JPWO2018008069A1 - Spectroscopic measuring apparatus and method - Google Patents

Abstract

The spectrometry apparatus (1) performs extraction while continuously changing the wavelength of the extracted light by the filter (12) capable of changing the wavelength of the extracted light, the imaging unit (14) having a plurality of pixel regions, and the filter A control unit (16, 20) for sequentially exposing a plurality of pixel regions with light and outputting an output signal as a basis of a frame image from each of the plurality of pixel regions; and one of a plurality of pixel regions in one frame image A generation unit (31) for generating a spectral image of a desired wavelength based on a pixel value corresponding to the pixel region and a pixel value corresponding to the one pixel region in another frame image different from the one frame image; Equipped with

Description

  TECHNICAL FIELD The present invention relates to the technical field of a spectrometry device and a spectrometry method.

  As this type of device, for example, a spectral image obtained by imaging light transmitted through an optical filter of an etalon type is converted to an optical filter based on information on spectral characteristics corresponding to the wavelength of light transmitted through the optical filter. An apparatus for correcting each wavelength of light to be transmitted has been proposed (see Patent Document 1).

JP 2008-128675 A

  For example, a global shutter system and a rolling shutter system are known as a system of the electronic shutter of the imaging device. In the rolling shutter method, an image is acquired by dividing an imaging element into blocks consisting of one or more lines and sequentially exposing each block. When a rolling shutter system is used as an exposure method for spectroscopic measurement, the imaging time is longer than that of the global shutter system.

  In the technology described in Patent Document 1, a CCD (Charge-Coupled Device) mainly using a global shutter system is used as an imaging element (see paragraph 0015). That is, in the technology described in Patent Document 1, the rolling shutter system is not considered.

  The present invention has been made in view of, for example, the above-mentioned facts, and for example, it is possible to shorten the imaging time in spectroscopic measurement using a method in which the exposure timing is different for each pixel area of the imaging device, such as a rolling shutter method. An object of the present invention is to provide a measuring device and method.

  In order to solve the above problems, the spectrometry device of the present invention continuously changes the wavelength of the extracted light by the filter capable of changing the wavelength of the extracted light, the imaging unit having a plurality of regions, and the filter. And a control unit that sequentially exposes the plurality of regions with the extracted light and outputs an output signal serving as a basis of the frame image from each of the plurality of regions, and one of the plurality of regions in one frame image And a generation unit configured to generate a spectral image of a desired wavelength based on the pixel value corresponding to and the pixel value corresponding to the one area in another frame image different from the one frame image.

  A spectrometry method of the present invention is a spectrometry method in a spectrometry device including a filter capable of changing the wavelength of extracted light, and an imaging unit having a plurality of regions, in order to solve the above-mentioned problems, A control step of sequentially exposing the plurality of regions with the extraction light while continuously changing the wavelength of the extraction light with a filter, and outputting an output signal as a basis of the frame image from each of the plurality of regions; A spectrum of a desired wavelength based on a pixel value corresponding to one of the plurality of regions in the frame image and a pixel value corresponding to the one region in another frame image different from the one frame image Generating an image.

  The operation and other advantages of the present invention will be apparent from the embodiments to be described below.

It is a block diagram showing composition of a spectrometry device concerning an example. It is a figure which shows an example of the relationship between the exposure timing of the CMOS sensor which concerns on an Example, and the transmission wavelength of a Fabry-Perot filter. It is a figure which shows an example of a frame image. It is a figure which shows the acquisition concept of the frame image by an ideal optical filter and an actual Fabry-Perot filter each. It is a figure which shows an example of the pixel value of the several flame | frame image acquired by changing the transmission wavelength of a filter for every flame | frame about an ideal optical filter and an actual Fabry-Perot filter, respectively. It is a flowchart which shows the operation processing of the spectrometry apparatus which concerns on an Example. It is a flow chart which shows spectrum image generation processing concerning an example.

  An embodiment according to a spectrometer and method of the present invention will be described.

(Spectrometry device)
The spectrometer according to the embodiment includes a filter capable of changing the wavelength of the extracted light, an imaging unit having a plurality of pixel regions, and a plurality of extracted light while continuously changing the wavelength of the extracted light by the filter. A control unit that sequentially exposes a pixel area and outputs an output signal that is a basis of a frame image from each of a plurality of areas; a pixel value corresponding to one pixel area among a plurality of pixel areas in one frame image; And a generator configured to generate a spectral image of a desired wavelength based on pixel values corresponding to the one pixel area in another frame image different from the frame image of.

  In the spectrometer, at the time of generation of one frame image (that is, at the time of imaging of one frame image), the wavelength of the extracted light for exposing the imaging unit is continuously changed. Therefore, the exposure wavelength is different for each pixel area. Therefore, the generated frame image does not meet the requirements as a spectral image as it is.

  Therefore, in the spectrophotometer, a pixel value (for example, obtained when a certain pixel area is exposed with extraction light of a desired wavelength using two frame images (that is, one frame image and another frame image)) The luminance value etc.) is estimated. Then, one spectral image is generated from the pixel values estimated for each pixel area.

  Therefore, after all the pixel areas of the imaging unit are sequentially exposed by the extraction light of a certain wavelength to generate a spectral image, the spectrometry apparatus changes the wavelength of the extraction light, and the imaging light is changed by the extracted light. As compared with the case where all the pixel regions are sequentially exposed to generate another spectral image, spectral images of a plurality of wavelengths can be generated while shortening the imaging time.

  Preferably, the control unit controls the filter such that the wavelength change amount of the extracted light per time required to capture one frame image is within the full width at half maximum of the wavelength-extraction rate characteristic of the filter.

  An example of the filter is a Fabry-Perot filter provided with an actuator.

  The pixel value corresponding to one pixel area in one frame image is one pixel area when the one pixel area is exposed by the extraction light of the first wavelength whose deviation from the desired wavelength is less than a predetermined value. It is a pixel value based on an output signal. The pixel value corresponding to one pixel area in another frame image is exposed by the extracted light of the second wavelength whose deviation from the desired wavelength is less than the predetermined value and which is different from the first wavelength. It is a pixel value based on the output signal of one pixel area at the same time. Here, the other frame image may be a frame temporally adjacent to one frame image.

  The generation unit is a shift from the reference timing of the timing at which one pixel region is exposed by the extraction light of the first wavelength, with the timing that at least a part of the imaging unit is exposed by the extraction light of the desired wavelength as a reference timing; A pixel value corresponding to one pixel region in one frame image and one pixel in another frame image based on the deviation from the reference timing of the timing when one pixel region is exposed by the extracted light of the second wavelength Each pixel value corresponding to the region may be weighted.

  Alternatively, the generation unit may generate a pixel value corresponding to one pixel area in one frame image and another frame image based on the deviation of the first wavelength from the desired wavelength and the deviation of the second wavelength from the desired wavelength. Each pixel value corresponding to one pixel area in may be weighted.

  With such a configuration, it is possible to improve the estimation accuracy of the pixel value obtained when a certain pixel area is exposed to the extraction light of the desired wavelength, which is extremely advantageous in practical use.

(Spectrometry method)
The spectrometry method which concerns on embodiment is a spectrometry method in a spectrometry apparatus provided with the filter which can change the wavelength of extraction light, and the imaging part which has several pixel area | region. The said spectroscopy measurement method is a control process which exposes several pixel area sequentially by extraction light, and changes the wavelength of extraction light with a filter, and outputs the output signal used as the basis of a frame image from each of the said several area, A desired wavelength is obtained based on a pixel value corresponding to one pixel region among a plurality of pixel regions in one frame image and a pixel value corresponding to one pixel region in another frame image different from one frame image. Generating a spectral image.

  According to the spectrometry method according to the embodiment, as in the spectrometry device according to the above-described embodiment, it is possible to generate spectral images of a plurality of wavelengths while shortening the imaging time. In addition, also in the spectrometry method which concerns on embodiment, the various aspects similar to the various aspects of the spectrometry apparatus which concerns on embodiment mentioned above can be taken.

  An embodiment according to the spectrometer of the present invention will be described.

(Constitution)
The structure of the spectrometry apparatus which concerns on an Example is demonstrated with reference to FIG. FIG. 1 is a block diagram showing the configuration of a spectrometry apparatus according to an embodiment.

  In FIG. 1, the spectrometry apparatus 1 is configured to include a spectroscopic camera 10, and a host computer 30 having a processor 31 and a memory 32.

  The spectral camera 10 includes a lens 11, a Fabry-Perot filter 12, a drive circuit 20, a rolling shutter type CMOS (Complementary Metal-Oxide-Semiconductor) sensor 13, and an image sensor controller 14.

  The Fabry-Perot filter 12 changes the wavelength of the transmitted light according to the gap. In other words, the Fabry-Perot filter 12 selectively transmits light of a wavelength corresponding to the gap. Strictly speaking, although there is a wavelength width in the transmitted light of the Fabry-Perot filter 12, for the sake of simplicity, the central wavelength of the transmitted wavelength width is referred to as a transmitted wavelength.

  The Fabry-Perot filter 12 has an actuator 12 a. The actuator 12 a is configured to be able to change the gap of the Fabry-Perot filter 12 in accordance with the drive voltage output from the drive circuit 20. The image sensor controller 14 generates a frame image from the output signal of the CMOS sensor 13 and outputs the frame image to the host computer 30.

  In the present embodiment, although the transmission type Fabry-Perot filter is used as the light wavelength variable selection filter, the present invention is not limited thereto, and various known transmission or reflection filters can be adopted.

(Imaging operation)
1. Problems Generally the rolling shutter system, if the CCD is used as an imaging device, a method of the electronic shutter is global shutter system. In the global shutter method, all pixels of the imaging device are exposed simultaneously. In recent years, a CMOS sensor has often been used as an imaging device in place of a CCD in terms of downsizing and power consumption.

  On the other hand, in a CMOS sensor, in general, signals from one or more pixels connected to each signal line are sequentially read out for each signal line (horizontal signal line or vertical signal line) due to its structure. Hereinafter, "one or more pixels connected to one signal line" will be referred to as "line" as appropriate. For this reason, when a CMOS sensor is used as an imaging device, a rolling shutter system in which exposure is performed line by line is used.

  In the rolling shutter system, the exposure timing and the signal readout timing are different for each line. When capturing a spectral image, the wavelength of light incident on the CMOS sensor can not be changed until exposure of all the lines is completed so that the wavelength when capturing one frame image does not change. Then, in the case of generating spectral images for a plurality of lights having different wavelengths, the imaging time (measurement time) may be relatively long.

2. Operation of Spectroscopic Measurement Device 1 According to the Embodiment In the present embodiment, the CMOS sensor 13 is sequentially exposed line by line while changing the transmission wavelength of the Fabry-Perot filter 12.

  Specifically, the processor 31 of the host computer 30 drives the gap of the Fabry-Perot filter 12 so as to continuously change based on the “drive voltage-transmission wavelength relationship formula” stored in advance in the memory 32. The circuit 20 is controlled. Then, when the gap of the Fabry-Perot filter 12 is changing, the CMOS sensor 13 is sequentially exposed line by line.

Specifically, as shown in FIG. 2, it is assumed that the drive circuit 20 is controlled such that the transmission wavelength of the Fabry-Perot filter 12 changes to the long wavelength side with time. At the time of imaging of “frame 1”, for example, the exposure wavelength of line 1 (strictly speaking, the central wavelength of exposure) is λ ₁ , ₁ and the exposure wavelength of line 2 is λ ₁ , ₂ and the exposure of line c The wavelengths are λ _{1 and c} . As a result, as shown in FIG. 3, frame images with different exposure wavelengths are captured (or generated) for each line. Note that "frame 1" and the like in FIGS. 2 and 3 indicate frame image numbers, and "line 1" and the like indicate line numbers.

  The frame image obtained in this way does not meet the requirements as a spectral image as it is. In this embodiment, a spectral image of a desired wavelength is generated by performing spectral image generation processing described later based on the obtained frame image. That is, in this embodiment, the imaging time is shortened by generating the spectral image by the spectral image generation process.

(Spectroscopic image generation)
The spectral image generation according to the present embodiment will be described. Here, it is assumed that the object of imaging does not change in state during imaging of a spectral image. For example, it is assumed that the exposure wavelengths λ _{2 and c} of the line c (see the shaded portion in FIG. 3) of the frame 2 in FIG. 3 are the desired wavelengths of the spectral image. The lines other than the line c of the frame 2 are exposed to light of a wavelength different from the desired wavelengths λ _{2 and c} . Therefore, at the time of generation of the spectral image, pixel values of each line other than the line c are estimated as follows.

For example, the pixel value of line 1 of frame 2 when exposed to light of wavelength λ _{2, c} (strictly speaking, “the pixel value of each pixel included in line 1”, and so on) is the line of frame 2 It is estimated from the weighted average based on the pixel value of 1 and the pixel value of line 1 of frame 3. The pixel value of line 1 when exposed to light of wavelength λ _{2, c} is wavelength λ _2, if the wavelength difference between wavelength λ ₂ , ₁ and wavelength λ ₃ , ₁ is smaller than the transmission wavelength width of the filter _{. This} is because it is estimated to be a value between the pixel value of line 1 of frame 2 exposed with ₁ light and the pixel value of line 1 of frame 3 exposed with light of wavelength λ _3,1 .

Similarly, the pixel values of line M of frame 2 when exposed with light of wavelength λ _{2, c} are weighted averages based on the pixel values of line M of frame 2 and the pixel M of line M of frame 1 Estimated from The pixel value of line M of frame 2 when exposed with light of wavelength λ _{2, c} is a wavelength if the wavelength difference between wavelength λ _{2, M} and wavelength λ _{1, M} is smaller than the transmission wavelength width of the filter It is estimated to be a value between the pixel value of line M of frame 2 exposed with light of λ _{2, M} and the pixel value of line M of frame 1 exposed with light of wavelengths λ _{1, M} It is.

  Thus, in the present embodiment, pixel values in the case of exposure to light of a desired wavelength are estimated by interpolation (ie, interpolation). In order to estimate the pixel value of a line whose line number is smaller than the line exposed at the desired wavelength (here, line c) to interpolate, a frame image including the line exposed at the desired wavelength (here, frame 2) And the next frame image (here, frame 3) temporally adjacent to the frame image. On the other hand, to estimate the pixel value of a line whose line number is larger than the line exposed at the desired wavelength, the frame image including the line exposed at the desired wavelength and the previous frame image temporally adjacent to the frame image (Here, frame 1) is used. For this reason, at least one preceding and following one-frame image of a frame image including a line exposed at a desired wavelength is captured.

  The processor 31 of the host computer 30 generates a spectral image of the desired wavelength based on the estimated pixel value.

The “frame 2”, the “frame 3”, the “wavelength λ _2,1 ” and the “wavelength λ _3,1 ” according to the embodiment are the “one frame image” and the “other frame” according to the present invention, respectively. It is an example of an image "," 1st wavelength ", and" 2nd wavelength ". The “frame 1”, the “wavelength λ _{2, M} ” and the “wavelength λ _{1, M} ” according to the embodiment are the “other frame image”, the “first wavelength” and the “second wavelength” according to the present invention, respectively. Another example of

(Reason why we can estimate pixel value)
Here, the reason why the pixel value in the case of exposure with light of a desired wavelength can be estimated by interpolation will be described with reference to FIGS. 4 and 5.

  As shown at the top of FIG. 4, the ideal optical filter transmits only a specific wavelength and does not transmit any other wavelength. On the other hand, in the actual Fabry-Perot filter, the transmission wavelength has a width, and the central wavelength of the transmission wavelength is called the transmission central wavelength.

  In FIG. 4, it is assumed that the spectrum of the subject has a sharp peak only at a specific wavelength such as a laser (see the middle part of FIG. 4) so that differences in filters can be easily understood.

The pixel value of the frame image acquired without changing the transmission wavelength of the filter is simply the product of the transmission of the filter and the spectrum of the object and integrated (see the bottom of FIG. 4). (Strictly, pixel quantization efficiency, exposure time, amplifier gain, etc. are also multiplied, but ignored here)
Now, as shown in the top row of FIG. 5, when a frame image is acquired while changing the transmission wavelength of the filter in the spectrometer, the horizontal axis is plotted as a frame number, and the vertical axis is plotted as a pixel value of that frame, When the frame number on the horizontal axis is replaced with the transmission wavelength (or transmission center wavelength) of the corresponding filter, the spectrum of the object obtained by the spectrometer is obtained. That is, it becomes so-called "convolution" of the filter transmittance spectrum and the subject spectrum.

  In an ideal optical filter, the spectrum of the object obtained by the spectrometer (see the bottom of FIG. 5) matches the spectrum of the object (see the middle of FIG. 5). On the other hand, in an actual Fabry-Perot filter, an abrupt change within a range narrower than the transmission wavelength width of the filter is "flattened" and the spectrum of the object obtained by the spectrometer is as shown on the lower right of FIG.

  Therefore, in practice, the pixel value of the frame image does not change abruptly (in other words, discontinuously) with respect to the wavelength change of about the transmission wavelength width of the Fabry-Perot filter 12. For this reason, the pixel value in the case of exposure with light of a desired wavelength can be estimated from the pixel value in the case of exposure with light of a wavelength near the desired wavelength.

  Incidentally, even if the transmission wavelength of the Fabry-Perot filter 12 continuously changes during imaging of one frame image as in the present embodiment, if the amount of change is about the transmission wavelength width of the Fabry-Perot filter 12, one frame Almost the same pixel values are obtained as in the case where the transmission wavelength does not change during imaging of the image. This is because the transmittance spectrum of the Fabry-Perot filter 12 is substantially symmetrical with respect to the transmission center wavelength.

  The processor 31 of the host computer 30 of the spectrometry apparatus 1 preferably makes the change amount of the transmission wavelength of the Fabry-Perot filter 12 during imaging of one frame image equal to or less than the transmission wavelength width of the Fabry-Perot filter 12. The gap of the Fabry-Perot filter 12 is controlled via the drive circuit 20 so as to be within the full width at half maximum of the transmission wavelength. Here, "the transmission wavelength width of the Fabry-Perot filter 12" is an example of the "wavelength-extraction rate characteristic of the filter" according to the present invention.

(Weighted average)
As the weighted average, (1) weighted average by line number and (2) weighted average by wavelength may be mentioned as an example.

(1) Weighted Average by Line Number In the spectroscopy measuring apparatus 1 as described above, since the rolling shutter method is used, the line number can be reworded as exposure timing. In addition, when the spectrometer 1 is in operation, the wavelength of the transmitted light of the Fabry-Perot filter 12 changes at a constant rate (ie, changes linearly) as shown in the lower part of FIG. Thus, the line number (and the exposure timing) and the wavelength of the transmitted light are closely related.

  It is assumed that the CMOS sensor 13 has M lines in total. The numbers given to the lines are 1 to M (see FIG. 2). Let m (1.ltoreq.m.ltoreq.M) be a number given to a line exposed with light of a desired wavelength, and y (1.ltoreq.y.ltoreq.M) be a number given to a line whose pixel value is to be estimated. A number given to a frame image including pixel values of a line exposed to light of a desired wavelength is i.

For y <m, a pixel value of the line y of the frame i _{I i, y,} a pixel value of the frame i + 1 of the line y as _{I i + 1, y,} the pixel value _I'i line y to be _{estimated, y} is It can be expressed as I'i _{, y} = (M + y-m) / M * _{Ii, n} + (m-y) / M * _{Ii + 1, y} .

For y> m, the pixel values of the frame i-1 of line y as _{I i-1, y,} the pixel value _I'i line y to be _{estimated, y is, I'i, y = (M} + m-y ) / M × I _{i, n} + (y−m) / M × I _{i−1, y}

(2) Weighted Average by Wavelength A number given to a frame image including pixel values of a line exposed with light of a desired wavelength is set to i. Let m be a number given to a line exposed to light of a desired wavelength, and λ _{i, m} be a desired wavelength. Let y be a number assigned to a line whose pixel value is to be estimated.

If y <m, the exposure wavelength of line y of frame i is λ _{i, y} , the pixel value is I _{i, y} , the exposure wavelength of line y of frame i + 1 is λ _{i + 1, y} , and the pixel value is I _{i + 1, y} The pixel value I ' _{i, y} of the line y to be estimated is I'i _{, y} = Ii _{, y} + ([lambda] _{i, m- [} lambda _{] i, y} ) / ([lambda] _{i + 1, y- [} lambda _{] i, y} ) It can be expressed as x ( _{Ii + 1, y-} Ii _{, y} ).

In the case of y> m, the pixel value I ′ _{i, y} of the line y estimated for the line y of the frame i−1 is λ _{i−1, y} and the pixel value is I _{i−1, y} I'i _{, y} = Ii _{, y} + (? _{I, y-} ? _{I, m} ) / (? _{I, y-} ? _{I-1, y} ) x ( _{Ii-1, y-} Ii _{, y} ) It can be expressed.

  Although the above-mentioned equation is a so-called equation of linear interpolation, the present invention is not limited to linear interpolation, and various existing methods can be applied to the estimation of pixel values.

(Operation of spectrometer)
Next, the operation process of the spectrometry apparatus 1 will be described with reference to the flowcharts of FIGS. 6 and 7.

  In FIG. 6, when a wavelength range (that is, a wavelength range of spectral image data that the user desires to obtain) is first specified via a user interface (not shown) (step S101), the processor 31 of the host computer 30 Is specified based on the exposure timing difference of the pixel area (ie line) of the CMOS sensor 13 and the amount of change per hour of the transmission wavelength of the Fabry-Perot filter 12 by the amount necessary for interpolation (ie estimation of pixel value) The corrected wavelength range is corrected (step S102).

Here, start wavelength lambda _stt is, the wavelength lambda _Stt' a wavelength shorter than said start wavelength lambda _stt, end wavelength lambda _{end The} is, the wavelength lambda _End' a longer wavelength than the end wavelength lambda _{end The,} modified Be done. The extent to which the originally designated wavelength range is corrected (or expanded) may be set as appropriate. It is desirable that the wavelength range be modified such that at least one more frame can be captured than the number of frames that can be captured in the originally specified wavelength range.

Next, the processor 31 of the host computer 30 drives such that the transmission wavelength of the Fabry-Perot filter 12 becomes the wavelength λ _stt 'based on the “drive voltage-transmission wavelength relationship” stored in the memory 32. The actuator 12a is controlled via the circuit 20 (step S103).

Subsequently, the processor 31 starts changing the drive voltage output from the drive circuit 20 so that the transmission wavelength of the Fabry-Perot filter 12 changes toward the wavelength λ _end '(step S104). As a result, changes the gap Fabry-Perot filter 12, transmission wavelength, it increased gradually toward the wavelength lambda _End' from the wavelength _λ stt'.

  In parallel with the process of step S104, the processes of steps S105 to S107 are performed. That is, the processor 31 obtains a frame image captured by the CMOS sensor 13 via the image sensor controller 16 (step S105). Subsequently, the processor 31 acquires information on the current transmission wavelength of the Fabry-Perot filter 12 (step S106). The “transmission wavelength information” may be the value of the transmission wavelength itself, or may be, for example, the value of the driving voltage associated with the transmission wavelength. That is, the "transmission wavelength information" means information indicating the transmission wavelength directly or indirectly.

Next, the processor 31 (determines i.e., whether or not to end the image pickup), the current transmission wavelength indicated by the information of the transmission wavelengths, determines whether the equal wavelength lambda _End' (Step S107) . In this determination, when it is determined that the current transmission wavelength is not equal to the wavelength λ _{end '} (step S107: No), the processor 31 performs the process of step S105 again.

On the other hand, if it is determined in step S107 that the current wavelength is equal to the wavelength λ _{end '} (step S107: Yes), the processor 31 stops the change of the drive voltage output from the drive circuit 20 (step S108).

  Next, the processor 31 performs spectral image generation processing with reference to the “spectral image generation program” stored in the memory 32 (step S2). Here, the spectral image generation process will be described with reference to the flowchart of FIG. 7.

  In the spectral image generation process, among the plurality of temporally consecutive frame images acquired in the process of step S105, frame images having numbers 1 to Nfrm assigned to the frame images are to be processed. . The number of lines of the CMOS sensor 13 is “Height”, and the number of pixels included in one line is “Width”. As a result, one frame image includes “Width × Height” pixels.

  In the spectral image generation processing, the desired wavelength is assumed to be the exposure wavelength when the pixel value of the line m (1 ≦ m ≦ Height) of each frame image is acquired. The line m may be set in advance by the manufacturer as a fixed value, or may be set by the user. In the spectral image generation process, the variable relating to the frame image is “i” (1 ≦ i ≦ Nfrm), the variable relating to the line is “y” (1 ≦ y ≦ Height), and the variable relating to the pixel included in one line is “ It is assumed that x ′ ′ (1 ≦ x ≦ Width).

  In FIG. 7, the processor 31 sets a variable i to "1" which is an initial value (step S201). The processor 31 sets the variable y to "1" which is an initial value (step S202). Next, the processor 31 calculates a weight W (y) according to the current line y (step S203). For the specific weight W (y), refer to the “weighted average” described above. However, constant "M" in "weighted average" shall be read as "Height".

  Next, the processor 31 sets the variable x to an initial value "1" (step S204). Next, the processor 31 determines whether the value of the variable y is smaller than m (step S205).

When it is determined in step S205 that the value of the variable y is smaller than m (step S205: Yes), the processor 31 determines the pixel value of the pixel x of the line y of the frame i as the wavelength λ _{i of the} frame _i. the pixel value of the pixel x of the line y, which is exposed to light of _y, replaced with the weighted average value of the pixel value of the frame i + 1 of the wavelength lambda _{i + 1, y} of a pixel x in the line y, which is exposed to light (step S206 ).

  On the other hand, when it is determined in step S205 that the value of the variable y is not smaller than m (step S205: No), the processor 31 determines whether the value of the variable y is larger than m (step S207). ).

When it is determined in step S207 that the value of the variable y is larger than m (step S207: Yes), the processor 31 determines the pixel value of the pixel x of the line y of the frame i as the wavelength λ _{i of the} frame _i. the pixel value of the pixel x of the line y, which is exposed to light of _y, the weighted average of the pixel values of the frame i-1 of the wavelength lambda _{i-1, y} pixel x of the line y, which is exposed to light Replace (step S208).

  If it is determined in step S207 that the value of the variable y is not larger than m (ie, the value of the variable y is equal to m) (step S207: No), or the process of step S206 or step S208 After the processing, the processor 31 determines whether the value of the variable x is equal to the Width (Step S209). That is, in step S209, it is determined whether the replacement of the pixel value is completed for all the pixels included in the line y.

  If it is determined in step S209 that the value of the variable x is not equal to the width (step S209: No), the processor 31 sets the value of the variable x to x + 1 (that is, increments) (step S212), The process of step S205 is performed.

  On the other hand, if it is determined in step S209 that the value of the variable x is equal to the width (step S209: Yes), the processor 31 determines whether the value of the variable y is equal to the height (step s210). . That is, in step S210, it is determined whether the replacement of the pixel values is completed for all the lines included in the frame i.

  If it is determined in step S210 that the value of the variable y is not equal to the height (step S210: No), the processor 31 sets the value of the variable y to y + 1 (that is, increments) (step S213), The process of step S203 is performed.

  On the other hand, when it is determined in step S210 that the value of the variable y is equal to the height (step S210: Yes), the processor 31 determines whether the value of the variable i is equal to Nfrm (step S211). . That is, in step S211, it is determined whether or not replacement of pixel values is completed for all frame images to be processed.

  When it is determined in step S211 that the value of the variable i is not equal to Nfrm (step S211: No), the processor 31 sets the value of the variable i to i + 1 (that is, increments) (step S214), The process of step S202 is performed. On the other hand, when it is determined in step S211 that the value of the variable i is equal to Nfrm (step S211: Yes), the processor 31 performs the process of step S109 in the flowchart of FIG.

  Returning to FIG. 6 again, after the spectral image generation processing, the processor 31 stores the spectral image generated in the spectral image generation processing (that is, the frame image in which the pixel value has been replaced), for example, a storage device such as the memory 32. , And the process ends (step S109).

(Technical effect)
In the said spectroscopy measuring apparatus 1, each pixel area (namely, each line) of CMOS sensor 13 is exposed one by one with transmitted light, changing the wavelength of the transmitted light of Fabry-Perot filter 12. FIG. For this reason, after all the pixel areas of the CMOS sensor 13 are exposed by the transmitted light of a certain wavelength, the spectrometer 1 changes the wavelength of the transmitted light by changing the gap of the Fabry-Perot filter 12. The imaging time can be shortened as compared to the case where all pixel areas of the CMOS sensor 13 are exposed by the later transmitted light.

  In the spectrometer 1, two frame images adjacent in time are used to estimate the pixel value. According to this configuration, estimation accuracy can be improved as compared with the case where two frame images not adjacent in time are used for estimation of the pixel value. Note that three or more frame images may be used to estimate the pixel value.

  The “Fabry-Perot filter 12”, the “CMOS sensor 13” and the “processor 31” according to the embodiment are examples of the “filter”, the “imaging unit”, and the “generation unit” according to the present invention, respectively. The sensor controller 14 ′ ′ and the “drive circuit 20” are examples of the “control unit” according to the present invention.

  The present invention is not limited to the above-described embodiment, and can be appropriately modified without departing from the scope and spirit of the invention which can be read from the claims and the specification as a whole. Moreover, it is contained in the technical scope of this invention.

  DESCRIPTION OF SYMBOLS 1 ... Spectroscopic measuring apparatus, 10 ... Spectroscopic camera, 11 ... Lens, 12 ... Fabry-Perot filter, 12a ... Actuator, 13 ... CMOS sensor, 14 ... Image sensor controller, 20 ... Drive circuit, 30 ... Host computer, 31 ... Processor, 32: Memory

Claims (9)

A filter capable of changing the wavelength of extracted light;
An imaging unit having a plurality of areas;
A controller for sequentially exposing the plurality of regions with the extraction light while continuously changing the wavelength of the extraction light by the filter, and outputting an output signal as a basis of a frame image from each of the plurality of regions;
A desired wavelength is obtained based on a pixel value corresponding to one of the plurality of regions in one frame image and a pixel value corresponding to the one region in another frame image different from the one frame image. A generation unit that generates a spectral image;
And a spectrometer. The pixel value corresponding to the one area in the one frame image is the one area when the one area is exposed by the extraction light of the first wavelength whose deviation from the desired wavelength is a predetermined value or less. Pixel values based on the output signal of
The pixel value corresponding to the one area in the other frame image is the extracted light of the second wavelength whose deviation from the desired wavelength is less than the predetermined value and which is different from the first wavelength in the one area It is a pixel value based on the output signal of said 1 area | region, when it exposes by this. The spectrometry apparatus of Claim 1 characterized by the above-mentioned.   3. The apparatus according to claim 2, wherein the other frame image is a frame temporally adjacent to the one frame image.   The generation unit is configured to use the timing at which at least a part of the imaging unit is exposed by the extraction light of the desired wavelength as a reference timing, and the reference timing at which the one region is exposed by the extraction light of the first wavelength. And a pixel value corresponding to the one area in the one frame image based on the deviation from the reference timing of the timing at which the one area is exposed by the extracted light of the second wavelength and the reference timing. The apparatus according to claim 2, wherein each of the pixel values corresponding to the one area in the other frame image is weighted.   The generation unit is configured to calculate a pixel value corresponding to the one area in the one frame image based on the deviation of the first wavelength from the desired wavelength and the deviation of the second wavelength from the desired wavelength. The apparatus according to claim 2, wherein each of the pixel values corresponding to the one area in the other frame image is weighted.   The control unit controls the filter such that a wavelength change amount of the extracted light per time required to capture the one frame image is within a full width at half maximum of a wavelength-extraction rate characteristic of the filter. The spectrometric measurement device according to claim 1.   The spectrometric device according to claim 6, wherein the filter is a Fabry-Perot filter including an actuator.   The spectrometer according to claim 1, wherein the imaging unit is exposed by a rolling shutter method. What is claimed is: 1. A spectrometry method of a spectrometry device, comprising: a filter capable of changing the wavelength of extracted light; and an imaging unit having a plurality of regions,
A control step of sequentially exposing the plurality of regions with the extraction light while continuously changing the wavelength of the extraction light by the filter, and outputting an output signal as a basis of a frame image from each of the plurality of regions;
A desired wavelength is obtained based on a pixel value corresponding to one of the plurality of regions in one frame image and a pixel value corresponding to the one region in another frame image different from the one frame image. Generating a spectral image;
A method of spectrometry comprising:

Images