Optical fiber spectrometer and spectrum detection method

Abstract

The invention provides a fiber optic spectrometer and a spectrum detection method, and belongs to the technical field of spectrum detection. The optical fiber spectrometer comprises an optical splitter and a detector, wherein the optical splitter comprises a multimode optical fiber and a diffraction grating which is formed by writing on the end face of the emergent end of the multimode optical fiber and is arranged opposite to the light sensing surface of the detector. The spectrum detection method comprises the following steps: calibrating a fiber optic spectrometer by using a tunable laser to obtain an initial transmission matrix for representing response characteristics of the fiber optic spectrometer under different wavelengths; correcting the initial transmission matrix by adopting a singular value decomposition algorithm to obtain a final transmission matrix; measuring a to-be-measured light source by using a fiber optic spectrometer to obtain a to-be-measured diffraction image; and reconstructing the spectrum of the to-be-measured light source according to the to-be-measured diffraction image and the final transmission matrix. According to the invention, the diffraction image of the light source to be detected is measured by means of the optical fiber spectrometer with a simple structure, and the spectrum of the light source to be detected can be rapidly and reliably detected in combination with the transmission matrix obtained by calibrating the optical fiber spectrometer.

Description

Technical Field

[0001] The present application relates to the field of spectrum detection technology, and in particular to a fiber optic spectrometer and a spectrum detection method. Background Art

[0002] In the prior art, the spectrum of the light source to be measured is usually detected with the help of a traditional grating spectrometer. However, during this detection process, precise moving parts are needed to control the diffraction grating inside the traditional grating spectrometer to rotate at a uniform speed, resulting in a relatively long detection time. In addition, the mechanical axis of the diffraction grating is prone to wear after a long period of detection, which affects the accuracy of spectral detection. Summary of the invention

[0003] The main purpose of the present application is to propose a fiber optic spectrometer and a spectrum detection method. By measuring and acquiring the diffraction image of the light source to be measured with the help of a fiber optic spectrometer with a simple and compact structure, and combining it with the transmission matrix obtained by calibrating the fiber optic spectrometer, the spectrum of the light source to be measured can be quickly and reliably detected.

[0004] To achieve the above-mentioned object, one aspect of the present application provides a fiber optic spectrometer, the fiber optic spectrometer comprising a beam splitter and a detector, the beam splitter comprising a multimode optical fiber and a diffraction grating, the multimode optical fiber having an incident end and an output end, the diffraction grating being formed by engraving on the end face of the output end of the multimode optical fiber, and the diffraction grating being arranged opposite to a photosensitive surface of the detector;

[0005] When the output end of the external light source is matched and docked with the incident end of the multimode optical fiber, the light beam emitted by the external light source is incident on the diffraction grating via the multimode optical fiber, Fraunhofer diffraction occurs on the diffraction grating and forms an output light including first-order diffraction light, and the output light is incident on the photosensitive surface of the detector, and the detector collects the light intensity information of the output light to generate a diffraction image.

[0006] Furthermore, the core diameter of the multimode optical fiber ranges from 30 microns to 100 microns.

[0007] Furthermore, the period of the diffraction grating ranges from 1 micron to 3 microns.

[0008] Furthermore, the distance between the diffraction grating and the photosensitive surface of the detector ranges from 200 micrometers to 500 micrometers.

[0009] Furthermore, the detector is a CCD camera or a CMOS camera.

[0010] To achieve the above object, another aspect of the present application provides a spectrum detection method, which is applied to the above-mentioned fiber optic spectrometer, and the method comprises:

[0011] Calibrate the fiber optic spectrometer using a tunable laser to obtain an initial transmission matrix, wherein the initial transmission matrix is ​​used to characterize the response characteristics of the fiber optic spectrometer at different wavelengths;

[0012] The initial transmission matrix is ​​corrected by using a singular value decomposition algorithm to obtain a final transmission matrix;

[0013] The optical fiber spectrometer is used to measure the light source to be measured to obtain a diffraction image to be measured;

[0014] The spectrum of the light source to be measured is reconstructed according to the diffraction image to be measured and the final transmission matrix.

[0015] Furthermore, the calibrating the fiber spectrometer using a tunable laser to obtain an initial transmission matrix includes:

[0016] Acquire multiple different reference wavelengths from the wavelength measurement range set by the optical fiber spectrometer;

[0017] For each of the reference wavelengths, the tunable laser is controlled to emit a quasi-monochromatic light beam at the reference wavelength, and then the fiber spectrometer is controlled to measure the quasi-monochromatic light beam to obtain a calibration diffraction image, and the intensity value of the quasi-monochromatic light beam is measured;

[0018] For the multiple quasi-monochromatic light beams emitted by the tunable laser at the multiple reference wavelengths, respectively, the initial transmission matrix is ​​determined according to the multiple calibrated diffraction images and multiple intensity values ​​corresponding to the multiple quasi-monochromatic light beams obtained by measurement.

[0019] Further, determining the initial transmission matrix according to the measured multiple calibration diffraction images and multiple intensity values ​​corresponding to the multiple quasi-monochromatic light beams includes:

[0020] For each calibration diffraction image and intensity value corresponding to the quasi-monochromatic light beam, the calibration diffraction image corresponding to the quasi-monochromatic light beam is divided into a plurality of first image regions, a pixel grayscale value of each first image region is obtained, and then the grayscale value is calculated in combination with the intensity value corresponding to the quasi-monochromatic light beam to generate an initial transmission vector corresponding to the quasi-monochromatic light beam;

[0021] The initial transmission matrix is ​​generated according to the multiple initial transmission vectors corresponding to the multiple quasi-monochromatic light beams obtained by calculation.

[0022] Further, the adopting a singular value decomposition algorithm to correct the initial transmission matrix to obtain a final transmission matrix includes:

[0023] Performing singular value decomposition on the initial transmission matrix to obtain an initial diagonal matrix, an initial left unitary matrix, and an initial right unitary matrix;

[0024] The number of non-zero singular values ​​contained in the initial diagonal matrix is ​​reduced to generate a final diagonal matrix; according to the column positions of the non-zero singular values ​​contained in the final diagonal matrix in the initial diagonal matrix, column data matching the column positions are extracted from the initial left unitary matrix to generate a final left unitary matrix;

[0025] According to the row positions of the non-zero singular values ​​contained in the final diagonal matrix in the initial diagonal matrix, extracting row data matching the row positions from the initial right unitary matrix to generate a final right unitary matrix;

[0026] The final transmission matrix is ​​generated according to the final diagonal matrix, the final left unitary matrix and the final right unitary matrix.

[0027] Further, reconstructing the spectrum of the light source to be measured according to the diffraction image to be measured and the final transmission matrix includes:

[0028] Calculating a pseudo-inverse matrix of the final transmission matrix;

[0029] The diffraction image to be measured is divided into a plurality of second image regions, the pixel grayscale value of each second image region is obtained, and then the grayscale value is calculated in combination with the pseudo-inverse matrix to generate a spectrum matrix of the light source to be measured.

[0030] The present application includes at least the following beneficial effects: by setting a detector and a multimode optical fiber with a diffraction grating engraved on the end face inside the optical fiber spectrometer, the diffraction image of the light source to be measured can be measured and obtained, without configuring any moving parts, the overall structure is simple and compact, no wear will be caused to any parts during the measurement process, and the measurement accuracy can be ensured. After obtaining the initial transmission matrix obtained by calibrating the optical fiber spectrometer and further correcting it based on the singular value decomposition algorithm, combined with the diffraction image of the light source to be measured, the spectrum of the light source to be measured can be quickly and reliably detected. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a schematic diagram of the structure of a fiber optic spectrometer provided in an embodiment of the present application;

[0032] Figure 2 It is a schematic diagram of the structural composition of a preparation system for a spectrometer provided in an embodiment of the present application;

[0033] Figure 3 It is a flow chart of a spectral detection method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are only examples of methods consistent with some aspects of the embodiments of the present application as detailed in the attached claims.

[0035] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0036] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0037] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0038] In the prior art, the spectrum of the light source to be measured is usually detected by means of a traditional grating spectrometer. Specifically, the traditional grating spectrometer is internally provided with a slit, a collimator, a diffraction grating, a focusing lens and a CCD (Charge-Coupled Device) camera. The diffraction grating generally adopts a blazed grating. The light emitted by the light source to be measured passes through the slit and is incident on the collimator. The collimated light is incident on the surface of the diffraction grating. Since the angles of light of different wavelengths leaving the diffraction grating are different, the focusing lens focuses the monochromatic light of a certain wavelength on the CCD camera. During the spectrum detection process, the diffraction grating will rotate at a uniform speed under the action of a stepping motor, so that the CCD camera can receive diffracted light of different wavelengths. By reading the angle of the diffraction grating and the light intensity signal detected by the CCD camera, the spectrum information of the light source to be measured can be determined. Since the spectral detection process relies on precise moving parts to control the movement of the diffraction grating, the detection time is relatively long, and the mechanical axis of the diffraction grating is prone to wear after a long period of detection, affecting the accuracy of spectral detection. In addition, the traditional grating spectrometer itself has at least the following defects: the preparation process of the diffraction grating is relatively complicated, resulting in high cost; the detection resolution of the traditional grating spectrometer is proportional to the internal optical path length of the device, which is not conducive to the miniaturization of the instrument; the instrument has poor shock resistance and is not suitable for use as an online instrument.

[0039] In view of this, the embodiments of the present application provide a fiber optic spectrometer and a spectrum detection method. By arranging a detector and a multimode optical fiber with a diffraction grating engraved on the end face inside the fiber optic spectrometer, the diffraction image of the light source to be measured can be measured and obtained. There is no need to configure any moving parts. The overall structure is simple and compact. No parts will be worn during the measurement process, and the measurement accuracy can be ensured. After obtaining the initial transmission matrix obtained by calibrating the fiber optic spectrometer and further correcting it based on the singular value decomposition algorithm, combined with the diffraction image of the light source to be measured, the spectrum of the light source to be measured can be quickly and reliably detected.

[0040] Figure 1 1 is a schematic diagram of the structure of a fiber optic spectrometer provided in an embodiment of the present application. The fiber optic spectrometer includes a detector 101 and a spectrometer. The spectrometer includes a multimode optical fiber 102 and a diffraction grating 103. The multimode optical fiber 102 has an incident end and an output end. The diffraction grating 103 is formed by engraving on the end face of the output end of the multimode optical fiber 102, and the diffraction grating 103 is arranged opposite to the photosensitive surface of the detector 101.

[0041] Among them, the core diameter range of the multimode optical fiber 102 is 30 microns to 100 microns, the cladding diameter of the multimode optical fiber 102 is 125 microns, the period range of the diffraction grating 103 is 1 micron to 3 microns, the distance range between the diffraction grating 103 and the photosensitive surface of the detector 101 is 200 microns to 500 microns, and the detector 101 is a CMOS (Complementary Metal Oxide Semiconductor) camera or a CCD camera; preferably, the core diameter range of the multimode optical fiber 102 is 50 microns to 62.5 microns, and the period of the diffraction grating 103 is 2 microns.

[0042] In a specific implementation process, when the output end of the external light source is matched and docked with the incident end of the multimode optical fiber 102, the light beam emitted by the external light source is incident on the diffraction grating 103 through the multimode optical fiber 102, Fraunhofer diffraction occurs on the diffraction grating 103 and forms an output light including first-order diffraction light, and the output light is incident on the photosensitive surface of the detector 101, and the detector 101 collects the light intensity information of the output light to generate a diffraction image.

[0043] As an improved implementation, by additionally setting a single-mode optical fiber jumper inside the optical fiber spectrometer, a fiber optic cutter can be used to fuse the incident end of the multi-mode optical fiber 102 with the output end of the single-mode optical fiber jumper, and then the output end of the external light source is matched and docked with the incident end of the single-mode optical fiber jumper, so that the light beam emitted by the external light source enters the spectrometer and the detector 101 in turn through the single-mode optical fiber jumper for spectral detection; by using a single-mode optical fiber jumper to connect the external light source and the multi-mode optical fiber 102, it is beneficial to improve the stability and efficiency of optical signal transmission.

[0044] In some embodiments, the preparation process of the spectrometer is described as follows:

[0045] Build a preparation system for the spectrometer, such as Figure 2 As shown, the preparation system includes a femtosecond laser 201, a half-wave plate 202, a Glan polarizer 203, a beam expander 204, a reflector 205, a 1 / 4 wave plate 206, a beam splitter 207, a microscope objective 208, a dichroic polarizer 209, an illumination light source 210, a focusing lens 211, a CCD camera 212 and a motion platform 213; wherein the microscope objective 208 preferably adopts a 40x objective lens, the motion platform 213 preferably adopts a 5-axis electric translation stage, the beam splitter 207 preferably adopts a dichroic short-pass beam splitter, the illumination light source 210 preferably adopts an LED illumination light source, and the angle between the optical axis of the 1 / 4 wave plate 206 and the optical axis of the Glan polarizer 203 is preferably set to 45 degrees.

[0046] The half-wave plate 202 is located between the femtosecond laser 201 and the Glan polarizer 203. The half-wave plate 202 can control its own rotation angle through a motor to further adjust the polarization direction of the light beam emitted by the femtosecond laser 201 and the optical axis angle of the Glan polarizer 203, thereby adjusting the light intensity of the light beam emitted by the femtosecond laser 201.

[0047] The preparation system is activated, the multimode optical fiber 102 is fixedly placed on the motion platform 213, and the end face of the output end of the multimode optical fiber 102 should be facing the microscope objective 208, the frequency of the femtosecond laser 201 is set to 0.1kHz~10kHz, the pulse energy of the femtosecond laser 201 is set to 1nJ~10nJ, and the motion platform 213 is used to control the femtosecond laser 201 to generate a writing speed of 1μm / s~100μm / s on the end face of the output end of the multimode optical fiber 102.

[0048] The femtosecond laser 201 is controlled to emit a femtosecond laser beam in a linear polarization state and a wavelength of 520nm. The femtosecond laser beam is sequentially adjusted in light intensity through the half-wave plate 202 and the Glan polarizer 203 to form a first laser beam. The first laser beam is expanded through the beam expander 204 to form a second laser beam. The second laser beam is formed into a third laser beam in a circular polarization state after passing through the reflector 205 and the 1 / 4 wave plate 206. Since the beam splitter 207 allows light with a wavelength less than or equal to 567nm to be transmitted but light with a wavelength greater than 567nm is reflected by 90 degrees, most of the energy carried by the third laser beam can act on the end face of the output end of the multimode optical fiber 102 through the beam splitter 207 and the microscope objective 208 in sequence.

[0049] The multimode optical fiber 102 is driven to move periodically by controlling the motion platform 213 , so that the femtosecond laser beam can periodically scan the end face of the output end of the multimode optical fiber 102 , thereby forming the diffraction grating 103 .

[0050] In addition, the illumination light source 210 is arranged on one side of the beam splitter 209, and an illumination light beam with a wavelength of 650nm is emitted by the illumination light source 210. After the illumination light beam passes through the beam splitter 209, most of its energy forms a first 90-degree reflected light through the beam splitter 207. The first 90-degree reflected light is irradiated to the end face of the output end of the multimode optical fiber 102 through the microscope objective 208 to form end face reflected light. Most of the energy of the end face reflected light forms a second 90-degree reflected light through the beam splitter 207. The second 90-degree reflected light is sequentially imaged into the CCD camera 212 through the beam splitter 209 and the focusing lens 211, so that the user can observe the entire preparation process.

[0051] The fiber optic spectrometer proposed in the embodiment of the present application can measure and obtain the diffraction image of the light source to be measured by arranging a detector and a multimode optical fiber with a diffraction grating engraved on the end face inside it. It does not need to be configured with any moving parts. The overall structure is simple and compact. No parts will be worn during the measurement process, and the measurement accuracy can be ensured.

[0052] Figure 3 is an optional flow chart of a spectrum detection method provided in an embodiment of the present application, which needs to be applied to the above-mentioned fiber optic spectrometer. Figure 3 The method may include but is not limited to steps S301 to S304:

[0053] Step S301, calibrating the fiber spectrometer using a tunable laser to obtain an initial transmission matrix;

[0054] It should be noted that the initial transmission matrix mainly characterizes the response characteristics of the fiber spectrometer at different wavelengths, that is, it characterizes the contribution of incident light at a certain point to the light field at a certain detection point after scattering;

[0055] Step S302: Correct the initial transmission matrix using a singular value decomposition algorithm to obtain a final transmission matrix;

[0056] Step S303, using a fiber optic spectrometer to measure the light source to be measured, and obtaining a diffraction image to be measured;

[0057] Step S304: reconstruct the spectrum of the light source to be measured according to the diffraction image to be measured and the final transmission matrix.

[0058] In steps S301 to S304 shown in the embodiment of the present application, by measuring and acquiring the diffraction image of the light source to be measured with the help of a fiber optic spectrometer with a simple and compact structure, and combining it with the transmission matrix obtained by calibrating the fiber optic spectrometer, the spectrum of the light source to be measured can be quickly and reliably detected.

[0059] In the embodiment of the present application, when a light beam emitted by any light source undergoes Fraunhofer diffraction on the diffraction grating inside the fiber spectrometer, the corresponding diffraction condition is described by the following grating equation (1):

[0060] d·sinθ=mλ (1)

[0061] Wherein, d is the period of the diffraction grating, θ is the diffraction angle, m is the order of the diffraction angle, and λ is the wavelength of the light beam emitted by any light source;

[0062] According to the above grating equation (1), when the wavelength λ of the light beam emitted by any light source is greater than the period d of the diffraction grating, the diffraction grating will not diffract; when the wavelength λ of the light beam emitted by any light source is less than or equal to the period d of the diffraction grating, it can be determined that the output light formed by the diffraction grating has first-order diffraction; and when the wavelength λ of the light beam emitted by any light source changes, the diffraction angle θ will change, that is, the angle of the output light spot will change, so that the position and intensity of the light incident on the detector inside the fiber spectrometer will change, thereby combining the diffraction image corresponding to the any light source generated by the detector, and reconstructing the spectrum of the any light source through the transfer matrix algorithm.

[0063] The following description is made for the above-mentioned transfer matrix algorithm: Since the light incident on the detector will be received by all the pixel units provided on the detector, the pixel grayscale value corresponding to each pixel unit can be read through the preset MATLAB code. All the pixel units provided on the detector are arranged in a two-dimensional array. Any pixel unit provided on the detector corresponds to a pixel point on the diffraction image generated by the detector. For example, if 100*100 pixel units are arranged on the detector, then the diffraction image generated by the detector will contain 100*100 pixel points. Here, all the pixel units provided on the detector are divided into regions. Preferably, all the pixel units provided on the detector are divided into rows, and a pixel region is formed by a row of pixel units. The light intensity received by each pixel region is different. The total pixel grayscale value corresponding to each pixel region can be expressed by the following formula (2):

[0064]

[0065] In the formula, C n is the total pixel grayscale value corresponding to the nth pixel area, λ max is the maximum wavelength of the light beam emitted by any light source, λ min is the minimum wavelength of the light beam emitted by any light source, I(λ) is the light intensity of the light beam emitted by any light source, F(λ) represents the transmission function of the multimode optical fiber inside the fiber spectrometer, S(λ) is the spectral sensitivity of the detector, O n (λ) represents the transmission function when the light beam emitted by any light source is transmitted to the nth pixel area on the detector via the multimode optical fiber;

[0066] Define T n (λ)=F(λ)S(λ)O n(λ) is the transmission function corresponding to the nth pixel area. Since the optical fiber spectrometer divides the spectral range of the light beam emitted by any light source into M spectral regions when detecting the light beam, the M spectral regions are preferably set to be M continuous but non-overlapping spectral regions, where each spectral region is regarded as quasi-monochromatic light, the above formula (2) is rewritten as:

[0067]

[0068] In the formula, I i is the intensity value corresponding to the i-th spectral region, T n,i is the transmission coefficient of the i-th spectral region on the n-th pixel region;

[0069] According to the above formula (3), the equivalent formula (4) on which the transfer matrix algorithm depends is determined as follows:

[0070]

[0071] Where N is the number of all pixel areas divided, which means that all pixel units on the detector are currently arranged in N rows. The vector (C) N×1 Record the N total pixel grayscale values ​​corresponding to the N pixel areas, T N×M is the transmission matrix of the fiber optic spectrometer, vector (I) M×1 The M intensity values ​​corresponding to the M spectral regions are recorded, that is, the discrete spectrum of the light beam emitted by any light source.

[0072] In some embodiments, the above step S301 may include but is not limited to steps S401 to S403:

[0073] Step S401: Acquire multiple different reference wavelengths from a wavelength measurement range set by the optical fiber spectrometer.

[0074] In this step, considering that the dimension of the transmission matrix of the fiber optic spectrometer is N×M, N is a fixed parameter value limited by the detector, and M different reference wavelengths are mainly obtained from the wavelength measurement range set by the fiber optic spectrometer, and the wavelength measurement range set by the fiber optic spectrometer should be included in the wavelength adjustable range allowed by the tunable laser; as an optional implementation, the wavelength measurement range set by the fiber optic spectrometer is evenly divided into M non-overlapping wavelength measurement sub-ranges, and then the wavelength falling in the middle position is obtained from each wavelength measurement sub-range as a reference wavelength, or a wavelength is randomly obtained from each wavelength measurement sub-range as a reference wavelength, or M different reference wavelengths are directly obtained by randomly selecting from the wavelength measurement range set by the fiber optic spectrometer, and the present application does not limit this.

[0075] Step S402: for each reference wavelength, control the tunable laser to emit a quasi-monochromatic beam at the reference wavelength, then control the fiber spectrometer to measure the quasi-monochromatic beam to obtain a calibrated diffraction image, and measure the intensity value of the quasi-monochromatic beam.

[0076] In this step, it is preferred that an optical power meter is provided at the output end of the tunable laser, and the optical power meter is used to measure the intensity value of the quasi-monochromatic light beam emitted by the tunable laser.

[0077] Step S403: for the multiple quasi-monochromatic light beams emitted by the tunable laser at multiple reference wavelengths, determine the initial transmission matrix according to the multiple intensity values ​​and the multiple calibrated diffraction images corresponding to the multiple quasi-monochromatic light beams measured in the above step S402.

[0078] In steps S401 to S403 shown in the embodiment of the present application, a tunable laser with adjustable output light wavelength is used to calibrate the fiber spectrometer to obtain a suitable initial transmission matrix, which can lay the foundation for the subsequent use of a transmission matrix algorithm to achieve the spectral reconstruction task of the light source to be measured.

[0079] In some embodiments, the above step S403 may include but is not limited to steps S501 to S502:

[0080] Step S501: for the intensity value and calibration diffraction image corresponding to each quasi-monochromatic light beam emitted by the tunable laser, the calibration diffraction image corresponding to the quasi-monochromatic light beam is divided into a plurality of first image areas, and the pixel grayscale value of each first image area is obtained, and then combined with the intensity value corresponding to the quasi-monochromatic light beam, an operation is performed to generate an initial transmission vector corresponding to the quasi-monochromatic light beam.

[0081] In this step, a pixel grayscale vector corresponding to the quasi-monochromatic light beam is formed by a plurality of pixel grayscale values ​​corresponding to a plurality of first image regions. The pixel grayscale vector corresponding to the quasi-monochromatic light beam is a column vector. The pixel grayscale vector corresponding to the quasi-monochromatic light beam is divided by the intensity value corresponding to the quasi-monochromatic light beam to obtain an initial transmission vector corresponding to the quasi-monochromatic light beam. This implementation method is expressed by the following formula (5):

[0082] (T1) N×i =(C1) N×1 / I ii (5)

[0083] Where, (T1) N×i is the initial transmission vector corresponding to the i-th quasi-monochromatic beam emitted by the tunable laser, (C1) N×1is the pixel grayscale vector corresponding to the i-th quasi-monochromatic light beam, N can also be understood as the number of the first image regions divided from the calibration diffraction image corresponding to the i-th quasi-monochromatic light beam, I 1i is the intensity value corresponding to the i-th quasi-monochromatic light beam.

[0084] Step S502: Generate an initial transmission matrix according to the multiple initial transmission vectors corresponding to the multiple quasi-monochromatic light beams obtained by the calculation in step S501.

[0085] In this step, since the initial transmission vector corresponding to each quasi-monochromatic light beam emitted by the tunable laser is a column vector, the multiple initial transmission vectors corresponding to the multiple quasi-monochromatic light beams emitted by the tunable laser are directly merged to obtain the initial transmission matrix represented by a two-dimensional array: (T1) N×M ={(T1) N×1 ,(T1) N×2 ,...,(T1) N×i ,...,(T1) N×M}.

[0086] In some embodiments, the above step S302 may include but is not limited to steps S601 to S605:

[0087] Step S601: perform singular value decomposition on the initial transmission matrix to obtain an initial left unitary matrix, an initial diagonal matrix and an initial right unitary matrix, which can be expressed by the following formula (6):

[0088]

[0089] Where: (U1) T U1=1,(V1) T V1=1, S1=diag(σ1,σ2,…,σ K );

[0090] Where U1 is the initial left unitary matrix and is an orthogonal matrix, S1 is the initial diagonal matrix, V1 is the initial right unitary matrix and is an orthogonal matrix, T is the transposition symbol, σ k is the kth singular value contained on the diagonal of the initial diagonal matrix S1, k = 1, 2, ..., K, K is the initial transmission matrix T j The rank of K is the minimum value of N and M, and σ1>σ2>…>σ K It should be noted that the main diagonal elements in the initial diagonal matrix S1 contain all the information of the initial transmission matrix T1, and the relative size of each element value represents the characteristic proportion of the original matrix retained.

[0091] Step S602: Reduce the number of non-zero singular values ​​contained in the initial diagonal matrix to generate a final diagonal matrix.

[0092] In this step, the initial diagonal matrix S1 can be checked and judged by human experience to determine the optimal number of singular values ​​allowed to be retained and recorded as N σ The initial diagonal matrix S1 can also be numerically analyzed using a preset spectrum reconstruction analysis program to determine the optimal number of singular values ​​N. σ , the optimal number of singular values ​​N can also be directly given based on previous practical experience before executing the entire spectrum detection task σ To meet the basic spectrum reconstruction requirements, this application does not limit this; then extract the N front rows from the diagonal of the initial diagonal matrix S1 σ non-zero singular values, according to the extracted N σ non-zero singular values ​​to directly generate the final diagonal matrix:

[0093] The following explanation is made for the above numerical analysis method: in the absence of signal noise, the larger the number of singular values ​​used, the closer the reconstructed spectrum is to the measured spectrum. However, any signal system has a certain amount of noise. Considering that the larger the number of singular values ​​used, the smaller the corresponding signal-to-noise ratio is. Here, based on all the non-zero singular values ​​contained in the initial diagonal matrix S1, several schemes to be analyzed are generated in a manner in which the number of non-zero singular values ​​is gradually increased. The first scheme to be analyzed includes a non-zero singular value arranged at the front on the diagonal of the initial diagonal matrix S1, the second scheme to be analyzed includes two non-zero singular values ​​arranged at the front on the diagonal of the initial diagonal matrix S1, and so on. The last scheme to be analyzed includes all the non-zero singular values ​​on the diagonal of the initial diagonal matrix S1. The spectrum reconstruction analysis program is used to analyze all the non-zero singular values ​​contained in each scheme to be analyzed to determine the corresponding signal-to-noise ratio, and then the signal-to-noise ratios corresponding to the several schemes to be analyzed are compared with a preset signal-to-noise ratio threshold, and the schemes to be analyzed whose corresponding signal-to-noise ratios are greater than or equal to the signal-to-noise ratio threshold are screened out, and the number of non-zero singular values ​​contained in the screened schemes to be analyzed is used as the optimal number of singular values ​​N σ It should be noted that when there are multiple solutions to be analyzed whose corresponding signal-to-noise ratio is greater than or equal to the signal-to-noise ratio threshold among the multiple solutions to be analyzed, the solution to be analyzed with the largest number of non-zero singular values ​​is selected from the multiple solutions to be analyzed, and then the number of non-zero singular values ​​contained in the solution to be analyzed obtained by the secondary screening is used as the optimal number of singular values ​​N σ .

[0094] Step S603: According to the column positions of the non-zero singular values ​​contained in the final diagonal matrix in the initial diagonal matrix, column data matching the column positions are extracted from the initial left unitary matrix to generate a final left unitary matrix.

[0095] In this step, since the diagonal of the final diagonal matrix S2 contains N σ non-zero singular values, get N σ The non-zero singular values ​​in the initial diagonal matrix S1 correspond to the N σ column positions, N σ The column positions are different from each other, and then extract the N σ The column positions corresponding to N σ columns of data, and N σ The columns of data are directly combined to form the final left unitary matrix represented by a two-dimensional array

[0096] Step S604: According to the row positions of the non-zero singular values ​​contained in the final diagonal matrix in the initial diagonal matrix, extract row data matching the row positions from the initial right unitary matrix to generate a final right unitary matrix.

[0097] In this step, since the diagonal of the final diagonal matrix S2 contains N σ non-zero singular values, get N σ The non-zero singular values ​​in the initial diagonal matrix S1 correspond to the N σ row positions, N σ The row positions are different from each other, and then extract the N σ The row positions correspond to N σ rows of data, and N σ The rows of data are directly combined to form the final right unitary matrix represented by a two-dimensional array

[0098] Step S605: Generate a final transmission matrix according to the final left unitary matrix, the final diagonal matrix and the final right unitary matrix, which can be expressed by the following formula (7):

[0099]

[0100] Where T2 is the final transmission matrix.

[0101] For example, when the initial transmission matrix is ​​(T1) 7×6 When , perform singular value decomposition on it to obtain the corresponding initial left unitary matrix (U1) 7×7 , initial diagonal matrix (S1) 7×6 and the initial right unitary matrix (V1)6×6 ; The initial diagonal matrix (S1) 7×6 The diagonal line of contains 6 non-zero singular values, denoted as {σ1,σ2,σ3,σ4,σ5,σ6}, set in the initial diagonal matrix (S1) 7×6 The optimal number of singular values ​​allowed to be retained is 3. From this initial diagonal matrix (S1) 7×6 Extract three appropriate non-zero singular values ​​from the diagonal of , denoted as {σ1,σ2,σ3}, and generate the final diagonal matrix (S2) based on the extracted three non-zero singular values 3×3 =diag(σ1,σ2,σ3), and in the final diagonal matrix (S2) 3×3 All elements except the diagonal are zero; since the three non-zero singular values ​​extracted fall in the first, second and third columns of the initial diagonal matrix, the initial left unitary matrix (U1) 7×7 Extract the first column data, the second column data and the third column data and directly combine them to form the final left unitary matrix (U2) 7×3 , which is equivalent to taking the initial left unitary matrix (U1) 7×7 The 4th, 5th, 6th and 7th columns of data are directly deleted to obtain the final left unitary matrix (U2) 7×3 ; Since the three non-zero singular values ​​extracted fall on the first, second and third rows of the initial diagonal matrix, the initial right unitary matrix (V1) 6×6 Extract the first row of data, the second row of data, and the third row of data and directly combine them to form the final right unitary matrix (V2) 3×6 , which is equivalent to taking the initial right unitary matrix (V1) 6×6 The 4th, 5th and 6th rows of data in are directly deleted to obtain the final right unitary matrix (V2) 3×6 ; The final left unitary matrix (U2) 7×3 , the final diagonal matrix (S2) 3×3 and the final right unitary matrix (V2) 3×6 Perform calculations to obtain the final transmission matrix (T2) 7×6 .

[0102] In steps S601 to S605 shown in the embodiment of the present application, considering that there are various noises in the calibration process of the fiber optic spectrometer, such as ambient light noise, vibration during the calibration process, fluctuations in the intensity of the light source, etc., the initial transmission matrix obtained by the calibration method is usually an ill-conditioned matrix. Directly reconstructing the spectrum of the light source to be measured using the ill-conditioned matrix will cause large errors. Here, the initial transmission matrix is ​​corrected by introducing a singular value decomposition algorithm, so that the data in the final transmission matrix obtained by correction retains the original characteristics to a great extent and simplifies the rank reduction at the same time, and can remove the noise and redundant information originally contained in the initial transmission matrix, which is conducive to the subsequent execution of more reliable spectrum reconstruction tasks.

[0103] In some embodiments, the above step S304 may include but is not limited to steps S701 to S702:

[0104] Step S701: Calculate the pseudo-inverse matrix of the final transmission matrix, which can be expressed by the following formula (8):

[0105]

[0106] In the formula, is the pseudo-inverse matrix of the final transmission matrix, is the pseudo-inverse matrix of the final diagonal matrix.

[0107] Step S702, dividing the diffraction image to be measured corresponding to the light source to be measured into a plurality of second image areas, obtaining the pixel grayscale value of each second image area, and then performing calculations in combination with the pseudo-inverse matrix of the final transmission matrix to generate a spectrum matrix of the light source to be measured.

[0108] In this step, a pixel grayscale matrix corresponding to the light source to be measured is formed by a plurality of pixel grayscale values ​​corresponding to a plurality of second image regions. The pixel grayscale matrix corresponding to the light source to be measured is a column matrix. The pseudo-inverse matrix of the final transmission matrix is ​​multiplied by the pixel grayscale matrix corresponding to the light source to be measured to obtain the spectrum matrix of the light source to be measured. This implementation method is expressed by the following formula (9):

[0109]

[0110] In the formula, (I ′ ) M×1 is the spectrum matrix of the light source to be measured, (C ′ ) N×1 is the pixel grayscale matrix corresponding to the light source to be measured, and N can also be understood as the number of a plurality of second image areas divided from the diffraction image to be measured corresponding to the light source to be measured.

[0111] The spectral detection method proposed in the embodiment of the present application can realize fast and reliable detection of the spectrum of the light source to be measured by acquiring the initial transmission matrix obtained by calibrating the optical fiber spectrometer and further correcting it based on the singular value decomposition algorithm, combined with the diffraction image of the light source to be measured.

[0112] Although the description of the present application has been quite detailed and specifically describes several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, but should be regarded as providing a broad possible interpretation of these claims by reference to the attached claims, taking into account the prior art, so as to effectively cover the intended scope of the present application. In addition, the above description of the present application is based on the embodiments foreseeable by the inventor, and its purpose is to provide a useful description, and those non-substantial changes to the present application that have not yet been foreseen may still represent equivalent changes to the present application.

Claims

1. A fiber optic spectrometer, characterized in that: The optical fiber spectrometer includes a beam splitter and a detector. The beam splitter includes a multimode optical fiber and a diffraction grating. The multimode optical fiber has an incident end and an output end. The diffraction grating is formed by engraving on the end face of the output end of the multimode optical fiber. The diffraction grating is arranged opposite to the photosensitive surface of the detector. When the output end of the external light source is matched and docked with the incident end of the multimode optical fiber, the light beam emitted by the external light source is incident on the diffraction grating via the multimode optical fiber, Fraunhofer diffraction occurs on the diffraction grating and forms an output light including first-order diffraction light, and the output light is incident on the photosensitive surface of the detector, and the detector collects the light intensity information of the output light to generate a diffraction image.

2. The optical fiber spectrometer according to claim 1, characterized in that: The core diameter of the multimode optical fiber ranges from 30 microns to 100 microns.

3. The optical fiber spectrometer according to claim 1, characterized in that: The period of the diffraction grating ranges from 1 micron to 3 microns.

4. The optical fiber spectrometer according to claim 1, characterized in that: The distance between the diffraction grating and the photosensitive surface of the detector ranges from 200 microns to 500 microns.

5. The optical fiber spectrometer according to claim 1, characterized in that: The detector is a CCD camera or a CMOS camera.

6. A spectrum detection method, characterized in that: Applied to the optical fiber spectrometer according to any one of claims 1 to 5, the method comprises: Calibrate the fiber optic spectrometer using a tunable laser to obtain an initial transmission matrix, wherein the initial transmission matrix is ​​used to characterize the response characteristics of the fiber optic spectrometer at different wavelengths; The initial transmission matrix is ​​corrected by using a singular value decomposition algorithm to obtain a final transmission matrix; The optical fiber spectrometer is used to measure the light source to be measured to obtain a diffraction image to be measured; The spectrum of the light source to be measured is reconstructed according to the diffraction image to be measured and the final transmission matrix.

7. The spectrum detection method according to claim 6, characterized in that: The method of calibrating the optical fiber spectrometer by using a tunable laser to obtain an initial transmission matrix includes: Acquire multiple different reference wavelengths from the wavelength measurement range set by the optical fiber spectrometer; For each of the reference wavelengths, the tunable laser is controlled to emit a quasi-monochromatic light beam at the reference wavelength, and then the fiber spectrometer is controlled to measure the quasi-monochromatic light beam to obtain a calibration diffraction image, and the intensity value of the quasi-monochromatic light beam is measured; For the multiple quasi-monochromatic light beams emitted by the tunable laser at the multiple reference wavelengths, respectively, the initial transmission matrix is ​​determined according to the multiple calibrated diffraction images and multiple intensity values ​​corresponding to the multiple quasi-monochromatic light beams obtained by measurement.

8. The spectrum detection method according to claim 7, characterized in that: Determining the initial transmission matrix according to the measured multiple calibration diffraction images and multiple intensity values ​​corresponding to the multiple quasi-monochromatic light beams includes: For each calibration diffraction image and intensity value corresponding to the quasi-monochromatic light beam, the calibration diffraction image corresponding to the quasi-monochromatic light beam is divided into a plurality of first image regions, a pixel grayscale value of each first image region is obtained, and then the grayscale value is calculated in combination with the intensity value corresponding to the quasi-monochromatic light beam to generate an initial transmission vector corresponding to the quasi-monochromatic light beam; The initial transmission matrix is ​​generated according to the multiple initial transmission vectors corresponding to the multiple quasi-monochromatic light beams obtained by calculation.

9. The spectrum detection method according to claim 6, characterized in that: The adopting a singular value decomposition algorithm to correct the initial transmission matrix to obtain a final transmission matrix comprises: Performing singular value decomposition on the initial transmission matrix to obtain an initial diagonal matrix, an initial left unitary matrix, and an initial right unitary matrix; Reducing the number of non-zero singular values ​​contained in the initial diagonal matrix to generate a final diagonal matrix; According to the column positions of the non-zero singular values ​​contained in the final diagonal matrix in the initial diagonal matrix, column data matching the column positions are extracted from the initial left unitary matrix to generate a final left unitary matrix; According to the row positions of the non-zero singular values ​​contained in the final diagonal matrix in the initial diagonal matrix, extracting row data matching the row positions from the initial right unitary matrix to generate a final right unitary matrix; The final transmission matrix is ​​generated according to the final diagonal matrix, the final left unitary matrix and the final right unitary matrix.

10. The spectrum detection method according to claim 6, characterized in that: The reconstructing the spectrum of the light source to be measured according to the diffraction image to be measured and the final transmission matrix comprises: Calculating a pseudo-inverse matrix of the final transmission matrix; The diffraction image to be measured is divided into a plurality of second image regions, the pixel grayscale value of each second image region is obtained, and then the grayscale value is calculated in combination with the pseudo-inverse matrix to generate a spectrum matrix of the light source to be measured.

Images