens, and the image sensor; specifying a usable region on the mask where a band-pass filter of a second wavelength band that is specified by an inverse calculation of the calculation can be installed from an image sensor usable region that is a region on the image sensor that has not been specified as a projection position of light that has passed through the band-pass filter of the other wavelength band and been dispersed; and updating the usable region on the mask to the specified region, repeating this process for the required number of band-pass filters.

11. The calibration system according to claim 10, further comprising a recording unit that, when the placement of the bandpass filters on the mask is determined, records, as mask design values ​​for each of the plurality of masks, the positions on the mask of the bandpass filters whose placement has been determined, the wavelength bands, and the projection positions on the image sensor onto which light that has passed through the bandpass filters whose placement has been determined and dispersed is projected, and the signal processing unit calculates, for each of the plurality of masks, a PSF for each wavelength used in the calibration based on the mask design values ​​recorded in the recording unit and the spectral image.

12. The calibration system according to claim 11, wherein the signal processing unit searches for a projection position on the image sensor that is registered as a mask design value for each of the plurality of masks based on the projection position in the spectral image, and identifies the wavelength of the PSF based on information about the wavelength band that is registered in association with the closest one.

13. The calibration system according to claim 1, wherein the spectroscopic measurement device is a snapshot-type spectroscopic measurement device.

14. A calibration method for a calibration system that realizes calibration of a spectroscopic measurement device that includes: a mask in which bandpass filters that transmit light in a plurality of different wavelength bands from light emitted by a light source are arranged; and a spectroscopic measurement device that performs spectroscopic measurement of the light from the light source that has passed through the mask, the method including performing signal processing based on the spectroscopic measurement results of the spectroscopic measurement device to calculate a PSF for each wavelength used in the calibration, and the bandpass filters that transmit light in a plurality of different wavelength bands and are arranged on the mask so that the PSFs for each different wavelength do not overlap in the spectroscopic measurement results.

Images