Spectrometry device, and spectrometry method

Abstract

An analyzing unit of this spectrometry device uses a first correction factor to correct a first electrical signal such that a linearity characteristic of a first amplifier matches a reference linearity characteristic. The analyzing unit uses a second correction factor to correct a second electrical signal such that a linearity characteristic of a second amplifier matches the reference linearity characteristic. The analyzing unit generates first spectrum data on the basis of the corrected first electrical signal, and generates second spectrum data on the basis of the corrected second electrical signal. The analyzing unit generates spectrum data of light to be measured, on the basis of the first spectrum data and the second spectrum data.

Description

Spectroscopic measurement device and spectroscopic measurement method

[0001] The present disclosure relates to a spectroscopic measurement device and a spectroscopic measurement method.

[0002] A spectroscopic measurement device is known that includes an optical system that disperses light under measurement, a photodetector that receives spectral images of the dispersed light under measurement in a first region with a short exposure time and a second region with a long exposure time, and an analyzer that generates spectral data of the light under measurement based on first spectral data generated based on the detection result in the first region and second spectral data generated based on the detection result in the second region (see, for example, Patent Document 1). Such a spectroscopic measurement device can generate spectral data of the light under measurement over a wide dynamic range.

[0003] Japanese Patent Application Laid-Open No. 2020-118477

[0004] In the above-described spectroscopic measurement device, errors may occur between the first spectral data and the second spectral data due to factors other than the difference in exposure time, and if spectral data of the measured light is generated based on the first spectral data and the second spectral data in a state where such errors occur, the accuracy of the spectral data of the measured light may be reduced.

[0005] An object of the present disclosure is to provide a spectroscopic measurement device and a spectroscopic measurement method that can acquire highly accurate spectral data of light to be measured.

[0006] A spectroscopic measurement device according to one aspect of the present disclosure includes: [1] "an optical system that disperses light under measurement; a photodetector that detects a spectral image of the light under measurement dispersed by the optical system; and an analyzer that generates spectral data of the light under measurement, wherein the photodetector has a plurality of first pixel rows arranged in a horizontal direction parallel to a wavelength axis of the spectral image, and each of the plurality of first pixel rows has a first light receiving unit including a plurality of first pixels arranged in a vertical direction perpendicular to the wavelength axis, and a plurality of second pixel rows that are juxtaposed with the first light receiving unit in the vertical direction and arranged in the horizontal direction, and a second light receiving section, each of two pixel columns including a plurality of second pixels arranged in the vertical direction; a first vertical transfer section that transfers, in the vertical direction for each of the plurality of first pixel columns, first charges accumulated in each of the plurality of first pixels by receiving the spectral image in the first light receiving section for a first exposure time; a first horizontal transfer section that transfers, in the horizontal direction, the first charges transferred by the first vertical transfer section; a first amplifier that outputs a first electric signal according to the amount of the first charges transferred by the first horizontal transfer section; a second vertical transfer unit that transfers, in the vertical direction for each of the plurality of second pixel columns, second charges accumulated in each of the plurality of second pixels by receiving the spectral image for a second exposure time longer than the first exposure time; a second horizontal transfer unit that transfers, in the horizontal direction, the second charges transferred by the second vertical transfer unit; and a second amplifier that outputs a second electric signal corresponding to an amount of the second charges transferred by the second horizontal transfer unit, wherein the analyzing unit stores a first correction coefficient for correcting the first electric signal so that a linearity characteristic of the first amplifier matches a reference linearity characteristic, stores a second correction coefficient for correcting the second electric signal so that the linearity characteristic of the second amplifier matches the reference linearity characteristic, corrects the first electric signal using the first correction coefficient, corrects the second electric signal using the second correction coefficient, generates first spectral data based on the corrected first electric signal, generates second spectral data based on the corrected second electric signal, and generates the spectral data of the light under measurement based on the first spectral data and the second spectral data."Spectroscopic measurement device".

[0007] In the spectroscopic measurement device described in [1] above, the photodetector receives a spectral image at the first light receiving unit over a first exposure time and outputs a first electrical signal, and receives a spectral image at the second light receiving unit over a second exposure time longer than the first exposure time and outputs a second electrical signal. The analyzer generates first spectral data based on the first electrical signal and second spectral data based on the second electrical signal. The analyzer then generates spectral data of the measured light based on the first spectral data and the second spectral data. This allows for the generation of spectral data of the measured light over a wide dynamic range. Furthermore, in the spectroscopic measurement device described in [1] above, the analyzer corrects the first electrical signal using a first correction coefficient and corrects the second electrical signal using a second correction coefficient. This allows for the correction of errors occurring in the linearity characteristics of the first amplifier and the second amplifier. Therefore, the spectroscopic measurement device described in [1] above allows for the acquisition of highly accurate spectral data of the measured light.

[0008] A spectroscopic measurement device according to one aspect of the present disclosure may be the spectroscopic measurement device described in [1] above, [2] in which "the analyzer generates the first spectral data based on a first correspondence relationship between each of the plurality of first pixel rows and a wavelength on the wavelength axis, and generates the second spectral data based on a second correspondence relationship between each of the plurality of second pixel rows and a wavelength on the wavelength axis." In the spectroscopic measurement device, a misalignment may occur between a position of a first pixel row on which an optical image of a certain wavelength is formed in a spectral image and a position of a second pixel row on which the optical image is formed. The spectroscopic measurement device described in [2] can acquire first spectral data and second spectral data in which the misalignment has been corrected, even when such a misalignment occurs.

[0009] A spectroscopic measurement device according to one aspect of the present disclosure may be [3] "the spectroscopic measurement device according to the above [1] or [2], wherein the analyzer corrects at least one of the first spectral data and the second spectral data based on a ratio between a representative value of the intensity of the first spectral data and a representative value of the intensity of the second spectral data." In the spectroscopic measurement device, a difference may occur between the light intensity in a first pixel row on which an optical image of a certain wavelength in a spectral image is formed and the light intensity in a second pixel row on which the optical image is formed. The spectroscopic measurement device according to the above [3] can acquire first spectral data and second spectral data in which the difference in light intensity has been corrected, even when such a difference in light intensity occurs.

[0010] A spectroscopic measurement device according to one aspect of the present disclosure may be [4] "the spectroscopic measurement device according to any one of [1] to [3] above, in which the analyzer corrects the second spectral data to remove stray light generated in the optical system." Focusing only on noise generated in the circuit, the S / N ratio of the second spectral data is higher than the S / N ratio of the first spectral data. However, the measured light may contain stray light. In such a case, the second exposure time is longer than the first exposure time, increasing noise due to the stray light, and the S / N ratio of the second spectral data is reduced. According to the spectroscopic measurement device according to [4], even when stray light is contained in the measured light, it is possible to obtain second spectral data from which such stray light has been removed.

[0011] The spectroscopic measurement device according to one aspect of the present disclosure may be [5] "the spectroscopic measurement device according to any one of [1] to [4] above, further including a bundle fiber including a plurality of optical fibers that guide the light under measurement to the optical system, the bundle fiber having a circular incident end face and a long exit end face, the plurality of optical fibers including a plurality of first optical fibers arranged on one side of the center of the exit end face at the exit end face and a plurality of second optical fibers arranged on the other side of the center of the exit end face at the exit end face, the plurality of first optical fibers and the plurality of second optical fibers being adjacent to each other in at least one of the circumferential direction and the radial direction at the entrance end face." In the spectroscopic measurement device according to [5], even if a bias in the light intensity of the light under measurement occurs at the entrance end face of the bundle fiber, the bias in the light intensity of the light under measurement is suppressed at the exit end face of the bundle fiber. Therefore, according to the spectroscopic measurement device described in [5], when focusing on the first pixel row and the second pixel row on which a light image of a certain wavelength is formed in the spectral image, it is possible to acquire the first spectral data and the second spectral data while suppressing the occurrence of a difference between the light intensity in the first pixel row and the light intensity in the second pixel row.

[0012] A spectroscopic measurement device according to one aspect of the present disclosure may be [6] "the spectroscopic measurement device according to any one of [1] to [5] above, further including: a bundle fiber including a plurality of optical fibers that guide the light under measurement to the optical system; and a light diffusion unit that is arranged upstream of the bundle fiber and that guides the light under measurement to an incident end face of the bundle fiber in a diffused state." In the spectroscopic measurement device according to [6], the light under measurement diffused by the light diffusion unit is incident on the incident end face of the bundle fiber, thereby suppressing bias in the light intensity of the light under measurement at the incident end face of the bundle fiber. Therefore, according to the spectroscopic measurement device according to [6], when focusing on a first pixel row and a second pixel row on which a light image of a certain wavelength is formed in a spectral image, it is possible to acquire first spectral data and second spectral data while suppressing a difference in light intensity between the first pixel row and the second pixel row.

[0013] According to one aspect of the present disclosure, the spectroscopic measurement device may be the spectroscopic measurement device described in any one of [1] to [6] above, wherein [7] "the photodetector has an electronic shutter function that, during one frame time from the start of transfer of the first charges by the first vertical transfer unit to the end of transfer of the first charges by the first horizontal transfer unit, discharges the first charges accumulated in each of the plurality of first pixels for a predetermined time from the start of the one frame time, thereby setting the time of the one frame time excluding the predetermined time as the first exposure time; and the analysis unit, when the electronic shutter function is used, stores a third correction coefficient associated with the predetermined time for correcting the first electrical signal so that the linearity characteristics of the first amplifier match the reference linearity characteristics, and corrects the first electrical signal using the third correction coefficient." When the electronic shutter function is used for the first light receiving unit, the intensity of the first electrical signal decreases as the length of the predetermined time increases, but the intensity of the first electrical signal does not decrease linearly, strictly speaking. According to the spectroscopic measurement device described in [7], by correcting the first electrical signal using a third correction coefficient associated with a predetermined time, it is possible to obtain first spectral data in which errors occurring in the linearity characteristics of the first amplifier have been corrected.

[0014] According to one aspect of the present disclosure, the spectroscopic measurement device may be the spectroscopic measurement device described in [2] above, wherein the photodetector has an electronic shutter function for discharging the first charges accumulated in each of the plurality of first pixels for a predetermined time from the start of one frame time until the end of the transfer of the first charges by the first vertical transfer unit, the first charges being accumulated in the plurality of first pixels, during one frame time from the start of the one frame time until the end of the transfer of the first charges by the first horizontal transfer unit. The analyzer generates the first spectral data based on the first correspondence relationship corresponding to at least one of the predetermined time and the first exposure time. When the electronic shutter function is used for the first light receiving unit, a deviation may occur in the correspondence relationship between each of the plurality of first pixel rows and the wavelength on the wavelength axis. According to the spectroscopic measurement device described in [8] above, the first spectral data can be obtained by correcting the deviation in the correspondence relationship by generating the first spectral data based on the first correspondence relationship corresponding to at least one of the predetermined time and the first exposure time.

[0015] The spectroscopic measurement device according to one aspect of the present disclosure may be the spectroscopic measurement device according to any one of [1] to [6] above, wherein the photodetector has an electronic shutter function that, during one frame time from the start of transfer of the first charges by the first vertical transfer unit to the end of transfer of the first charges by the first horizontal transfer unit, discharges the first charges accumulated in each of the plurality of first pixels for a predetermined time from the start of the one frame time, thereby setting the time of the one frame time excluding the predetermined time as the first exposure time. According to the spectroscopic measurement device described in [9], by using the electronic shutter function for the first light receiving unit, it is possible to prevent each first pixel in the first light receiving unit from saturating, and it is possible to reliably acquire first spectral data in a saturation wavelength band.

[0016] A spectroscopic measurement device according to one aspect of the present disclosure may be

[10] "the spectroscopic measurement device according to any one of [1] to [9] above, wherein, when the first light receiving unit and the second light receiving unit are exposed for the same exposure time, the analyzer generates a first function of the first spectral data and a second function of the second spectral data, and corrects at least one of the first spectral data and the second spectral data based on a ratio between the first function and the second function." In the spectroscopic measurement device, the light sensitivity of the first light receiving unit and the light sensitivity of the second light receiving unit do not necessarily match. According to the spectroscopic measurement device described in

[10] , even when a difference in light sensitivity occurs, it is possible to acquire first spectral data and second spectral data in which such a difference in light sensitivity has been corrected.

[0017] The spectroscopic measurement device according to one aspect of the present disclosure may be

[11] "the spectroscopic measurement device according to any one of [1] to

[10] above, wherein the analysis unit corrects at least one of the first spectral data and the second spectral data based on a ratio between the first exposure time and the second exposure time." The spectroscopic measurement device according to

[11] can acquire first spectral data and second spectral data in which the influence of a difference in exposure time on light intensity has been corrected.

[0018] A spectroscopic measurement device according to one aspect of the present disclosure includes,

[12] "an optical system that spectrally separates light under measurement; a photodetector that detects a spectral image of the light under measurement separated by the optical system; and a storage unit, wherein the photodetector has a plurality of first pixel rows arranged in a horizontal direction parallel to a wavelength axis of the spectral image, each of the plurality of first pixel rows including a plurality of first pixels arranged in a vertical direction perpendicular to the wavelength axis; a second photodetector that is juxtaposed with the first photodetector in the vertical direction and has a plurality of second pixel rows arranged in the horizontal direction, each of the plurality of second pixel rows including a plurality of second pixels arranged in the vertical direction; a first vertical transfer unit that transfers, in the vertical direction for each of the plurality of first pixel rows, first charges accumulated in each of the plurality of first pixels by receiving the spectral image in the first exposure time; and a first horizontal transfer unit that transfers the first charges transferred by the first vertical transfer unit in the horizontal direction. a first amplifier that outputs a first electrical signal corresponding to the amount of the first charge transferred by the first horizontal transfer unit; a second vertical transfer unit that transfers, in the vertical direction for each of the plurality of second pixel columns, second charges accumulated in each of the plurality of second pixels by receiving the spectral image in the second light receiving unit for a second exposure time longer than the first exposure time; a second horizontal transfer unit that transfers the second charges transferred by the second vertical transfer unit in the horizontal direction; and a second amplifier that outputs a second electrical signal corresponding to the amount of the second charge transferred by the second horizontal transfer unit, wherein the memory unit stores a first correction coefficient for correcting the first electrical signal so that the linearity characteristic of the first amplifier matches a reference linearity characteristic, and stores a second correction coefficient for correcting the second electrical signal so that the linearity characteristic of the second amplifier matches the reference linearity characteristic.

[0019] In the spectroscopic measurement device described in

[12] above, the photodetector receives a spectral image at the first light receiving unit over a first exposure time and outputs a first electrical signal, and receives a spectral image at the second light receiving unit over a second exposure time longer than the first exposure time and outputs a second electrical signal. The memory stores a first correction coefficient for correcting the first electrical signal and a second correction coefficient for correcting the second electrical signal. Therefore, when generating spectral data of the light under measurement based on the first electrical signal and the second electrical signal, errors occurring in the linearity characteristics of the first amplifier and the second amplifier can be corrected by correcting the first electrical signal using the first correction coefficient and the second electrical signal using the second correction coefficient. Therefore, the spectroscopic measurement device described in

[12] above can acquire spectral data of the light under measurement with high accuracy.

[0020] A spectroscopic measurement method according to one aspect of the present disclosure is

[13] "a spectroscopic measurement method using a spectroscopic measurement device, the spectroscopic measurement device including: an optical system that spectrally separates light to be measured; a photodetector that detects a spectral image of the light to be measured separated by the optical system; and an analyzer that generates spectral data of the light to be measured, the photodetector having a plurality of first pixel rows arranged in a horizontal direction parallel to a wavelength axis of the spectral image, each of the plurality of first pixel rows being a first light receiving unit including a plurality of first pixels arranged in a vertical direction perpendicular to the wavelength axis; a second light receiving section arranged in parallel with the first vertical transfer section and having a plurality of second pixel columns arranged in the horizontal direction, each of the plurality of second pixel columns including a plurality of second pixels arranged in the vertical direction; a first vertical transfer section that transfers, in the vertical direction for each of the plurality of first pixel columns, first charges accumulated in each of the plurality of first pixels by receiving the spectral image in the first light receiving section for a first exposure time; a first horizontal transfer section that transfers, in the horizontal direction, the first charges transferred by the first vertical transfer section; a first amplifier configured to output a first electrical signal corresponding to the amount of second charge accumulated in each of the plurality of second pixels by receiving the spectral image in the second light receiving unit for a second exposure time longer than the first exposure time, a second vertical transfer unit configured to transfer, in the vertical direction for each of the plurality of second pixel columns, second charges accumulated in each of the plurality of second pixels by receiving the spectral image in the second light receiving unit for a second exposure time longer than the first exposure time, a second horizontal transfer unit configured to transfer the second charges transferred by the second vertical transfer unit in the horizontal direction, and a second amplifier configured to output a second electrical signal corresponding to the amount of second charge transferred by the second horizontal transfer unit, wherein the spectroscopic measurement method includes the steps of: correcting the first electrical signal so that a linearity characteristic of the first amplifier matches a reference linearity characteristic; correcting the second electrical signal so that a linearity characteristic of the second amplifier matches the reference linearity characteristic; generating first spectral data based on the corrected first electrical signal; generating second spectral data based on the corrected second electrical signal; and generating the spectral data of the light under measurement based on the first spectral data and the second spectral data.

[0021] According to the spectroscopic measurement method described in

[13] above, for the same reason as the spectroscopic measurement device described in [1] above, it is possible to obtain highly accurate spectral data of the light to be measured.

[0022] According to the present disclosure, it is possible to provide a spectroscopic measurement device and a spectroscopic measurement method that can acquire highly accurate spectral data of light to be measured.

[0023] FIG. 1 is a diagram illustrating a configuration of a spectroscopic measurement device according to an embodiment. FIG. 2 is a diagram illustrating the configurations of an incident end face and an exit end face of a fiber bundle illustrated in FIG. 1 . FIG. 3 is a diagram illustrating the configuration of a photodetector illustrated in FIG. 1 . FIG. 4 is a diagram illustrating the timing of exposure of each first pixel. FIG. 5 is a diagram illustrating the timing of exposure of each first pixel. FIG. 6 is a diagram illustrating the timing of exposure of each first pixel when an electronic shutter is used. FIG. 7 is a diagram illustrating the timing of exposure of each first pixel when an electronic shutter is used. FIG. 8 is a diagram illustrating the timing of exposure of each first pixel when an electronic shutter is used. FIG. 9 is a diagram illustrating processing performed by an analysis unit. FIG. 10 is a diagram illustrating the linearity characteristics of each of a first amplifier and a second amplifier. FIG. 11 is a diagram illustrating the correspondence between a first exposure time and a first correction coefficient, and the correspondence between a second exposure time and a second correction coefficient. FIG. 12 is a diagram illustrating wavelength axis correction. FIG. 13 is a diagram showing a first correspondence relationship between the position of the first pixel row and the wavelength of the spectral image, and a second correspondence relationship between the position of the second pixel row and the wavelength of the spectral image. FIG. 14 is a diagram for explaining generation of each of the first spectral data and the second spectral data. FIG. 15 is a diagram for explaining correction of input dependency. FIG. 16 is a diagram for explaining stray light correction. FIG. 17 is a diagram for explaining generation of spectral data of the light to be measured. FIG. 18 is a diagram showing the relationship between the charge discarding time and the intensity of the first electrical signal when an electronic shutter is used. FIG. 19 is a diagram showing the correspondence relationship between the charge discarding time and the third correction coefficient. FIG. 20 is a diagram showing the first correspondence relationship corrected based on the first exposure time. FIG. 21 is a diagram for explaining the anti-blooming function of the photodetector. FIG. 22 is a diagram for explaining the light diffusing unit.

[0024] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In each drawing, the same or corresponding parts are denoted by the same reference numerals, and duplicated explanations will be omitted. [Configuration of Spectroscopic Measurement Device]

[0025] As shown in FIG. 1 , the spectroscopic measurement device 1 includes a spectrometer 10 and a computer device 20. The computer device 20 is, for example, a personal computer and is configured with a processing unit, a memory unit, a display unit, and an input unit. The processing unit is configured with a processor, memory, storage, a communication device, etc., and processes various data by executing software (programs). The memory unit is, for example, a hard disk, and stores various data. The display unit is, for example, a display, and displays various data to an operator. The input unit is, for example, a mouse and keyboard, and accepts input of various data from an operator. The spectrometer 10 includes a bundle fiber 2, an optical system 3, a photodetector 4, a memory unit 7, a control board 8, and a housing 9. In this embodiment, the memory unit 7, the control board 8, and the computer device 20 configure an analysis unit 70 that generates spectral data of the measured light L1. The spectroscopic measurement device 1 is a device that generates spectral data of the measured light L1 by spectrally separating the measured light L1.

[0026] The bundle fiber 2 guides the light to be measured L1 to the optical system 3. The bundle fiber 2 has a circular incident end face 2a and a long exit end face 2b. The bundle fiber 2 includes a plurality of optical fibers. As shown in FIGS. 2A and 2B, the bundle fiber 2 includes a plurality of first optical fibers 21, a plurality of second optical fibers 22, a plurality of third optical fibers 23, and a fourth optical fiber 24.

[0027] The incident end face 2a of the bundle fiber 2 has the following configuration. As shown in FIG. 2A , the fourth optical fiber 24 is arranged in the center of the incident end face 2a. The multiple third optical fibers 23 are lined up along the outer edge of the incident end face 2a. The multiple third optical fibers 23 are arranged at predetermined intervals. The multiple first optical fibers 21 and the multiple second optical fibers 22 are arranged between the fourth optical fiber 24 and the third optical fiber 23, and between the multiple third optical fibers 23. The multiple first optical fibers 21 and the multiple second optical fibers 22 are arranged adjacent to each other in the circumferential direction of the incident end face 2a. The multiple first optical fibers 21 and the multiple second optical fibers 22 may also be arranged adjacent to each other in the radial direction of the incident end face 2a.

[0028] The configuration of the output end face 2b of the bundle fiber 2 is as follows. As shown in FIG. 2B , the fourth optical fiber 24 is arranged in the center of the output end face 2b. The plurality of third optical fibers 23 are arranged on both ends of the output end face 2b. The plurality of first optical fibers 21 are arranged between the fourth optical fiber 24 and the plurality of third optical fibers 23 arranged at one end of the output end face 2b. In other words, the plurality of first optical fibers 21 are arranged on one side of the center of the output end face 2b. The plurality of second optical fibers 22 are arranged between the fourth optical fiber 24 and the plurality of third optical fibers 23 arranged at the other end of the output end face 2b. In other words, the plurality of second optical fibers 22 are arranged on the other side of the center of the output end face 2b.

[0029] 1, the optical system 3 has a light incident portion 31, a diffraction grating 32, and a lens 33. The optical system 3 guides the light to be measured L1 to the photodetector 4 and forms a spectral image of the light to be measured L1 on the light receiving portion of the photodetector 4. In this embodiment, the optical system 3 is a Dyson optical system. Also, in this embodiment, the diffraction grating 32 is a reflective diffraction grating. The light to be measured L1 is split by the diffraction grating 32 in a direction perpendicular to the direction in which the light to be measured L1 is incident. Here, the direction in which the light to be measured L1 is split is referred to as the X-axis direction, the direction perpendicular to the X-axis direction is referred to as the Y-axis direction, and the direction perpendicular to the X-axis and Y-axis directions is referred to as the Z-axis direction.

[0030] The light incident portion 31 is connected to the output end surface 2b of the bundle fiber 2. When viewed from the Z-axis direction, the output end surface 2b of the bundle fiber 2 is connected to the light incident portion 31 with the Y-axis direction as the longitudinal direction. The light incident portion 31 causes the light to be measured L1 guided from the bundle fiber 2 to enter the optical system 3. The light incident portion 31 adjusts the amount of incident light to be measured L1. The light incident portion 31 is, for example, a slit member. When viewed from the Z-axis direction, the slit formed in the slit member opens in a rectangular shape with the short side in the X-axis direction and the long side in the Y-axis direction.

[0031] The diffraction grating 32 faces the light incident portion 31 in the Z-axis direction. The diffraction grating 32 is composed of a plurality of grating grooves (not shown). The plurality of grating grooves are aligned along the X-axis direction with each grating groove extending along the Y-axis direction. The measured light L1 incident on the diffraction grating 32 is split according to wavelength along the X-axis direction, which is the direction in which the plurality of grating grooves are aligned.

[0032] The lens 33 is disposed in the Z-axis direction between the light incident portion 31 and the photodetector 4 and the diffraction grating 32. The lens 33 guides the measured light L1 incident from the light incident portion 31 to the diffraction grating 32, and forms a spectral image of the measured light L1 dispersed by the diffraction grating 32 on the light receiving portion of the photodetector 4.

[0033] The photodetector 4 faces the diffraction grating 32 in the Z-axis direction. The photodetector 4 detects the measured light L1 dispersed by the diffraction grating 32. In this embodiment, the photodetector 4 is a full-frame transfer CCD image sensor.

[0034] The storage unit 7 and the photodetector 4 are mounted on a control board 8. The storage unit 7 stores correction coefficients for correcting data detected by the photodetector 4. The storage unit 7 is a storage medium such as a RAM (Random Access Memory) or a ROM (Read Only Memory). The computer device 20 is electrically connected to the control board 8. The computer device 20 acquires data from the photodetector 4 and also acquires correction coefficients from the storage unit 7, corrects the data, and then generates spectral data of the measured light L1.

[0035] The housing 9 houses the diffraction grating 32, the lens 33, the photodetector 4, and the storage unit 7. The light incident unit 31 and the control board 8 are attached to the wall of the housing 9. [Configuration and Operation of the Photodetector]

[0036] 3, the photodetector 4 has a first light receiving section 5 and a second light receiving section 6. The first light receiving section 5 and the second light receiving section 6 are arranged side by side along the Y-axis direction. The wavelength axis A of the spectral image α formed on the first light receiving section 5 and the second light receiving section 6 extends in the X-axis direction. The images of the spectral image α for each wavelength extend in the Y-axis direction. It is preferable that the spectral image α has a shape that is line-symmetric with respect to the boundary line between the first light receiving section 5 and the second light receiving section 6.

[0037] The first light receiving unit 5 has a plurality of first pixel rows 51 arranged in the X-axis direction. Each of the plurality of first pixel rows 51 includes a plurality of first pixels 52 arranged in the Y-axis direction. The first light receiving unit 5 has, for example, 2168 first pixel rows 51. Each of the plurality of first pixel rows 51 includes, for example, 128 first pixels 52. The second light receiving unit 6 has a plurality of second pixel rows 61 arranged in the X-axis direction. Each of the plurality of second pixel rows 61 includes a plurality of second pixels 62 arranged in the Y-axis direction. The second light receiving unit 6 has, for example, 2168 second pixel rows 61. Each of the plurality of second pixel rows 61 includes, for example, 128 second pixels 62. By receiving the spectral image α during the first exposure time, each of the plurality of first pixels 52 generates and accumulates a first charge Q1 in an amount corresponding to the intensity of the received spectral image α. Each of the second pixels 62 receives the spectral image α during the second exposure time, and thereby generates and accumulates a second charge Q2 in an amount corresponding to the intensity of the received spectral image α. The first exposure time is shorter than the second exposure time.

[0038] The photodetector 4 further includes a first vertical transfer unit 41, a first horizontal transfer unit 42, and a first amplifier 43. The first vertical transfer unit 41 transfers the first charges Q1 accumulated in each of the first pixels 52 in the Y-axis direction. Hereinafter, transfer in the Y-axis direction will be referred to as "transfer in the vertical direction." In this embodiment, because the photodetector 4 is a full-frame transfer CCD image sensor, the first vertical transfer unit 41 is not separated from each of the first pixel columns 51. In this case, the first vertical transfer unit 41 is a plurality of transfer electrodes (not shown) that transfer the first charges Q1 in the vertical direction along each of the first pixel columns 51. The first horizontal transfer unit 42 is adjacent to the first light receiving unit 5 on the opposite side from the second light receiving unit 6 and extends in the X-axis direction. The first horizontal transfer unit 42 transfers the first charges Q1 transferred from each of the first pixel columns 51 by the first vertical transfer unit 41 in the X-axis direction. Hereinafter, transfer in the X-axis direction will be referred to as "transfer in the horizontal direction."

[0039] The photodetector 4 further includes a second vertical transfer unit 44, a second horizontal transfer unit 45, and a second amplifier 46. The second vertical transfer unit 44 vertically transfers the second charges Q2 accumulated in each of the second pixels 62. In this embodiment, because the photodetector 4 is a full-frame transfer CCD image sensor, the second vertical transfer unit 44 is not separated from each second pixel column 61. In this case, the second vertical transfer unit 44 is a plurality of transfer electrodes (not shown) that vertically transfer the second charges Q2 along each second pixel column 61. The second horizontal transfer unit 45 is adjacent to the second light receiving unit 6 on the opposite side from the first light receiving unit 5 and extends in the X-axis direction. The second horizontal transfer unit 45 horizontally transfers the second charges Q2 transferred from each second pixel column 61 by the second vertical transfer unit 44.

[0040] In this embodiment, the photodetector 4 performs vertical and horizontal transfer using FVB (Full Vertical Binning). In the first light receiving section 5, the first charges Q1 of the multiple first pixels 52 are summed for each first pixel column 51 in the first horizontal transfer unit 42. The summed first charges Q1 for each first pixel column 51 in the first horizontal transfer unit 42 are then sequentially horizontally transferred from the first horizontal transfer unit 42. The photodetector 4 operates in one frame from the start of transfer of the first charges Q1 by the first vertical transfer unit 41 to the end of transfer of the first charges Q1 by the first horizontal transfer unit 42. The first amplifier 43 generates a first electrical signal ES1 corresponding to the amount of the first charges Q1 transferred from the first horizontal transfer unit 42. The first electrical signal ES1 is, for example, a voltage value. The first electrical signal ES1 is converted from analog to digital by an AD converter (not shown) to a digital value.

[0041] On the second light receiving section 6 side, the second charges Q2 of the multiple second pixels 62 are added together for each second pixel column 61 in the second horizontal transfer section 45. The second charges Q2 added together for each second pixel column 61 in the second horizontal transfer section 45 are then sequentially horizontally transferred from the second horizontal transfer section 45. The photodetector 4 operates in one frame from the start of transfer of the second charges Q2 by the second vertical transfer section 44 to the end of transfer of the second charges Q2 by the second horizontal transfer section 45. The second amplifier 46 generates a second electric signal ES2 corresponding to the amount of the second charges Q2 transferred from the second horizontal transfer section 45. The second electric signal ES2 is, for example, a voltage value. The second electric signal ES2 is AD converted into a digital value by an AD converter (not shown).

[0042] Focusing on one first pixel column 51, the exposure timing of each first pixel 52 per frame operation will be described. First, with reference to FIG. 4, the exposure timing of each first pixel 52 from the start of vertical transfer of the first charge Q1 by the first vertical transfer unit 41 to the end of vertical transfer will be described. For simplicity's sake, the following pixels in a certain first pixel column 51 will be used as an example. The first pixels 52a to 52d are arranged in the order of first pixel 52d, first pixel 52c, first pixel 52b, and first pixel 52a, from closest to the first horizontal transfer unit 42. Exposure times t0 to t4 indicate predetermined timings within the first exposure time.

[0043] Before the exposure time t0, the first charge Q ｔ０ is stored, and the first charge Q ｔ０ is left without being vertically transferred to the first horizontal transfer unit 42. The vertical transfer in one frame operation is performed by transferring the first charge Q ｔ０ On the other hand, even after the exposure time t1, the first pixel row 51 continues to be exposed, and new first charges Q1 are accumulated. The new first charges Q1 generated during the exposure time t1 are transferred as the first charges Q ｔ１, the first charge Q1 newly generated during exposure time t2 is the first charge Q ｔ２ , the first charge Q1 newly generated during exposure time t3 is the first charge Q ｔ３ , the first charge Q1 newly generated at exposure time t4 is the first charge Q ｔ４ Let's say.

[0044] First, the first charge Q accumulated in the first pixel 52d during the exposure time t0 ｔ０ is transferred to the first horizontal transfer unit 42. At this time, a new first charge Q ｔ１ Here, the first charge Q accumulated in the first pixel 52 a of the first pixel column 51 is ｔ１ is transferred in the next frame operation. Subsequently, during exposure time t2, the first charge Q accumulated in the first pixel 52c is transferred. ｔ０ , and the first charge Q accumulated in the first pixel 52d ｔ１ is transferred to the first horizontal transfer unit 42. At this time, a new first charge Q ｔ２ Here, the first charge Q accumulated in the first pixel 52b of the first pixel column 51 is ｔ２ Next, during exposure time t3, the first charge Q accumulated in the first pixel 52b is transferred. ｔ０ , the first charge Q accumulated in the first pixel 52c ｔ１ , and the first charge Q accumulated in the first pixel 52d ｔ２ is transferred to the first horizontal transfer unit 42. At this time, a new first charge Q ｔ３ Here, the first charge Q accumulated in the first pixel 52c of the first pixel column 51 is ｔ３ will be transferred in the next frame operation.

[0045] Finally, at exposure time t4, the first charge Q accumulated in the first pixel 52a ｔ０ , the first charge Q accumulated in the first pixel 52b ｔ１ , the first charge Q accumulated in the first pixel 52c ｔ２ , and the first charge Q accumulated in the first pixel 52d ｔ３ is transferred to the first horizontal transfer unit 42. At this time, a new first charge Q ｔ４ Here, the first charge Q accumulated in the first pixel 52d of the first pixel column 51 isｔ４ As described above, the exposure timing for each first pixel 52 at which the transfer of the first charge Q1 starts in the next frame operation is the earliest for the first pixel 52a (exposure time t1) and the latest for the first pixel 52d (exposure time t4).

[0046] Next, with reference to FIG. 5 , the exposure timing of each first pixel 52, including operations after the end of vertical transfer, will be described. FIG. 5 represents a coordinate system in which the horizontal axis represents the first exposure time and the vertical axis represents the position of each first pixel 52. The position of each first pixel 52 coincides with the position of each first pixel 52 in the Y-axis direction in FIG. 3 . In FIG. 5 , the origin side is the side opposite the first horizontal transfer unit 42 (the boundary line between the first light receiving unit 5 and the second light receiving unit 6 in FIG. 3 ). As in FIG. 4 , exposure times t1 to t4 represent the vertical transfer time VT from the start of vertical transfer of the first charge Q1 by the first vertical transfer unit 41 to the end of vertical transfer. Subsequently, horizontal transfer by the first horizontal transfer unit 42 begins at exposure time t4, and the horizontal transfer ends at exposure time t5. Exposure times t4 to t5 represent the horizontal transfer time HT. The exposure time, which is the sum of the vertical transfer time VT and the horizontal transfer time HT, represents one frame time FT required for one frame operation. Subsequently, vertical transfer in the next frame operation starts from exposure time t5, and ends at exposure time t6.

[0047] During exposure times t1 to t4, as described with reference to FIG. 4 , the exposure timing for each first pixel 52 that will start transferring the first charge Q1 in the next frame operation is slower for each first pixel 52 on the first horizontal transfer unit 42 side and faster for each first pixel 52 on the opposite side of the first horizontal transfer unit 42. During exposure times t4 to t5, the first charge Q1 that was not vertically transferred in the previous frame operation is horizontally transferred, while all of the multiple first pixels 52 are exposed at the same timing. During exposure times t5 to t6, exposure is further performed for each first pixel 52 that will start transferring the first charge Q1 in the next frame operation. In other words, during exposure times t5 to t6, exposure ends for each first pixel 52 that will start transferring the first charge Q1 in the next frame operation. The exposure ends slower for each first pixel 52 on the first horizontal transfer unit 42 side and faster for each first pixel 52 on the opposite side of the first horizontal transfer unit 42. From the above, the shape of the exposure timing area 5a formed in a coordinate system with the first exposure time on the horizontal axis and the position of each first pixel 52 on the vertical axis is a parallelogram with the opposing short sides tilted, when the horizontal axis is the long side and the vertical axis is the short side.

[0048] The photodetector 4 can set the first exposure time to a time shorter than one frame time FT. In this case, the photodetector 4 uses an electronic shutter function to discard the accumulated first charge Q1 for a predetermined time (charge discard time) from the start of one frame time FT. As a result, the photodetector 4 may use the remaining time of one frame time FT as the first exposure time. As shown in FIG. 6 , when the electronic shutter is operated between exposure times t1 and t4, the first charge Q1 vertically transferred to the first horizontal transfer unit 42 during vertical transfer and the first charge Q1 accumulated in the multiple first pixels 52 during vertical transfer are all discarded during the charge discard time DT. When the first exposure time ET1 begins, vertical transfer and accumulation of the first charge Q1 are resumed.

[0049] As shown in FIG. 7 , when the electronic shutter is operated between exposure times t4 and t5, in addition to the first charge Q1 discarded in the case of FIG. 6 , all of the first charge Q1 accumulated in the multiple first pixels 52 during horizontal transfer is discarded. When the first exposure time ET1 begins, horizontal transfer and accumulation of the first charge Q1 resume. When the electronic shutter is operated between exposure times t4 and t5, the region 5a becomes trapezoidal. Specifically, when the first horizontal transfer unit 42 is defined as the upper edge and the opposite side to the first horizontal transfer unit 42 is defined as the lower edge, the region 5a becomes a trapezoid with the upper edge longer than the lower edge. When the region 5a is trapezoidal, the amount of first charge Q1 transferred to the first horizontal transfer unit 42 at the end of exposure time t6 is such that the amount of charge accumulated in each first pixel 52 on the first horizontal transfer unit 42 side is greater and the amount of charge accumulated in each first pixel 52 on the opposite side to the first horizontal transfer unit 42 is smaller.

[0050] As shown in FIG. 8 , when the electronic shutter is operated between exposure times t5 and t6, in addition to the first charges Q1 discarded in the case of FIG. 7 , the first charges Q1 vertically transferred to the first horizontal transfer unit 42 during vertical transfer in the next frame operation and the first charges Q1 accumulated in the multiple first pixels 52 during vertical transfer are all discarded. When the first exposure time ET1 begins, vertical transfer and accumulation of the first charges Q1 resume in the next frame operation. When the electronic shutter is operated between exposure times t5 and t6, the region 5a becomes triangular. Specifically, the region 5a becomes triangular with the first horizontal transfer unit 42 side as the base and the opposite side to the first horizontal transfer unit 42 as the apex. When the region 5a becomes triangular, the first charges Q1 transferred to the first horizontal transfer unit 42 at the end of exposure time t6 are, for example, only the charges accumulated in each first pixel 52 on the first horizontal transfer unit 42 side. [Generation of Spectral Data by the Analysis Unit]

[0051] As shown in FIG. 9 , the analyzer 70 generates spectral data of the measured light L1 through multiple correction steps. The analyzer 70 first performs linearity correction (S01). As an example of linearity correction, as shown in FIG. 10 , the linearity characteristic R1 of the first amplifier 43 and the linearity characteristic R2 of the second amplifier 46 do not completely match. Furthermore, the linearity characteristic R1 of the first amplifier 43 and the linearity characteristic R2 of the second amplifier 46 deviate from the reference linearity characteristic R0. The linearity characteristic R1 of the first amplifier 43 and the linearity characteristic R2 of the second amplifier 46 are represented by a graph with the horizontal axis representing the converted intensity of the electrical signal and the vertical axis representing the linearity error. For the linearity characteristic R1 of the first amplifier 43, the converted intensity of the first electrical signal ES1 is calculated by (reference intensity / reference exposure time)×first exposure time ET1. For the linearity characteristic R2 of the second amplifier 46, the converted intensity of the second electrical signal ES2 is calculated by (reference intensity / reference exposure time)×second exposure time ET2. In the case of the first amplifier 43, the linearity error is expressed as the absolute value of the difference between the intensity of the first electrical signal ES1 and the converted intensity of the first electrical signal ES1, while in the case of the second amplifier 46, the linearity error is expressed as the absolute value of the difference between the intensity of the second electrical signal ES2 and the converted intensity of the second electrical signal ES2. The reference intensity is the signal intensity acquired with the reference exposure time. The reference exposure time is set to the shortest possible exposure time within the range in which data reliability can be ensured.

[0052] The linearity error of the reference linearity characteristic R0 is zero based on the linearity error calculation formula described above. The linearity error of the linearity characteristic R1 of the first amplifier 43 and the linearity characteristic R2 of the second amplifier 46 increases as the converted strength of the electrical signal increases. By matching the linearity characteristic R1 of the first amplifier 43 and the linearity characteristic R2 of the second amplifier 46 to the reference linearity characteristic R0, the linearity error of each of the linearity characteristic R1 of the first amplifier 43 and the linearity characteristic R2 of the second amplifier 46 can be corrected.

[0053] In the linearity correction, the analyzer 70 corrects the first electrical signal ES1 using a first correction coefficient k1. As shown in FIG. 11A , the analyzer 70 stores a table TL1 of first correction coefficients k1 corresponding to the converted intensities of the first electrical signal ES1 in the storage unit 7. The first correction coefficient k1 is determined in advance based on the linearity characteristic R1 of the first amplifier 43. The analyzer 70 multiplies the first electrical signal ES1 by the first correction coefficient k1 to match the linearity characteristic R1 of the first amplifier 43 with the reference linearity characteristic R0. Therefore, the first correction coefficient k1 is determined so that the correction amount of the linearity characteristic R1 increases as the converted intensity of the first electrical signal ES1 increases. The converted intensity of the first electrical signal ES1 is calculated, for example, as follows: An operator performing spectroscopic measurement using the spectroscopic measurement device 1 inputs an instruction regarding the value of the first exposure time ET1 to the analyzer 70 using an input unit connected to the analyzer 70. The analysis unit 70 then calculates the converted intensity of the first electrical signal ES1 based on the specified value. For example, if the calculated converted intensity of the first electrical signal ES1 is s11, the analysis unit 70 reads k11 from the table TL1 as the corresponding first correction coefficient k1. The analysis unit 70 then corrects the first electrical signal ES1 using the read k11.

[0054] In the linearity correction, the analyzer 70 corrects the second electrical signal ES2 using a second correction coefficient k2. As shown in FIG. 11B , the analyzer 70 stores a table TL2 of second correction coefficients k2 corresponding to the converted intensities of the second electrical signal ES2 in the storage unit 7. The second correction coefficient k2 is determined in advance based on the linearity characteristic R2 of the second amplifier 46. The analyzer 70 multiplies the second electrical signal ES2 by the second correction coefficient k2 to match the linearity characteristic R2 of the second amplifier 46 with the reference linearity characteristic R0. Therefore, the second correction coefficient k2 is determined such that the correction amount of the linearity characteristic R2 increases as the converted intensity of the second electrical signal ES2 increases. The converted intensity of the second electrical signal ES2 is calculated, for example, as follows: An operator performing spectroscopic measurement using the spectroscopic measurement device 1 provides instructions to the analyzer 70 using an input unit connected to the analyzer 70. The analysis unit 70 then calculates the converted intensity of the second electrical signal ES2 based on the specified value. For example, if the calculated converted intensity of the second electrical signal ES2 is s21, the analysis unit 70 reads k21 from the table TL2 as the corresponding second correction coefficient k2. The analysis unit 70 then corrects the second electrical signal ES2 using the read k21. Note that the correction of the first electrical signal ES1 and the correction of the second electrical signal ES2 may be performed in either order, or may be performed simultaneously.

[0055] The analysis unit 70 then performs wavelength axis correction (S02). As shown in FIG. 12 , the image for each wavelength of the spectral image α may be an image parallel to the Y-axis direction (image for wavelength λ1), or an image having an arched shape due to the characteristics of the optical system 3 (image for wavelength λ2). The analysis unit 70 sets a coordinate system in which the boundary between the first light receiving unit 5 and the second light receiving unit 6 is the x-axis and the axis perpendicular to the x-axis is the y-axis. In the coordinate system, the analysis unit 70 sets a first correspondence relationship between the position on the wavelength axis of each of the multiple first pixel rows 51 and the wavelength on the wavelength axis, and a second correspondence relationship between the position on the wavelength axis of each of the multiple second pixel rows 61 and the wavelength on the wavelength axis. The positions on the wavelength axis of each of the multiple first pixel rows 51 and the positions on the wavelength axis of each of the multiple second pixel rows 61 are determined based on a reference x-coordinate WL ０ (λ) Reference x-coordinate WL ０(λ) is the x-coordinate of the point where the image for each wavelength intersects with the x-axis. For example, the reference x-coordinate WL of the image of wavelength λ1 ０ (λ1) is x1, and the reference x-coordinate WL of the image of wavelength λ2 ０ (λ2) is x2.

[0056] The position on the wavelength axis of the first pixel row 51, where an image of a certain wavelength of the spectral image α is formed, and the position on the wavelength axis of the second pixel row 61 may be misaligned due to the characteristics of the optical system 3. Furthermore, the positions on the wavelength axis of the first pixel row 51 and the second pixel row 61 may vary from one spectrometer 1 to another. Therefore, the position on the wavelength axis of the first pixel row 51 is measured in advance for each spectrometer 1, and the analysis unit 70 stores the measurement results as a first correspondence table TL3, as shown in FIG. 13A. Similarly, the position on the wavelength axis of the second pixel row 61 is measured in advance for each spectrometer 1, and the analysis unit 70 stores the measurement results as a second correspondence table TL4, as shown in FIG. 13B. For example, in FIG. 13A, the position of the first pixel row 51 corresponding to wavelength λ3 is x3. x3 is the reference x-coordinate WL in the first pixel row 51. ０ 13B, the position of the second pixel row 61 corresponding to the wavelength λ3 is shifted from the first pixel row 51 and is x5. x5 is the reference x-coordinate WL of the second pixel row 61. ０ The analyzing unit 70 generates first spectral data (to be described later) based on the first correspondence relationship TL3, and generates second spectral data (to be described later) based on the second correspondence relationship TL4.

[0057] The analysis unit 70 then generates first and second spectral data (S03). In step S03, the analysis unit 70 generates the first spectral data based on the corrected first electrical signal ES1. The analysis unit 70 also generates second spectral data based on the corrected second electrical signal ES2. As shown in FIG. 14 , the analysis unit 70 can obtain intensities in all wavelength bands for the first spectral data S1 without saturating the first pixels 52 in any wavelength band. In contrast, the second spectral data S2 includes a wavelength band (saturation wavelength band Δλ1) in which the second pixels 62 are saturated. Therefore, the analysis unit 70 cannot accurately obtain intensities in the saturation wavelength band Δλ1. If the wavelength band excluding the saturation wavelength band Δλ1 is defined as the non-saturation wavelength band Δλ2, the first spectral data S1 contains noise and has a poor S / N ratio in the non-saturation wavelength band Δλ2. On the other hand, in the second spectral data S2, noise is not superimposed even in the non-saturated wavelength band Δλ2, and the analyzer 70 can obtain highly accurate data.

[0058] The analyzer 70 then performs sensitivity correction and exposure time correction (S04). In the sensitivity correction, the analyzer 70 calculates the function I of the first spectrum data S1 when the first light receiving unit 5 and the second light receiving unit 6 are exposed for the same exposure time. calib_S (λ) (first function) and the function I of the second spectrum data S2 calib_L (λ) (second function). Then, the analysis unit 70 generates the function I calib_S (λ) and function I calib_L At least one of the first spectrum data S1 and the second spectrum data S2 is corrected based on the ratio of the function I calib_S (λ) and function I calib_L (λ) are ideally the same value, so the ratio between them is ideally 1. However, the light receiving sensitivity of the first light receiving section 5 and the light receiving sensitivity of the second light receiving section 6 do not necessarily match, and this difference is corrected by sensitivity correction. Sensitivity correction coefficient SR calib (λ) is shown by Equation 1. SR calib (λ) = I calib_L (λ) / I calib_S (λ)…(1)

[0059] In the exposure time correction, the analysis unit 70 corrects at least one of the first spectrum data S1 and the second spectrum data S2 based on the ratio between the first exposure time ET1 and the second exposure time ET2. The exposure time correction coefficient DRGain is expressed by Equation 2: DRGain=ET2 / ET1 (2) Sensitivity correction coefficient SR calib (λ) and the exposure time correction coefficient DRGain, the intensity I in the second spectrum data S2 is calculated. Ｌ When correcting (λ), the corrected intensity I Ｌ (λ)′ is shown by Equation 3. Ｌ (λ)′=I Ｌ (λ) / (SR calib (λ)×DRGain) (3) where the intensity I in the first spectrum data S1 Ｓ When correcting (λ), the sensitivity correction coefficient SR calib (λ) × exposure time correction coefficient DRGain = I Ｓ Multiply by (λ).

[0060] The analyzer 70 then corrects the input dependency of the measured light L1 (S05). It is preferable that the measured light L1 be received at the same intensity by the first light-receiving unit 5 and the second light-receiving unit 6. However, when the measured light L1 enters the optical system 3, the intensity distribution of the measured light L1 in its cross section may become non-uniform within the plane. In this case, a difference may occur between the light intensity at the first pixel row 51 where an optical image of a certain wavelength in the spectral image α is formed and the light intensity at the second pixel row 61 where the optical image is formed. In other words, the first light-receiving unit 5 and the second light-receiving unit 6 may be affected by the input dependency of the measured light L1. In the input dependency correction, the analyzer 70 corrects the difference in light intensity.

[0061] In correcting the input dependency, the analyzer 70 corrects at least one of the first spectral data S1 and the second spectral data S2 based on the ratio between a representative value of the intensity of the first spectral data S1 in the non-saturated wavelength band Δλ2 and a representative value of the intensity of the second spectral data S2 in the non-saturated wavelength band Δλ2. Here, the representative value is, for example, the first area S11 of the first spectral data S1 and the second area S21 of the second spectral data S2, as shown in FIG. 15 . Alternatively, the representative value may be the median of the intensity of the first spectral data S1 and the median of the intensity of the second spectral data S2. The intensity range used by the analyzer 70 to calculate the first area S11 and the second area S21 is between Imax×0.9 and Imax×0.1, where Imax is the value at which the intensity in the second spectral data S2 saturates. The range between Imax×0.9 and Imax×0.1 is a range in which the linearity characteristics are relatively good, and therefore the accuracy of correction can be improved. Here, good linearity characteristics refers to a range in which the rate of increase of the linearity error in the linearity characteristics R1 and R2 in FIG. 10 is relatively gradual.

[0062] The wavelength range for calculating the first area S11 and the second area S21 is the non-saturated wavelength band Δλ2. The first area S11 is the area of ​​the first spectral data S1 surrounded by the range between Imax×0.9 and Imax×0.1 and the non-saturated wavelength band Δλ2. The second area S21 is the area of ​​the second spectral data S2 surrounded by the range between Imax×0.9 and Imax×0.1 and the non-saturated wavelength band Δλ2. The analyzer 70 corrects at least one of the first spectral data S1 and the second spectral data S2 based on an input-dependent correction coefficient IR, which is the ratio between the first area S11 and the second area S21. The input-dependent correction coefficient IR and the sensitivity correction coefficient SR calib (λ) and the exposure time correction coefficient DRGain, the intensity I in the second spectrum data S2 is calculated. Ｌ When correcting (λ), the corrected intensity I Ｌ (λ)' is given by Equation 4. Intensity I Ｌ (λ)′=I Ｌ (λ) / (SR calib(λ)×DRGain×IR) (4) where the intensity I in the first spectrum data S1 Ｓ When correcting (λ), the sensitivity correction coefficient SR calib (λ) × exposure time correction coefficient DRGain × input dependency correction coefficient IR = I Ｓ Multiply by (λ).

[0063] The analyzer 70 then performs stray light correction (S06). As described above, the second spectral data S2 has a higher S / N ratio than the first spectral data S1 in the non-saturated wavelength band Δλ2. Therefore, when generating the spectral data of the measured light L1 (described later), using the second spectral data S2 in the non-saturated wavelength band Δλ2 can suppress a decrease in detection accuracy. However, if stray light occurs in the second spectral data S2, the stray light will also occur in the spectral data of the measured light L1, resulting in a decrease in detection accuracy. Stray light occurs in the optical system 3, for example, when a portion of the measured light L1 is multiple-reflected within the lens 33. In the stray light correction, the analyzer 70 corrects the second spectral data S2 in the non-saturated wavelength band Δλ2 to remove stray light generated in the optical system 3.

[0064] As shown in FIG. 16 , stray light L0 occurs in the non-saturated wavelength band Δλ2 of the second spectral data S2. The analysis unit 70 removes stray light L0, for example, as follows: The analysis unit 70 calculates a stray light signal distribution matrix that quantifies the intensity of stray light L0 in the second spectral data S2. Next, the analysis unit 70 derives a stray light correction matrix by calculating the inverse matrix of the sum of the stray light signal distribution matrix and a unit matrix having the same number of columns and columns as the stray light signal distribution matrix. The analysis unit 70 then corrects the stray light L0 by matrix multiplication based on the stray light correction matrix.

[0065] Finally, the analyzer 70 generates spectral data of the measured light L1 (S07). As shown in FIG. 17 , the analyzer 70 generates spectral data S3 of the measured light L1 by combining data of the saturation wavelength band Δλ1 of the first spectral data S1 and data of the non-saturation wavelength band Δλ2 of the second spectral data S2. For example, for the first spectral data S1, the analyzer 70 divides the corrected first electrical signal ES1 by the first exposure time ET1 and multiplies the result by the reference exposure time. For the second spectral data S2, the analyzer 70 divides the corrected second electrical signal ES2 by the second exposure time ET2 and multiplies the result by the reference exposure time. Thus, the analyzer 70 generates spectral data S3 of the measured light L1 by aligning the scale of the first spectral data S1 with the scale of the second spectral data S2. In the spectrum data S3, the intensities in all wavelength bands can be acquired without saturating the pixels in all wavelength bands and without superimposing noise.

[0066] In the spectroscopic measurement device 1, the photodetector 4 receives the spectral image α at the first light receiving unit 5 during a first exposure time ET1 and outputs a first electrical signal ES1, and receives the spectral image α at the second light receiving unit 6 during a second exposure time ET2, which is longer than the first exposure time ET1, and outputs a second electrical signal ES2. The analyzer 70 generates first spectral data S1 based on the first electrical signal ES1 and second spectral data S2 based on the second electrical signal ES2. The analyzer 70 then generates spectral data S3 of the measured light L1 based on the first spectral data S1 and the second spectral data S2. This allows the spectral data S3 of the measured light L1 to be generated over a wide dynamic range. Furthermore, in the spectroscopic measurement device 1, the analyzer 70 corrects the first electrical signal ES1 using a first correction coefficient k1 and corrects the second electrical signal ES2 using a second correction coefficient k2. This makes it possible to correct errors that occur in the linearity characteristic R1 of the first amplifier 43 and the linearity characteristic R2 of the second amplifier 46. Therefore, the spectroscopic measurement device 1 can acquire highly accurate spectral data S3 of the measured light L1.

[0067] In the spectroscopic measurement device 1, the analyzer 70 generates first spectral data S1 based on a first correspondence relationship between each of the multiple first pixel arrays 51 and a wavelength on the wavelength axis, and generates second spectral data S2 based on a second correspondence relationship between each of the multiple second pixel arrays 61 and a wavelength on the wavelength axis. In the spectroscopic measurement device 1, there may be a misalignment between the position of the first pixel array 51 where an optical image of a certain wavelength in the spectral image α is formed and the position of the second pixel array 61 where that optical image is formed. Even when such a misalignment occurs, the spectroscopic measurement device 1 can acquire the first spectral data S1 and the second spectral data S2 in which the misalignment has been corrected.

[0068] In the spectroscopic measurement device 1, the analyzer 70 corrects at least one of the first spectral data S1 and the second spectral data S2 based on the ratio between the representative intensity value of the first spectral data S1 and the representative intensity value of the second spectral data S2. In the spectroscopic measurement device 1, a difference may occur between the light intensity in the first pixel row 51 where an optical image of a certain wavelength in the spectral image α is formed and the light intensity in the second pixel row 61 where the optical image is formed. Even when such a difference in light intensity occurs, the spectroscopic measurement device 1 can acquire the first spectral data S1 and the second spectral data S2 in which the difference in light intensity has been corrected.

[0069] In the spectroscopic measurement device 1, the analyzer 70 corrects the second spectral data S2 to remove stray light L0 generated in the optical system 3. Focusing only on noise generated in the circuit (e.g., noise generated on the control board 8), the S / N of the second spectral data S2 is higher than the S / N of the first spectral data S1. However, the measured light L1 may contain stray light L0. In such cases, the noise caused by the stray light L0 increases by the amount that the second exposure time ET2 is longer than the first exposure time ET1, and the S / N of the second spectral data S2 decreases. According to the spectroscopic measurement device 1, even when the measured light L1 contains stray light L0, the second spectral data S2 from which the stray light L0 has been removed can be obtained.

[0070] The spectroscopic measurement device 1 further includes a bundle fiber 2 including a plurality of optical fibers that guide the light to be measured L1 to the optical system 3, and having a circular incident end face 2 a and an elongated exit end face 2 b, the plurality of optical fibers including a plurality of first optical fibers 21 arranged on one side of the center of the exit end face 2 b at the exit end face 2 b and a plurality of second optical fibers 22 arranged on the other side of the center of the exit end face 2 b at the exit end face 2 b, and at the entrance end face 2 a, the plurality of first optical fibers 21 and the plurality of second optical fibers 22 are adjacent to each other in at least one of the circumferential direction and the radial direction. In the spectroscopic measurement device 1, even if a bias in the light intensity of the light to be measured L1 occurs at the entrance end face 2 a of the bundle fiber 2, the bias in the light intensity of the light to be measured L1 is suppressed at the exit end face 2 b of the bundle fiber 2. Therefore, according to the spectroscopic measurement device 1, when focusing on the first pixel row 51 and the second pixel row 61 on which a light image of a certain wavelength is formed in the spectral image α, it is possible to obtain the first spectral data S1 and the second spectral data S2 while suppressing the occurrence of a difference between the light intensity in the first pixel row 51 and the light intensity in the second pixel row 61.

[0071] In the spectroscopic measurement device 1, the photodetector 4 has an electronic shutter function that discharges the first charges Q1 accumulated in each of the plurality of first pixels 52 for a predetermined time (charge discard time DT) from the start of one frame time FT1, which is from the start of transfer of the first charges Q1 by the first vertical transfer unit 41 to the end of transfer of the first charges Q1 by the first horizontal transfer unit 42, thereby setting the time excluding the predetermined time within one frame time FT as the first exposure time ET1. According to the spectroscopic measurement device 1, by using the electronic shutter function for the first light receiving unit 5, it is possible to prevent each of the first pixels 52 in the first light receiving unit 5 from saturating, and it is possible to reliably acquire the first spectral data S1 in the saturation wavelength band Δλ1.

[0072] In the spectroscopic measurement device 1, when the first light receiving unit 5 and the second light receiving unit 6 are exposed for the same exposure time, the analyzer 70 generates a first function of the first spectral data S1 and a second function of the second spectral data S2, and corrects at least one of the first spectral data S1 and the second spectral data S2 based on the ratio between the first function and the second function. In the spectroscopic measurement device 1, the light receiving sensitivity of the first light receiving unit 5 and the light receiving sensitivity of the second light receiving unit 6 do not necessarily match. According to the spectroscopic measurement device 1, even if a difference in light receiving sensitivity occurs, it is possible to obtain the first spectral data S1 and the second spectral data S2 in which such a difference in light receiving sensitivity has been corrected.

[0073] In the spectroscopic measurement device 1, the analyzer 70 corrects at least one of the first spectral data S1 and the second spectral data S2 based on the ratio between the first exposure time ET1 and the second exposure time ET2. The spectroscopic measurement device 1 can acquire the first spectral data S1 and the second spectral data S2 in which the influence of the difference in exposure time on the light intensity has been corrected.

[0074] In the spectroscopic measurement device 1, the photodetector 4 receives the spectral image α at the first light receiving unit 5 during a first exposure time ET1 and outputs a first electrical signal ES1, and receives the spectral image α at the second light receiving unit 6 during a second exposure time ET2 that is longer than the first exposure time ET1 and outputs a second electrical signal ES2. The storage unit 7 stores a first correction coefficient k1 for correcting the first electrical signal ES1 and a second correction coefficient k2 for correcting the second electrical signal ES2. Therefore, when generating spectral data S3 of the measured light L1 based on the first electrical signal ES1 and the second electrical signal ES2, errors occurring in the linearity characteristic R1 of the first amplifier 43 and the linearity characteristic R2 of the second amplifier 46 can be corrected by correcting the first electrical signal ES1 using the first correction coefficient k1 and correcting the second electrical signal ES2 using the second correction coefficient k2. Therefore, the spectroscopic measurement device 1 can acquire highly accurate spectral data S3 of the light to be measured L1.

[0075] According to the spectroscopic measurement method, for the same reason as the spectroscopic measurement device 1, it is possible to acquire highly accurate spectral data S3 of the measured light L1.

[0076] The present disclosure is not limited to the above-described embodiment. When the photodetector 4 uses the electronic shutter function to set the first exposure time ET1 to a time shorter than one frame time FT, as shown in FIG. 18 , the intensity of the first electrical signal ES1 decreases as the charge dump time DT increases. In FIG. 18 , the horizontal axis represents the charge dump time DT, and the vertical axis represents the intensity of the first electrical signal ES1. Here, the intensity of the first electrical signal ES1 is expressed as a percentage according to the charge dump time DT, assuming that the intensity of the first electrical signal ES1 when the charge dump time DT is 0 ms is 100%. When the charge dump time DT is 0 ms, the photodetector 4 does not operate the electronic shutter. When the charge dump time DT is around 0.1 ms, the intensity of the first electrical signal ES1 decreases rapidly. In other words, when the photodetector 4 switches from a state in which the electronic shutter is not operating to a state in which it is operating, the first electrical signal ES1 decreases rapidly. When the charge dump time DT is from approximately 0.2 ms to approximately 8.3 ms, the first electrical signal ES1 decreases linearly and inversely proportional to the charge dump time DT. If the photodetector 4 operates the electronic shutter during this period, the region 5a will be trapezoidal, as described in FIG. 7. When the charge dump time DT is from approximately 8.4 ms to approximately 10 ms, the first electrical signal ES1 decreases nonlinearly and inversely proportional to the charge dump time DT. If the photodetector 4 operates the electronic shutter during this period, the region 5a will be triangular, as described in FIG. 8.

[0077] As shown in FIG. 19 , the analysis unit 70 stores in the storage unit 7 a table TL5 that associates the third correction coefficient k3 with the charge dump time DT. As described above, when the electronic shutter function is used, the intensity of the first electrical signal ES1 decreases depending on the length of the charge dump time DT. The intensity of the first electrical signal ES1 may decrease linearly, nonlinearly, or suddenly. This changes the linearity characteristic R1 of the first amplifier 43, making it difficult to match the linearity characteristic R1 with the reference linearity characteristic R0 when performing linearity correction (S01) of the first amplifier 43 using the first correction coefficient k1. Therefore, the third correction coefficient k3 is calculated for each charge dump time DT by actual measurement or simulation based on the decreasing tendency of the intensity of the first electrical signal ES1 depending on the charge dump time DT. The analysis unit 70 stores the calculated results as a table TL5 of the third correction coefficient k3 corresponding to the charge dump time DT. The analyzer 70 corrects the first electrical signal ES1 using the third correction coefficient k3 instead of the first correction coefficient k1. This allows the linearity characteristic R1 of the first amplifier 43 to match the reference linearity characteristic R0 even when the electronic shutter function is used. This makes it possible to acquire the first spectral data S1 in which errors occurring in the linearity characteristic R1 of the first amplifier 43 have been corrected. The analyzer 70 may store the third correction coefficient k3 corresponding to the first exposure time ET1 as a table TL5. Alternatively, the analyzer 70 may store the third correction coefficient k3 corresponding to the converted intensity of the first electrical signal ES1 as a table TL5.

[0078] When the photodetector 4 uses the electronic shutter function to set the first exposure time ET1 to a time shorter than one frame time FT, there is a case where only the first charges Q1 accumulated in some of the first pixels 52 are vertically and horizontally transferred. For example, as shown in FIG. 8 , when the region 5a is triangular, the first charges Q1 transferred to the first horizontal transfer unit 42 at the end of the exposure time t6 are only the charges accumulated in the first pixels 52 on the first horizontal transfer unit 42 side. In this case, in FIG. 12 , of the first pixels 52 on which an image of wavelength λ2 is formed, only the first charges Q1 accumulated in some of the first pixels 52 close to the first horizontal transfer unit 42 are transferred. The some of the first pixels 52 close to the first horizontal transfer unit 42 are located at the reference x-coordinate WL ０ Therefore, the first electrical signal ES1 corresponding to the wavelength λ2 is shifted from the reference x-coordinate WL ０ In other words, the first correspondence relationship when the electronic shutter function is used is different from the first correspondence relationship when the electronic shutter function is not used.

[0079] Therefore, the first correspondence relationship corresponding to the length of the first exposure time ET1 is measured in advance, and the analysis unit 70 stores the measurement results in the storage unit 7 as a table TL6 of corrected first correspondence relationships, as shown in Fig. 20. For example, in Fig. 20, when the first exposure time ET1 is set to t32, the region 5a becomes triangular as shown in Fig. 8. In this case, the position of the first pixel row 51 corresponding to the wavelength λ2 is calculated based on the reference x-coordinate WL ０ (λ2) is not x2, which corresponds to the reference x-coordinate WL ０ x4, which takes into account the deviation from (λ2). The analysis unit 70 generates the first spectral data S1 based on the table TL6 of the corrected first correspondence relationships. This makes it possible to acquire the first spectral data S1 in which the deviation between the first correspondence relationship when the electronic shutter function is used and the correspondence relationship when the electronic shutter function is not used has been corrected. The analysis unit 70 may store the first correspondence relationship according to the length of the charge discarding time DT as table TL6. In other words, the analysis unit 70 may generate the first spectral data S1 based on the first correspondence relationship according to at least one of the charge discarding time DT and the first exposure time ET1.

[0080] The analysis unit 70 may use a formula instead of the corrected first correspondence relationship table TL6 to calculate the position on the wavelength axis of the first pixel row 51 corresponding to each wavelength. The calculation formula is shown in Equation 5: WL(t, λ)=WL ０ (λ)+Σy{w(y,t)×WL error (y, λ)} (5) In Equation 5, the position of each first pixel 52 in the vertical direction is y, the charge discard time DT is t, and the weight of the first charge Q1 transferred from each first pixel 52 is modeled as w(y, t). Equation 5 also expresses the amount of image distortion for each wavelength obtained by optical simulation as WL error (y, λ), and the reference x coordinate is WL ０ (λ). This allows the reference x-coordinate WL ０ The deviation from (λ) is expressed as weight w(y, t) and distortion WL error By calculating for each first pixel 52 from the result of multiplication by (y, λ), the position on the wavelength axis of the first pixel row 51 can be calculated with high precision.

[0081] In the above embodiment and modified example, the photodetector 4 uses the electronic shutter function to set the first exposure time ET1 to a time shorter than one frame time FT. Similarly, the photodetector 4 may use the electronic shutter function to set the second exposure time ET2 to a time shorter than one frame time FT. However, in this case, the photodetector 4 must set the first exposure time ET1 to a time shorter than the second exposure time ET2. When the electronic shutter function is used to set the second exposure time ET2, the intensity of the second electrical signal ES2 decreases depending on the charge discard time DT. Therefore, the analysis unit 70 may store a fourth correction coefficient k4 corresponding to the charge discard time DT and correct the second electrical signal ES2 using the fourth correction coefficient k4 instead of the second correction coefficient k2. Furthermore, the second correspondence relationship when the electronic shutter function is used differs from the second correspondence relationship when the electronic shutter function is not used. Therefore, the second correspondence relationship corresponding to the length of the second exposure time ET2 may be measured in advance, and the analysis unit 70 may store the measurement result as the corrected second correspondence relationship. Then, the analyzing section 70 may generate second spectral data S2 based on the corrected second correspondence relationship.

[0082] The analysis unit 70 may calculate the ratio between the third correction coefficient k3 and the fourth correction coefficient k4, and use the ratio between the third correction coefficient k3 and the fourth correction coefficient k4 to correct the intensity of the first spectrum data S1 or the intensity of the second spectrum data S2. For example, Ｌ When correcting (λ), the corrected intensity I Ｌ (λ)' is given by Equation 6. Intensity I Ｌ (λ)′=I Ｌ (λ) / {SR calib (λ)×DRGain×(k4 / k3)}…(6)

[0083] The photodetector 4 may have an anti-blooming function. When the second light receiving unit 6 is exposed for the second exposure time ET2, the amount of second charge Q2 accumulated in each second pixel 62 may exceed a specific amount. This may cause excess charge to overflow into adjacent second pixels 62 (blooming). When blooming occurs, the second spectral data S2 becomes data S20 affected by blooming, as shown in FIG. 21 . When generating the spectral data S3 of the measured light L1, the analyzer 70 cannot accurately combine the first spectral data S1 and the data S20 affected by blooming. Therefore, for example, the photodetector 4 may prevent blooming by providing a drain in the second light receiving unit 6 to discard excess charge. In this case, the photodetector 4 may use the anti-blooming function for the amount of second charge Q2 equal to or less than the value obtained by multiplying the specific amount by the maximum value of DRGain in Equation 2 (e.g., 100 times).

[0084] 22 , a light diffusion unit 11 may be disposed in front of the bundle fiber 2. The light diffusion unit 11 guides the measured light L1 to the incident end face 2a in a diffused state. The light diffusion unit 11 is, for example, a diffusion plate. Since the measured light L1 diffused by the light diffusion unit 11 is incident on the incident end face 2a of the bundle fiber 2, bias in the light intensity of the measured light L1 is suppressed at the incident end face 2a of the bundle fiber 2. In addition, bias in the light intensity of the measured light L1 is suppressed at the exit end face 2b of the bundle fiber 2. Therefore, when focusing on the first pixel row 51 and the second pixel row 61 on which a light image of a certain wavelength in the spectral image α is formed, the first spectral data S1 and the second spectral data S2 can be acquired while suppressing a difference in light intensity between the first pixel row 51 and the second pixel row 61.

[0085] The optical system 3 is not limited to a Dyson optical system and may be another optical system (e.g., a Czerny-Turner optical system). The diffraction grating 32 may be a transmission type diffraction grating. The photodetector 4 is not limited to a full-frame transfer CCD image sensor and may be another type of CCD image sensor, such as an interline type or a frame transfer type. The photodetector 4 may perform vertical and horizontal transfer by binning every n pixels (n is an integer between 2 and 127) instead of FVB. Alternatively, the photodetector 4 may vertically transfer charge one pixel at a time. The computer device 20 may be a device incorporated into the spectrometer 10 rather than a personal computer or the like. The computer device 20 may store correction coefficients instead of the memory unit 7. In this case, the computer device 20 may generate spectral data of the measured light L1 after correcting the data acquired from the photodetector 4. The exit end surface 2b of the bundle fiber 2 does not have to be elongated. For example, the exit end surface 2b may be circular. The bundle fiber 2 may include only one optical fiber. For example, the bundle fiber 2 may include any one of one first optical fiber 21 , one second optical fiber 22 , one third optical fiber 23 , and one fourth optical fiber 24 .

[0086] 1...spectroscopic measurement device, 2...bundle fiber, 2a...incident end face, 2b...exit end face, 3...optical system, 4...photodetector, 5...first light receiving section, 6...second light receiving section, 7...storage section, 8...control board, 11...light diffusion section, 21...first optical fiber, 22...second optical fiber, 41...first vertical transfer section, 42...first horizontal transfer section, 43...first amplifier, 44...second vertical transfer section, 45...second horizontal transfer section, 46...second amplifier, 51...first pixel row, 52...first pixel, 61...second pixel row, 62...second pixel, 70... Analysis unit, A...wavelength axis, ES1...first electrical signal, ES2...second electrical signal, ET1...first exposure time, ET2...second exposure time, FT...1 frame time, k1...first correction coefficient, k2...second correction coefficient, k3...third correction coefficient, L0...stray light, L1...measured light, Q1...first charge, Q2...second charge, R1, R2...linearity characteristic, S1...first spectral data, S2...second spectral data, S3...spectral data, α...spectral image, Δλ1...saturated wavelength band, Δλ2...non-saturated wavelength band.

Claims

1. an optical system for dispersing light to be measured; a photodetector for detecting a spectral image of the measured light dispersed by the optical system; an analysis unit that generates spectrum data of the measured light, The photodetector includes: a first light receiving unit having a plurality of first pixel rows arranged in a horizontal direction parallel to a wavelength axis of the spectral image, each of the plurality of first pixel rows including a plurality of first pixels arranged in a vertical direction perpendicular to the wavelength axis; a second light receiving section that is juxtaposed with the first light receiving section in the vertical direction and has a plurality of second pixel columns arranged in the horizontal direction, each of the second pixel columns including a plurality of second pixels arranged in the vertical direction; a first vertical transfer unit that transfers, in the vertical direction for each of the first pixel columns, a first charge that has been accumulated in each of the first pixels by receiving the spectral image in a first exposure time in the first light receiving unit; a first horizontal transfer unit that transfers the first charges transferred by the first vertical transfer unit in the horizontal direction; a first amplifier that outputs a first electric signal according to an amount of the first charges transferred by the first horizontal transfer unit; a second vertical transfer unit that transfers, in the vertical direction for each of the second pixel columns, second charges that have been accumulated in each of the second pixels by receiving the spectral image in the second light receiving unit for a second exposure time that is longer than the first exposure time; a second horizontal transfer unit that transfers the second charges transferred by the second vertical transfer unit in the horizontal direction; a second amplifier that outputs a second electric signal according to the amount of the second electric charges transferred by the second horizontal transfer unit, The analysis unit is a first correction coefficient for correcting the first electric signal so that the linearity characteristic of the first amplifier coincides with a reference linearity characteristic; a second correction coefficient for correcting the second electric signal so that the linearity characteristic of the second amplifier coincides with the reference linearity characteristic; correcting the first electrical signal using the first correction factor; correcting the second electrical signal using the second correction factor; generating first spectral data based on the corrected first electrical signal; generating second spectral data based on the corrected second electrical signal; a spectroscopic measurement device that generates the spectral data of the measured light based on the first spectral data and the second spectral data.

2. The analysis unit is generating the first spectral data based on a first correspondence relationship between each of the plurality of first pixel rows and a wavelength on the wavelength axis; The spectroscopic measurement device according to claim 1 , wherein the second spectral data is generated based on a second correspondence relationship between each of the plurality of second pixel rows and a wavelength on the wavelength axis.

3. 3. The spectroscopic measurement device according to claim 1, wherein the analysis unit corrects at least one of the first spectral data and the second spectral data based on a ratio between a representative value of intensity of the first spectral data and a representative value of intensity of the second spectral data.

4. The spectroscopic measurement device according to claim 1 , wherein the analyzer performs a correction on the second spectral data to remove stray light generated in the optical system.

5. a bundle fiber including a plurality of optical fibers for guiding the light to be measured to the optical system, the bundle fiber having a circular incident end face and a long exit end face; the plurality of optical fibers include a plurality of first optical fibers arranged on one side of the center of the output end face at the output end face, and a plurality of second optical fibers arranged on the other side of the center of the output end face at the output end face, 3 . The spectroscopic measurement device according to claim 1 , wherein, on the incident end surface, each of the plurality of first optical fibers and each of the plurality of second optical fibers are adjacent to each other in at least one of a circumferential direction and a radial direction.

6. a bundle fiber including a plurality of optical fibers for guiding the light to be measured to the optical system; 3. The spectroscopic measurement device according to claim 1, further comprising a light diffusing section disposed in front of the fiber bundle and guiding the light to be measured in a diffused state to an incident end face of the fiber bundle.

7. the photodetector has an electronic shutter function of discharging the first charges accumulated in each of the plurality of first pixels for a predetermined time from the start of one frame time from the start of transfer of the first charges by the first vertical transfer unit to the end of transfer of the first charges by the first horizontal transfer unit, thereby setting a time of the one frame time excluding the predetermined time as the first exposure time; The analysis unit is a third correction coefficient for correcting the first electric signal so that the linearity characteristic of the first amplifier coincides with the reference linearity characteristic when the electronic shutter function is used, the third correction coefficient being associated with the predetermined time and stored; The spectroscopic measurement device according to claim 1 , wherein the first electrical signal is corrected using the third correction coefficient.

8. the photodetector has an electronic shutter function of discharging the first charges accumulated in each of the plurality of first pixels for a predetermined time from the start of one frame time from the start of transfer of the first charges by the first vertical transfer unit to the end of transfer of the first charges by the first horizontal transfer unit, thereby setting a time of the one frame time excluding the predetermined time as the first exposure time; The spectroscopic measurement device according to claim 2 , wherein the analysis section generates the first spectral data based on the first correspondence relationship according to at least one of the predetermined time period and the first exposure time period.

9. 3. The spectroscopic measurement device of claim 1, wherein the photodetector has an electronic shutter function for discharging the first charges accumulated in each of the plurality of first pixels for a predetermined time from the start of one frame time from the start of transfer of the first charges by the first vertical transfer unit to the end of transfer of the first charges by the first horizontal transfer unit, thereby setting the time of the one frame time excluding the predetermined time as the first exposure time.

10. The analysis unit is 3. The spectroscopic measurement device according to claim 1, wherein when the first light receiving unit and the second light receiving unit are exposed for a same exposure time, a first function of the first spectral data and a second function of the second spectral data are generated, and at least one of the first spectral data and the second spectral data is corrected based on a ratio between the first function and the second function.

11. The spectroscopic measurement device according to claim 1 , wherein the analysis section corrects at least one of the first spectral data and the second spectral data based on a ratio between the first exposure time and the second exposure time.

12. an optical system for dispersing light to be measured; a photodetector for detecting a spectral image of the measured light dispersed by the optical system; A memory unit and a The photodetector includes: a first light receiving unit having a plurality of first pixel rows arranged in a horizontal direction parallel to a wavelength axis of the spectral image, each of the plurality of first pixel rows including a plurality of first pixels arranged in a vertical direction perpendicular to the wavelength axis; a second light receiving section that is juxtaposed with the first light receiving section in the vertical direction and has a plurality of second pixel columns arranged in the horizontal direction, each of the second pixel columns including a plurality of second pixels arranged in the vertical direction; a first vertical transfer unit that transfers, in the vertical direction for each of the first pixel columns, a first charge that has been accumulated in each of the first pixels by receiving the spectral image in a first exposure time in the first light receiving unit; a first horizontal transfer unit that transfers the first charges transferred by the first vertical transfer unit in the horizontal direction; a first amplifier that outputs a first electric signal according to an amount of the first charges transferred by the first horizontal transfer unit; a second vertical transfer unit that transfers, in the vertical direction for each of the second pixel columns, second charges that have been accumulated in each of the second pixels by receiving the spectral image in the second light receiving unit for a second exposure time that is longer than the first exposure time; a second horizontal transfer unit that transfers the second charges transferred by the second vertical transfer unit in the horizontal direction; a second amplifier that outputs a second electric signal according to the amount of the second electric charges transferred by the second horizontal transfer unit, The storage unit is a first correction coefficient for correcting the first electric signal so that the linearity characteristic of the first amplifier coincides with a reference linearity characteristic; a second correction coefficient for correcting the second electrical signal so that the linearity characteristic of the second amplifier coincides with the reference linearity characteristic;

13. A spectroscopic measurement method using a spectroscopic measurement device, comprising: The spectroscopic measurement device is an optical system for dispersing light to be measured; a photodetector for detecting a spectral image of the measured light dispersed by the optical system; an analysis unit that generates spectrum data of the measured light, The photodetector includes: a first light receiving unit having a plurality of first pixel rows arranged in a horizontal direction parallel to a wavelength axis of the spectral image, each of the plurality of first pixel rows including a plurality of first pixels arranged in a vertical direction perpendicular to the wavelength axis; a second light receiving section that is juxtaposed with the first light receiving section in the vertical direction and has a plurality of second pixel columns arranged in the horizontal direction, each of the second pixel columns including a plurality of second pixels arranged in the vertical direction; a first vertical transfer unit that transfers, in the vertical direction for each of the first pixel columns, a first charge that has been accumulated in each of the first pixels by receiving the spectral image in a first exposure time in the first light receiving unit; a first horizontal transfer unit that transfers the first charges transferred by the first vertical transfer unit in the horizontal direction; a first amplifier that outputs a first electric signal according to an amount of the first charges transferred by the first horizontal transfer unit; a second vertical transfer unit that transfers, in the vertical direction for each of the second pixel columns, second charges that have been accumulated in each of the second pixels by receiving the spectral image in the second light receiving unit for a second exposure time that is longer than the first exposure time; a second horizontal transfer unit that transfers the second charges transferred by the second vertical transfer unit in the horizontal direction; a second amplifier that outputs a second electric signal according to the amount of the second electric charges transferred by the second horizontal transfer unit, The spectroscopic measurement method includes: correcting the first electrical signal so that a linearity characteristic of the first amplifier coincides with a reference linearity characteristic, and correcting the second electrical signal so that a linearity characteristic of the second amplifier coincides with the reference linearity characteristic; generating first spectral data based on the corrected first electrical signal and generating second spectral data based on the corrected second electrical signal; generating the spectral data of the measured light based on the first spectral data and the second spectral data.