Calculation spectrometer based on diffraction type planar micro lens and spectrum reconstruction method

Abstract

The invention discloses a calculation spectrometer based on a diffraction type plane micro lens and a spectrum reconstruction method. According to the invention, a calculation spectrum reconstruction method based on a spectrum-space mapping principle is introduced, and all parameters generating intensity influence on a measurement process are integrated into a mapping matrix, so that the structure of the spectrometer and the measurement process of the spectrum can be greatly simplified; according to the invention, the single-pixel detector only needs to be used for detecting the axial intensity distribution, so that the influence of aberration such as astigmatism and coma when transverse light spots are detected in a traditional method can be effectively eliminated; the device also has the characteristic of no spherical aberration, and can ensure that focusing light spots with different wavelengths at different axial distances realize ideal focusing light spot morphology; moreover, the spectrum coverage range is wide, the cost is low, the device is simple, the structure is compact, the resolution ratio is high, the device can be conveniently integrated into portable equipment, handheld equipment and wearable equipment, and very important significance is achieved for applications needing real-time spectrum detection.

Description

Technical Field

[0001] The present invention relates to spectrum detection technology, and in particular to a computational spectrometer based on a diffraction-type planar microlens and a spectrum reconstruction method. Background Art

[0002] Spectrometers are instruments used to distinguish complex spectra and measure the intensity of components at different wavelengths. They have been widely used in various fields such as chemical analysis, environmental monitoring, material characterization, and astronomical observation. Traditional spectrometers are composed of spectroscopic elements, scanning elements, collimation elements, and detection elements. They need to ensure a long spectroscopic distance to obtain higher performance. Therefore, traditional spectrometers are large in size, expensive, and difficult to integrate. However, with the development of portable, handheld, and wearable devices, people pay more attention to the portability and real-time performance of spectral measurement, the low power consumption, low cost, and integration of the system, rather than the extreme spectral resolution and spectral measurement bandwidth. Therefore, the research on miniaturized spectrometers has received more and more attention in recent years.

[0003] The research on miniaturized spectrometers began in the early 1990s. At present, many types of miniaturized spectrometers have been developed. These miniaturized spectrometers are generally composed of wavelength spectrometers and detection devices. The most common miniaturized spectrometer adopts the working principle of the traditional spectrometer, separates the different spectral components in space through a small dispersion device, and detects them using a miniaturized detection element. In addition, multiple narrow-band filter arrays that filter specific wavelengths can be selected for the production of miniaturized spectrometers. In the above miniaturized spectrometers, the readings of different pixels of the detector have a one-to-one correspondence with the incident wavelength, which directly reflects the incident spectral information. However, in these spectrometers, the spectral resolution continues to decrease with the miniaturization of the spectrometer size, thus limiting the further miniaturization of the above miniaturized spectrometers in volume.

[0004] In recent years, people have introduced the principles of computational optics into spectral detection and developed a new type of miniaturized spectrometer based on computational optics. Its principle is mainly based on the spectrum-space mapping relationship, which converts different wavelength components to a specific spatial distribution, and solves the spectrum information by solving the mapping relationship between spectrum and space. Compared with the above-mentioned miniature spectrometers, computational miniature spectrometers have significant advantages in small size, high resolution, wide spectral width, low cost and fast data acquisition. However, the existing computational spectrometers are mainly based on two-dimensional lateral intensity distribution, and require multi-channel (one-dimensional or two-dimensional) detectors to detect intensity information, which hinders the further reduction of the size of miniaturized spectrometers. Summary of the invention

[0005] In view of the above problems existing in the prior art, the present invention proposes a computational spectrometer based on a diffraction-type planar microlens and a spectrum reconstruction method.

[0006] An object of the present invention is to provide a computational spectrometer based on a diffraction-type planar microlens.

[0007] The computational spectrometer based on the diffraction-type plane microlens of the present invention comprises: a diffraction-type plane microlens, a single-pixel intensity detector, a one-dimensional linear scanning system, a scanning control system and a signal processing system; wherein the center of the diffraction-type plane microlens is located on the optical axis, and the diffraction-type plane microlens is perpendicular to the optical axis; the detection surface of the single-pixel intensity detector is placed parallel to the diffraction-type plane microlens, and the center of the single-pixel intensity detector is located on the optical axis; the single-pixel intensity detector is installed on the one-dimensional linear scanning system, the one-dimensional linear scanning system is connected to the scanning control system, and the one-dimensional linear scanning system is driven by the scanning control system to drive the single-pixel intensity detector to move one-dimensionally along the optical axis direction; the scanning control system is connected to the signal processing system; the single-pixel intensity detector is connected to the signal processing system;

[0008] The diffractive planar microlens comprises a plurality of concentric opaque circular rings and a transparent substrate; the plurality of concentric opaque circular rings are arranged on the transparent substrate, the opaque circular rings are made of opaque materials, a transparent circular ring is formed between every two adjacent opaque circular rings, the transparent circular rings are concentric, and the center of the circular rings is located on the optical axis, the width of each circular ring, the radius of the circular ring and the radius of the outermost opaque circular ring of the diffractive planar microlens are set according to the principle of a zone plate or the principle of a detour phase, so that light of different wavelengths is focused along the optical axis to different positions on the optical axis;

[0009] The incident light to be measured of the broadband spectrum passes through the diffraction-type plane microlens parallel to the optical axis. The diffraction-type plane microlens based on the principle of the zone plate or the principle of the roundabout phase has the characteristics of axial dispersion. After passing through a plurality of concentric light-transmitting rings, light of different wavelengths accumulates different phases. The incident light is focused behind the diffraction-type plane microlens and produces a series of focal points along the optical axis. Light of different wavelengths is focused to different positions on the optical axis along the optical axis. The focal points of light of different wavelengths are arranged in sequence along the optical axis. The diffraction-type plane microlens has a negative dispersion characteristic. The longer the wavelength, the shorter the focal length, and the shorter the wavelength, the longer the focal length. The one-dimensional linear scanning is controlled by the scanning control system. The scanning distance and scanning step length of the system drive the one-dimensional linear scanning system to drive the single-pixel intensity detector to move one-dimensionally along the optical axis direction, and scan and detect the axial light intensity distribution of the incident light after passing through the diffraction-type planar microlens; during the scanning process, the axial light field intensity distribution detected by the single-pixel intensity detector is matched one-to-one with the axial scanning position coordinates of the one-dimensional linear scanning system; the axial light field intensity distribution detected by the single-pixel intensity detector and the corresponding axial scanning position coordinates of the one-dimensional linear scanning system are input into the signal processing system; based on the spectrum-space mapping principle, the signal processing system obtains the incident spectrum according to the corresponding axial light field intensity distribution and the mapping matrix.

[0010] The material of the opaque ring is a thin film material of graphene oxide, graphene or perovskite. The thin film material is placed on a transparent substrate, and the femtosecond laser direct writing technology is used to etch the thin film material at the position of the corresponding light-transmitting ring in a single process, thereby forming a ring-shaped light-transmitting ring, which can be prepared in large quantities and at low cost.

[0011] According to the different wavelength ranges of spectral detection, the single-pixel intensity detector uses different materials. In the visible light and near-infrared band less than 900nm, silicon is used as the detector material; in the near-infrared band of 900-1700nm, germanium or indium gallium arsenide is used as the detector material. The detection surface size of the single-pixel intensity detector needs to be as small as possible to ensure that the detector has the best detection sensitivity to the axial intensity change. Specifically, the diameter of the focused spot on the focal plane of the diffraction-type planar microlens is about 1 micron (1 / e 2 Intensity distribution), the side length of the detection surface should be in the same order of magnitude as the diameter of the focused light spot, and the side length of the detection surface should be greater than or equal to the diameter of the focused light spot.

[0012] One-dimensional linear scanning systems use micro-electromechanical system-based devices, such as one-dimensional piezoelectric ceramic actuators, which can reduce the footprint of the entire system.

[0013] The transparent substrate is made of a transparent glass plate, and the transparent substrate is a circular or square substrate. When the transparent substrate is circular, the diameter of the transparent substrate is the same as the diameter of the outermost opaque circular ring. When the transparent substrate is square, the side length of the transparent substrate is the same as the diameter of the outermost opaque circular ring. The thickness of the transparent substrate is 200-500μm, so as to ensure the mechanical strength of the diffraction type planar microlens.

[0014] According to the parameters of the diffraction-type planar microlens, within the measured spectral range, the scanning stroke of the one-dimensional linear scanning system is also determined. The number of intensity signals measured by the single-pixel intensity detector is related to the scanning step length. The smaller the scanning step length, the more accurate the measured spectrum; the larger the scanning step length, the lower the accuracy of the measured spectrum. In practical applications, the scanning step length is flexibly set according to the resolution requirements of the spectral measurement. For the design scheme shown in the present invention, when the spectral measurement range is 380-780nm, the axial scanning distance of the one-dimensional linear scanning system is about 250μm, and the scanning step length is set to the micrometer level to achieve the spectral measurement requirements.

[0015] Based on the principle of spectrum-space mapping, the axial light field intensity distribution measured at different axial distances behind the diffractive plane microlens is discretized and measured. The axial light field intensity distribution I(z) is a one-dimensional vector, the number of elements of the axial light field intensity distribution I(z) is n, and z is the axial scanning position coordinate; after the unknown incident spectrum is discretized and sampled in the order of wavelength, the unknown incident spectrum S(λ) is a one-dimensional vector, the number of elements of the incident spectrum S(λ) is m, and λ is the wavelength. According to the principle of spectrum-space mapping, the axial light field intensity distribution I(z) and the incident spectrum S(λ) have the following relationship: I(z) = H·S(λ), where H is the mapping matrix. The axial light intensity distribution is actually the result of the linear superposition of the axial response of the diffractive plane microlens to all wavelength components of the incident spectrum. The i-th column element of the mapping matrix H represents the contribution weight of the i-th wavelength component in the incident spectrum S(λ) to the axial focusing light intensity. The mapping matrix is ​​obtained by pre-calibration using incident light of known spectrum, including the following steps: first, the incident light of known spectrum is filtered out narrowband wavelength components of different central wavelengths, the different wavelength components filtered out are represented by S'(λ), and the number of filtered wavelengths is the same as the number of elements of the unknown incident spectrum S(λ) to be measured; then, the incident light with narrowband wavelength components is sequentially irradiated onto the planar microlens respectively, and the axial light field intensity distribution I'(z) behind the lens along the optical axis when different narrowband wavelength components are incident is detected respectively; the matrix elements of the mapping matrix H are calculated by the relationship I'(z)=H·S'(λ). Since the matrix elements of the mapping matrix H in a specific system will not change due to different incident spectrum components, the pre-calibration only needs to be performed once for a certain diffraction type planar microlens. In actual measurement, for the incident spectrum S(λ) of unknown wavelength component, a one-dimensional linear scanning system is used to drive a single-pixel detector to axially scan the axial focused light field intensity distribution I(z) after passing through the lens, and then the data is transmitted to the signal processing system. Through the relationship: S(λ) = H -1 ·I(z) can be used to calculate the incident spectrum, where H -1 is the inverse matrix of the mapping matrix. According to the above content, the axial light field intensity distribution I(z) and the incident spectrum S(λ) can have the same number of elements, that is, m=n, or different numbers of elements, that is, m≠n. In actual spectral measurement, it can be selected according to the needs, which brings great flexibility to spectral measurement. When the number of elements is not equal (m≠n), H -1 It is obtained by calculating the generalized inverse matrix of H. m and n are natural numbers ≥ 2. The number of elements determines the resolution of reconstruction. Before the resolution of the system itself is reached, the more elements the better.

[0016] In the diffractive planar microlens designed using the zone plate principle, the transparent rings and the opaque rings are alternately distributed along the radial direction, and the widths of the transparent rings and the opaque rings gradually decrease. The calculation formula for the radius of the outer ring of the kth transparent or opaque ring is: where λ d is the designed working wavelength of the diffractive planar microlens, f d is the focal length of the diffractive plane microlens at the designed working wavelength; in the diffractive plane microlens designed by the detour phase principle, the distribution of the light-transmitting ring on the surface of the diffractive plane microlens along the radial direction is regarded as a series of gratings with a grating period of Λ and unequal slit intervals in adjacent periods, and the position offset of each slit relative to the center coordinate in its period is ΔD. At this time, the light passing through two adjacent grating slits will produce Therefore, the phase modulation of the transmitted light can be achieved by moving the spatial position of the slits within their respective grating periods. For a converging lens, its phase distribution is expressed as Where r is the radius of the diffractive planar microlens, λ d is the designed working wavelength of the diffractive planar microlens, f d is the focal length of the diffractive planar microlens at the designed working wavelength. According to the required phase distribution, the offset of the light-transmitting slit relative to the center of the grating period in each grating period is determined. According to the above derivation, the radius of the center of the kth light-transmitting ring is calculated as: r k =kΛ+ΔD. Finally, r m A diffractive plane microlens is obtained by performing a circular symmetric transformation, in which the light-transmitting slits constitute the light-transmitting ring of the diffractive plane microlens, and the remaining parts of the grating period constitute the opaque ring of the diffractive plane microlens. When the width of the light-transmitting ring is set to half of the grating period, the highest focusing efficiency is obtained. When the radius and width of the light-transmitting ring are determined, the width of the opaque ring is also uniquely determined. For the diffractive plane microlens, its numerical aperture (NA) is calculated according to the formula NA = r k / f d Calculation shows that for the diffractive plane microlens designed based on the zone plate principle and the detour phase principle, in order to obtain an ideal focused spot shape, its NA should be less than 0.5. According to the NA, the radius of the diffractive plane microlens, i.e. the radius of the outermost opaque ring, is restricted, and the number of transparent and opaque rings is limited. The dispersion relation of the diffractive plane microlens can be expressed by the formula Calculate, where λ d is the designed working wavelength of the diffractive planar microlens, f d is the focal length of the diffractive plane microlens at the designed working wavelength, f 1 is the incident wavelength λ 1 Deviation from the designed working wavelength λ dThe focal length of the diffractive plane microlens is calculated, so as to calculate the dispersion relation of the diffractive plane microlens under different incident wavelengths.

[0017] Another object of the present invention is to provide a spectrum reconstruction method based on a diffractive planar microlens.

[0018] The spectrum reconstruction method based on the diffraction type planar microlens of the present invention comprises the following steps:

[0019] 1) Preparation of a computational spectrometer based on a diffraction-type planar microlens:

[0020] Placing an opaque thin film material on a transparent substrate, etching the thin film material at the position of the corresponding transparent ring by a single processing method using a femtosecond laser direct writing technique, and forming a plurality of concentric opaque rings with the remaining thin film material, wherein a transparent ring is formed between every two adjacent opaque rings, each of which is concentric and has its center on the optical axis, and setting the width of each ring, the radius of the ring, and the radius of the outermost opaque ring of the diffraction-type planar microlens according to the zone plate principle or the detour phase principle, so that light of different wavelengths is focused along the optical axis to different positions on the optical axis;

[0021] The center of the diffraction-type plane microlens is located on the optical axis, and the diffraction-type plane microlens is perpendicular to the optical axis; the detection surface of the single-pixel intensity detector is placed parallel to the diffraction-type plane microlens, and the center of the single-pixel intensity detector is located on the optical axis; the single-pixel intensity detector is installed on a one-dimensional linear scanning system, and the one-dimensional linear scanning system is connected to a scanning control system, and the one-dimensional linear scanning system is driven by the scanning control system to drive the single-pixel intensity detector to move one-dimensionally along the optical axis direction; the scanning control system is connected to a signal processing system; the single-pixel intensity detector is connected to the signal processing system;

[0022] 2) The incident light to be measured of the broadband spectrum passes through the diffractive plane microlens parallel to the optical axis. The diffractive plane microlens based on the principle of the zone plate or the principle of the roundabout phase has the characteristics of axial dispersion. After passing through a plurality of concentric light-transmitting rings, light of different wavelengths accumulates different phases. The incident light is focused behind the diffractive plane microlens and produces a series of focal points along the optical axis. Light of different wavelengths is focused to different positions on the optical axis along the optical axis. The focal points of light of different wavelengths are arranged in sequence along the optical axis. The diffractive plane microlens has a negative dispersion characteristic. The longer the wavelength, the shorter the focal length, and the shorter the wavelength, the longer the focal length.

[0023] 3) controlling the scanning distance and scanning step length of the one-dimensional linear scanning system through a scanning control system, driving the one-dimensional linear scanning system to drive the single-pixel intensity detector to move one-dimensionally along the optical axis direction, scanning and detecting the axial light intensity distribution of the incident light after passing through the diffraction-type planar microlens, and in the scanning process, making the axial light field intensity distribution detected by the single-pixel intensity detector correspond one-to-one to the axial scanning position coordinates of the one-dimensional linear scanning system;

[0024] 4) inputting the axial light field intensity distribution detected by the single pixel intensity detector and the axial scanning position coordinates of the corresponding one-dimensional linear scanning system into the signal processing system;

[0025] 5) The signal processing system is based on the spectrum-space mapping principle, according to the corresponding axial light field intensity distribution and mapping matrix,

[0026] Get the incident spectrum.

[0027] Before reconstructing the spectrum of the incident light to be measured, it is necessary to obtain a mapping matrix based on the spectrum-space mapping principle, including the following steps:

[0028] i. Based on the spectrum-space mapping principle, the axial light field intensity distribution measured at different axial distances behind the diffractive plane microlens is discretized and measured, and the axial light field intensity distribution I(z) is a one-dimensional vector, the number of elements of the axial light field intensity distribution I(z) is n, and z is the axial scanning position coordinate; after the unknown incident spectrum is discretized and sampled in the order of wavelength, the unknown incident spectrum S(λ) is a one-dimensional vector, the number of elements of the incident spectrum S(λ) is m, and λ is the wavelength; According to the spectrum-space mapping principle, the axial light field intensity distribution I(z) and the incident spectrum S(λ) have the following relationship: I(z)=H·S(λ), where H is the mapping matrix; the axial light intensity distribution is actually the result of the linear superposition of the axial response of the diffractive plane microlens to all wavelength components of the incident spectrum, and the i-th column element of the mapping matrix H represents the contribution weight of the i-th wavelength component in the incident spectrum S(λ) to the axial focusing light intensity;

[0029] ii. Using incident light of known spectrum to obtain the mapping matrix through pre-calibration:

[0030] First, the incident light of known spectrum is filtered to obtain narrowband wavelength components with different central wavelengths. The different wavelength components filtered out are represented by S'(λ), and the number of wavelengths filtered out is the same as the number of elements of the unknown incident spectrum S(λ) to be measured; then, the incident light with narrowband wavelength components is sequentially irradiated onto the plane microlens respectively, and the axial light field intensity distribution I'(z) along the optical axis behind the lens is detected respectively when different narrowband wavelength components are incident;

[0031] iii. The matrix elements of the mapping matrix H are calculated by the relationship I'(z)=H·S'(λ).

[0032] Since the matrix elements of the mapping matrix H in a specific system will not change due to different incident spectral components, the pre-calibration only needs to be performed once for a certain diffractive planar microlens.

[0033] In step 5), the signal processing system receives the axial focusing light field intensity distribution I(z) obtained by the single pixel detector, and uses the relationship: S(λ) = H -1 I(z), calculate the incident spectrum, where H -1 is the inverse matrix of the mapping matrix.

[0034] Advantages of the present invention:

[0035] 1. The present invention introduces a computational spectrum reconstruction method based on the spectrum-space mapping principle. All parameters that affect the intensity of the measurement process are integrated into the mapping matrix. No additional measurement and calculation are required. Only the axial light field intensity distribution of the focused light spot needs to be concerned.

[0036] 2. The spectrum measurement in the present invention can be realized by using a single-pixel detector to detect the axial light field intensity distribution along the optical axis direction, without relying on a linear array or planar array detector, which can not only avoid the influence of aberrations such as astigmatism and coma, but also greatly reduce the size of the miniaturized spectrometer;

[0037] 3. The diffractive planar microlens used in the present invention also has the characteristic of no spherical aberration, which can ensure that the focused light spots of different wavelengths at different axial distances achieve ideal focused light spot morphology;

[0038] 4. Compared with traditional spectral measurement technology, the present invention can greatly improve the spectral measurement resolution and can obtain a spectral resolution of 5nm, which is five times the resolution improvement compared to methods that do not use computational optics;

[0039] 5. The spectrometer of the present invention is small in size, with its length, width and height all being in the order of hundreds of microns; in addition, only an axial scanning distance of about 200 microns is required to achieve broadband spectrum measurement covering the entire visible spectrum range;

[0040] 6. The spectral measurement range of the present invention is wide, the cost is low, the device is simple, the structure is compact, and the resolution is high. It can be easily integrated into portable devices, handheld devices, and wearable devices. It is of great significance for applications that require real-time spectral detection, such as human health monitoring, food nutrient analysis, counterfeit banknote and counterfeit drug detection, and other application fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1It is a structural block diagram of a computational spectrometer based on a diffraction-type planar microlens of the present invention;

[0042] Figure 2 A schematic diagram of a diffractive planar microlens of an embodiment of a computational spectrometer based on a diffractive planar microlens of the present invention;

[0043] Figure 3 The intensity distribution diagram of the focused light spot in the axial direction after the incident light of different wavelengths is obtained according to an embodiment of the spectrum reconstruction method based on the diffraction type planar microlens of the present invention;

[0044] Figure 4 Figure 1 is a graph showing the relationship between the focal length changes after different wavelengths of incident light and the simulation results of the spectrum reconstruction obtained by an embodiment of the spectrum reconstruction method based on a diffractive planar microlens of the present invention, wherein (a) is a graph showing the relationship between the focal length changes with wavelength obtained by theoretical calculation and simulation based on the diffraction theorem, and (b) is a graph showing the incident spectrum set artificially and the reconstructed spectrum simulated by the spectrum reconstruction method based on a diffractive planar microlens;

[0045] Figure 5 The diagram is a result of two different incident spectra obtained by an embodiment of the spectrum reconstruction method based on a diffractive planar microlens of the present invention, and is compared with the results measured by a commercial spectrometer. DETAILED DESCRIPTION

[0046] The present invention will be further described below through specific embodiments in conjunction with the accompanying drawings.

[0047] like Figure 1 As shown, the computational spectrometer based on the diffraction-type plane microlens of the present embodiment comprises: a diffraction-type plane microlens, a single-pixel intensity detector, a one-dimensional linear scanning system, a scanning control system and a signal processing system; wherein, the center of the diffraction-type plane microlens is located on the optical axis, and the diffraction-type plane microlens is perpendicular to the optical axis; the detection surface of the single-pixel intensity detector is placed parallel to the diffraction-type plane microlens, and the center of the single-pixel intensity detector is located on the optical axis; the single-pixel intensity detector is installed on the one-dimensional linear scanning system, the one-dimensional linear scanning system is connected to the scanning control system, and the one-dimensional linear scanning system is driven by the scanning control system to drive the single-pixel intensity detector to move one-dimensionally along the optical axis direction; the scanning control system is connected to the signal processing system; the single-pixel intensity detector is connected to the signal processing system;

[0048] like Figure 2As shown, the diffraction type plane microlens includes a plurality of concentric opaque circular rings and a transparent substrate; the plurality of concentric opaque circular rings are arranged on the transparent substrate, the opaque circular rings are made of opaque materials, and a transparent circular ring is formed between each two adjacent opaque circular rings, each of which is concentric, and the center of the circle is located on the optical axis. The width and radius of each transparent circular ring are set according to the detour phase principle, so that different wavelengths are focused to different positions along the optical axis. The radius of the outermost opaque circular ring of the diffraction type plane microlens is 100μm; in the detour phase principle, the distribution of the transparent circular ring along the radial direction on the surface of the diffraction type plane microlens is regarded as a series of gratings with unequal slit intervals, the grating period is set to 2μm, the line width of the transparent circular ring is 1μm, and the focal length of the diffraction type plane microlens at the design working wavelength of 550nm is 300μm, and the numerical aperture NA is 0.32. The distribution of the transparent circular ring of the diffraction type plane microlens is shown in the following table:

[0049] Number of rings Radius(μm) Number of rings Radius(μm) Number of rings Radius(μm) Number of rings Radius(μm) Number of rings Radius(μm) 1 0 11 19.3218 21 38.6404 31 58.9301 41 80.1907 2 1 12 20.8733 22 41.7175 32 61.4920 42 82.2375 3 3.0151 13 22.3732 23 42.7036 33 64.0025 43 84.2328 4 4.9786 14 23.8215 24 45.6764 34 64.4206 44 86.1766 5 6.8906 15 27.2593 25 46.5581 35 66.8280 45 88.0689 6 8.7512 16 28.6047 26 49.4292 36 69.1839 46 89.9097 7 10.5602 17 29.8985 27 50.2079 37 71.4883 47 91.6989 8 12.3176 18 33.1817 28 52.9759 38 73.7412 48 93.4366 9 14.0236 19 34.3725 29 55.6924 39 75.9425 49 95.1229 10 15.6781 20 37.5526 30 56.3166 40 78.0924 50 98.7984

[0050] In this embodiment, the material of the opaque ring is graphene oxide. Graphene oxide with a thickness of 250nm is placed on a transparent substrate, and the graphene oxide is etched away at the position of the corresponding light-transmitting ring by a single process using femtosecond laser direct writing technology, thereby forming a ring-shaped light-transmitting ring. The one-dimensional linear scanning system uses a one-dimensional piezoelectric ceramic driver. The transparent substrate uses a transparent glass plate with a thickness of 300μm; when the spectral measurement range is 380-780nm, the axial scanning distance of the one-dimensional linear scanning system is about 250μm, and the scanning step size is set to 1μm; the single-pixel intensity detector is based on silicon material, and the working wavelength range of the single-pixel intensity detector is 200-1100nm.

[0051] For a wide spectrum incident light with a wavelength range of 380-780nm covering the entire visible light wavelength range, different wavelength components (380nm, 480nm, 580nm, 680nm, 780nm) are selected and the intensity distribution of the focal point along the axial direction after passing through the lens is as follows Figure 3 As shown, the focal length corresponding to 380nm is 434μm, the focal length corresponding to 780nm is 211μm, and the number of elements of the axial light field intensity distribution I(z) is selected to be 235 at an axial scanning step of 1μm.

[0052] Before reconstructing the spectrum of the incident light to be measured, it is necessary to obtain a mapping matrix based on the spectrum-space mapping principle, including the following steps:

[0053] i. Based on the spectrum-space mapping principle, the axial light field intensity distribution measured at different axial distances behind the diffractive plane microlens is discretized and measured, and the axial light field intensity distribution I(z) is a one-dimensional vector, the number of elements of the axial light field intensity distribution I(z) is n, and z is the axial scanning position coordinate; after the unknown incident spectrum is discretized and sampled in the order of wavelength, the unknown incident spectrum S(λ) is a one-dimensional vector, the number of elements of the incident spectrum S(λ) is m, and λ is the wavelength; According to the spectrum-space mapping principle, the axial light field intensity distribution I(z) and the incident spectrum S(λ) have the following relationship: I(z)=H·S(λ), where H is the mapping matrix; the axial light intensity distribution is actually the result of the linear superposition of the axial response of the diffractive plane microlens to all wavelength components of the incident spectrum, and the i-th column element of the mapping matrix H represents the contribution weight of the i-th wavelength component in the incident spectrum S(λ) to the axial focusing light intensity;

[0054] ii. Using incident light of known spectrum to obtain a mapping matrix through pre-calibration, the mapping matrix includes the following steps: first, filtering out narrowband wavelength components of different central wavelengths from the incident light of known spectrum, wherein the different wavelength components filtered out are represented by S'(λ), and the number of filtered out wavelengths is the same as the number of elements of the unknown incident spectrum S(λ) to be measured; then, sequentially irradiating the incident light with narrowband wavelength components onto a planar microlens, respectively, and detecting the axial light field intensity distribution I'(z) along the optical axis behind the lens when different narrowband wavelength components are incident;

[0055] iii. The matrix elements of the mapping matrix H are calculated by the relationship I'(z)=H·S'(λ).

[0056] Since the matrix elements of the mapping matrix H in a specific system will not change due to different incident spectral components, the pre-calibration only needs to be performed once for a certain diffractive planar microlens.

[0057] The spectrum reconstruction method based on the diffractive planar microlens of this embodiment includes the following steps:

[0058] 1) Prepare a computational spectrometer based on a diffractive planar microlens, such as Figure 1 As shown;

[0059] 2) The incident light to be measured of the broadband spectrum passes through the diffractive plane microlens parallel to the optical axis. The diffractive plane microlens based on the principle of the zone plate has the characteristics of axial dispersion. After passing through multiple concentric light-transmitting rings, light of different wavelengths accumulates different phases. The incident light is focused behind the diffractive plane microlens and produces a series of focal points along the optical axis. Incident light of different wavelengths is focused to different positions along the optical axis. The focal points of incident light of different wavelengths are arranged in sequence along the optical axis. The diffractive plane microlens has a negative dispersion characteristic. The longer the wavelength, the shorter the focal length, and the shorter the wavelength, the longer the focal length.

[0060] 3) controlling the scanning distance and scanning step length of the one-dimensional linear scanning system through a scanning control system, driving the one-dimensional linear scanning system to drive the single-pixel intensity detector to move one-dimensionally along the optical axis direction, scanning and detecting the axial light intensity distribution of the incident light after passing through the diffraction-type planar microlens, and in the scanning process, making the axial light field intensity distribution detected by the single-pixel intensity detector correspond one-to-one to the axial scanning position coordinates of the one-dimensional linear scanning system;

[0061] 4) inputting the axial light field intensity distribution detected by the single pixel intensity detector and the axial scanning position coordinates of the corresponding one-dimensional linear scanning system into the signal processing system;

[0062] 5) The signal processing system receives the axial focused light field intensity distribution I(z) obtained by the single pixel detector, and uses the relationship: S(λ) = H -1 ·I(z), calculate the incident spectrum, where H- 1 is the inverse matrix of the mapping matrix.

[0063] If the spectrum reconstruction method proposed in the present invention is not used, the traditional spectrum measurement method will be limited by diffraction. According to the Rayleigh criterion, the corresponding spectrum resolution when the spectrum reconstruction method of the present invention is not used is shown in the following table:

[0064] Wavelength (nm) 380 580 780 Resolution, Δλ(nm) 12.54 19.14 25.74

[0065] At different wavelengths, the corresponding resolution is also different. When the incident wavelength is 780nm, the system resolution is only 25.74nm. Figure 4 According to the simulation results in , after using the spectrum reconstruction method of the present invention, the spectrum can be resolved when the wavelength interval is 5nm. Therefore, the method proposed in the present invention can achieve a five-fold resolution improvement.

[0066] Figure 5 The comparison results between the reconstructed spectrum obtained by using the spectrum reconstruction method of the present invention for two different incident spectra based on this embodiment and the commercial spectrometer are shown.

[0067] Finally, it should be noted that the purpose of publishing the embodiments is to help further understand the present invention, but those skilled in the art can understand that various substitutions and modifications are possible without departing from the spirit and scope of the present invention and the appended claims. Therefore, the present invention should not be limited to the contents disclosed in the embodiments, and the scope of protection claimed by the present invention shall be subject to the scope defined in the claims.

Claims

1. A computational spectrometer based on a diffraction-type planar microlens, characterized in that: The computational spectrometer comprises: a diffraction-type plane microlens, a single-pixel intensity detector, a one-dimensional linear scanning system, a scanning control system and a signal processing system; wherein the center of the diffraction-type plane microlens is located on the optical axis, and the diffraction-type plane microlens is perpendicular to the optical axis; the detection surface of the single-pixel intensity detector is placed parallel to the diffraction-type plane microlens, and the center of the single-pixel intensity detector is located on the optical axis; the single-pixel intensity detector is installed on the one-dimensional linear scanning system, the one-dimensional linear scanning system is connected to the scanning control system, and the one-dimensional linear scanning system is driven by the scanning control system to drive the single-pixel intensity detector to move one dimension along the optical axis direction; The scanning control system is connected to the signal processing system; the single pixel intensity detector is connected to the signal processing system; The diffractive planar microlens comprises a plurality of concentric opaque circular rings and a transparent substrate; the plurality of concentric opaque circular rings are arranged on the transparent substrate, the opaque circular rings are made of opaque materials, a transparent circular ring is formed between every two adjacent opaque circular rings, the transparent circular rings are concentric, and the center of the circular rings is located on the optical axis, the width of each circular ring, the radius of the circular ring and the radius of the outermost opaque circular ring of the diffractive planar microlens are set according to the principle of a zone plate or the principle of a detour phase, so that light of different wavelengths is focused along the optical axis to different positions on the optical axis; The incident light to be measured of the broadband spectrum passes through the diffraction-type plane microlens parallel to the optical axis. The diffraction-type plane microlens based on the principle of the zone plate or the principle of the roundabout phase has the characteristics of axial dispersion. After passing through a plurality of concentric light-transmitting rings, light of different wavelengths accumulates different phases. The incident light is focused behind the diffraction-type plane microlens and produces a series of focal points along the optical axis. Light of different wavelengths is focused to different positions on the optical axis along the optical axis. The focal points of light of different wavelengths are arranged in sequence along the optical axis. The diffraction-type plane microlens has a negative dispersion characteristic. The longer the wavelength, the shorter the focal length, and the shorter the wavelength, the longer the focal length. The one-dimensional linear scanning is controlled by the scanning control system. The scanning distance and scanning step length of the system drive the one-dimensional linear scanning system to drive the single-pixel intensity detector to move one-dimensionally along the optical axis direction, and scan and detect the axial light intensity distribution of the incident light after passing through the diffraction-type planar microlens; during the scanning process, the axial light field intensity distribution detected by the single-pixel intensity detector is matched one-to-one with the axial scanning position coordinates of the one-dimensional linear scanning system; the axial light field intensity distribution detected by the single-pixel intensity detector and the corresponding axial scanning position coordinates of the one-dimensional linear scanning system are input into the signal processing system; based on the spectrum-space mapping principle, the signal processing system obtains the incident spectrum according to the corresponding axial light field intensity distribution and the mapping matrix.

2. The computational spectrometer according to claim 1, wherein: The material of the opaque ring is a thin film material of graphene oxide, graphene or perovskite.

3. The computational spectrometer according to claim 1, wherein: The one-dimensional linear scanning system uses a device based on a micro-electromechanical system.

4. The computational spectrometer according to claim 1, wherein: The transparent substrate is made of transparent glass.

5. The computational spectrometer according to claim 1, wherein: The thickness of the transparent substrate is 200-500 μm.

6. A spectrum reconstruction method of a computational spectrometer based on a diffraction-type planar microlens as claimed in claim 1, characterized in that: The spectrum reconstruction method comprises the following steps: 1) Prepare a computational spectrometer based on a diffractive planar microlens; 2) The incident light to be measured of the broadband spectrum passes through the diffractive plane microlens parallel to the optical axis. The diffractive plane microlens based on the principle of the zone plate or the principle of the roundabout phase has the characteristics of axial dispersion. After passing through a plurality of concentric light-transmitting rings, light of different wavelengths accumulates different phases. The incident light is focused behind the diffractive plane microlens and produces a series of focal points along the optical axis. Light of different wavelengths is focused to different positions on the optical axis along the optical axis. The focal points of light of different wavelengths are arranged in sequence along the optical axis. The diffractive plane microlens has a negative dispersion characteristic. The longer the wavelength, the shorter the focal length, and the shorter the wavelength, the longer the focal length. 3) controlling the scanning distance and scanning step length of the one-dimensional linear scanning system through a scanning control system, driving the one-dimensional linear scanning system to drive the single-pixel intensity detector to move one-dimensionally along the optical axis direction, scanning and detecting the axial light intensity distribution of the incident light after passing through the diffraction-type planar microlens, and in the scanning process, making the axial light field intensity distribution detected by the single-pixel intensity detector correspond one-to-one to the axial scanning position coordinates of the one-dimensional linear scanning system; 4) inputting the axial light field intensity distribution detected by the single pixel intensity detector and the axial scanning position coordinates of the corresponding one-dimensional linear scanning system into the signal processing system; 5) The signal processing system is based on the spectrum-space mapping principle and obtains the incident spectrum according to the corresponding axial light field intensity distribution and mapping matrix.

7. The real spectrum reconstruction method according to claim 6, characterized in that: In step 1), a computational spectrometer based on a diffraction-type planar microlens is prepared, comprising the following steps: Placing an opaque thin film material on a transparent substrate, etching the thin film material at the position of the corresponding transparent ring by a single processing method using a femtosecond laser direct writing technique, and forming a plurality of concentric opaque rings with the remaining thin film material, wherein a transparent ring is formed between every two adjacent opaque rings, each of which is concentric and has its center on the optical axis, and setting the width of each ring, the radius of the ring, and the radius of the outermost opaque ring of the diffraction-type planar microlens according to the zone plate principle or the detour phase principle, so that different wavelengths are focused to different positions along the optical axis; The center of the diffraction-type plane microlens is located on the optical axis, and the diffraction-type plane microlens is perpendicular to the optical axis; the detection surface of the single-pixel intensity detector is placed parallel to the diffraction-type plane microlens, and the center of the single-pixel intensity detector is located on the optical axis; the single-pixel intensity detector is installed on a one-dimensional linear scanning system, and the one-dimensional linear scanning system is connected to a scanning control system, and the one-dimensional linear scanning system is driven by the scanning control system to drive the single-pixel intensity detector to move one-dimensionally along the optical axis direction; The scanning control system is connected to the signal processing system; and the single pixel intensity detector is connected to the signal processing system.

8. The spectrum reconstruction method according to claim 6, characterized in that: Before reconstructing the spectrum of the incident light to be measured, it is necessary to obtain a mapping matrix based on the spectrum-space mapping principle, including the following steps: i. Based on the spectrum-space mapping principle, the axial light field intensity distribution measured at different axial distances behind the diffractive plane microlens is discretized and measured, and the axial light field intensity distribution I(z) is a one-dimensional vector, the number of elements of the axial light field intensity distribution I(z) is n, and z is the axial scanning position coordinate; after the unknown incident spectrum is discretized and sampled in the order of wavelength, the unknown incident spectrum S(λ) is a one-dimensional vector, the number of elements of the incident spectrum S(λ) is m, and λ is the wavelength; According to the spectrum-space mapping principle, the axial light field intensity distribution I(z) and the incident spectrum S(λ) have the following relationship: I(z)=H·S(λ), where H is the mapping matrix; the axial light intensity distribution is actually the result of the linear superposition of the axial response of the diffractive plane microlens to all wavelength components of the incident spectrum, and the i-th column element of the mapping matrix H represents the contribution weight of the i-th wavelength component in the incident spectrum S(λ) to the axial focusing light intensity; ii. Using incident light of known spectrum to obtain the mapping matrix through pre-calibration: First, the incident light of known spectrum is filtered to obtain narrowband wavelength components with different central wavelengths. The different wavelength components filtered out are represented by S'(λ), and the number of wavelengths filtered out is the same as the number of elements of the unknown incident spectrum S(λ) to be measured; then, the incident light with narrowband wavelength components is sequentially irradiated onto the plane microlens respectively, and the axial light field intensity distribution I'(z) along the optical axis behind the lens is detected respectively when different narrowband wavelength components are incident; iii. The matrix elements of the mapping matrix H are calculated by the relationship I'(z)=H·S'(λ).

9. The spectrum reconstruction method according to claim 8, characterized in that: In step 5), the signal processing system receives the axial focusing light field intensity distribution I(z) obtained by the single pixel detector, and uses the relationship: S(λ) = H -1 I(z), calculate the incident spectrum, where H -1 is the inverse matrix of the mapping matrix.

Images