Computational spectrometer based on photochromic microcavity coupling and spectral reconstruction method thereof

Abstract

The application discloses a kind of based on photochromic microcavity coupling computing spectrometer and its spectral reconstruction method, belong to snapshot type computing spectrometer technical field, computing spectrometer includes photochromic Fabry-Perot microcavity filter array and CMOS image camera;Filter array is composed of multiple filter units, each unit is PET substrate, first silver film, photochromic molecular layer and second silver film from bottom to top in turn, and the resonant frequency of microcavity is regulated by changing the thickness of molecular layer.Spectral reconstruction method includes: using monochromatic light to irradiate the array, the coupling strength between photochromic molecular transition frequency and microcavity mode is regulated by applying different length ultraviolet light irradiation, and the spectral response function matrix is obtained by reading transmission intensity;Unknown light is irradiated after applying the same length ultraviolet light, and the response value is read and reconstructed spectrum by compressed sensing algorithm.The application expands the number of spectral channels by non-contact light regulation, without electrode structure, which is beneficial to the miniaturization of spectrometer.

Description

Technical Field

[0001] This invention relates to the field of snapshot-type computational spectrometer technology, and in particular to a computational spectrometer based on photochromic microcavity coupling and its spectral reconstruction method. Background Technology

[0002] Computational spectrometry is a novel spectral measurement technology based on computational imaging principles. Its core lies in capturing and encoding spectral data using a limited number of encoding elements, and then reconstructing the unknown incident spectrum from the captured signal using reconstruction algorithms such as compressed sensing. Unlike traditional spectrometers that rely on dispersive elements such as gratings, prisms, or interferometers, computational spectrometers tightly integrate encoding elements with the signal detection unit, avoiding the long optical paths required by traditional spectrometers to achieve high spectral resolution, making them suitable for chip-based applications.

[0003] On-chip computational spectrometers typically integrate a filter array onto the top of a CMOS image sensor. They encode unknown spectral information using a pre-calibrated response function, and then decode and reconstruct the information using a reconstruction algorithm. However, the spectral resolution of computational spectrometers is limited by the number of spectral channels, and it is difficult to further improve it given the limited space available for the filters.

[0004] To address these issues, existing research has introduced dynamic control mechanisms to increase the number of spectral channels. For example, a team from Beijing University of Aeronautics and Astronautics proposed a miniature computational spectrometer based on electrochromic filters, which modulates the spectral response of the filters through voltage changes. In invention patent CN121577157A, a dynamically adjustable electrochromic miniature spectrometer and its reconstruction method are proposed, which is a spectrometer scheme based on a combination of an electrochromator and a microcavity array. Furthermore, invention patent CN118915219A uses phase change materials to control the transmission characteristics of filters through annealing temperature. However, electrochromic schemes rely on complex electrode structures and voltage-driven circuits, while phase change material schemes rely on precise temperature control devices, both hindering further miniaturization of computational spectrometers.

[0005] Based on this, a computational spectrometer based on photochromic microcavity coupling and its spectral reconstruction method are proposed. Summary of the Invention

[0006] The purpose of this invention is to provide a computational spectrometer based on photochromic microcavity coupling and its spectral reconstruction method to solve the problems in the background technology. By controlling the spectral response characteristics through photochromism, the number of spectral channels can be effectively expanded, thereby improving the spectral resolution of the computational spectrometer.

[0007] To achieve the above objectives, the present invention provides a computational spectrometer based on photochromic microcavity coupling, comprising a photochromic Fabry-Perot microcavity filter array and a CMOS image camera; the photochromic Fabry-Perot microcavity filter array is composed of multiple filter units, and the structure of each filter unit from bottom to top is as follows: a polyethylene terephthalate (PET) substrate, a first silver film, a photochromic molecular layer, and a second silver film; Among them, the photochromic molecular layer consists of spiropyran molecules and polymethyl methacrylate. Different filter units adjust the microcavity resonance peak position by changing the thickness of the photochromic molecular layer, thus forming a filter array with different spectral response characteristics. A CMOS image camera is positioned on the light-emitting side of the photochromic Fabry-Perot microcavity filter array to receive and record the intensity of the light signal transmitted through the filter array.

[0008] Preferably, the thickness of the first silver film is 30±10 nm; the thickness of the second silver film is 30±10 nm; both the first and second silver films are prepared by pulsed DC reactive magnetron sputtering.

[0009] Preferably, in the photochromic molecular layer, the weight ratio of spiropyran molecules to polymethyl methacrylate is 3:2.

[0010] Preferably, in the photochromic molecular layer, the thickness of the photochromic molecular layer is 115nm~170nm by changing the spin coating speed, and the spin coating speed is 1300rpm~2100rpm.

[0011] This invention provides a spectral reconstruction method for a computational spectrometer based on photochromic microcavity coupling, comprising the following steps: Step 1: Spectral Calibration: The photochromic Fabry-Perot microcavity filter array was irradiated with monochromatic light in the wavelength range of 400nm~700nm. Ultraviolet light (UV=365nm) was applied to the filter array for different durations to modulate the coupling strength between the photochromic molecular transition frequencies and the microcavity mode frequencies. The transmission intensity of each filter unit under monochromatic light irradiation for different UV irradiation durations was read by a CMOS image camera to obtain the spectral response function matrix of the filter array. ,in, Indicates the first Duration of UV exposure ( =1,2,3 correspond to 0s, 6s, and 15s respectively. Indicates wavelength; Step 2, Unknown Spectrum Measurement and Reconstruction: The photochromic Fabry-Perot microcavity filter array is irradiated with unknown light. Ultraviolet light irradiation for the same duration as in Step 1 is applied to the filter array. The spectral response values ​​of each filter unit under unknown light irradiation for different ultraviolet light irradiation durations are read by a CMOS image camera. , This indicates the filter unit number; the spectral information of the unknown light is reconstructed using a compressed sensing algorithm.

[0012] Preferably, in step 1, the fabrication method of the photochromic Fabry-Perot microcavity filter array is as follows: 1) The first silver film was prepared on a polyethylene terephthalate (PET) substrate by pulsed DC reactive magnetron sputtering. 2) Then, a toluene solution containing the photochromic molecules spiropyran and polymethyl methacrylate is spin-coated onto a silver film at different rotation speeds to obtain a photochromic molecular layer; 3) Finally, a silver film is sputtered onto the photochromic molecular layer to form a photochromic microcavity filter, the structure of which is represented as PET / silver / photochromic molecular layer / silver.

[0013] Preferably, in step 1, the ultraviolet irradiation of different durations includes three durations: 0s, 6s, and 15s.

[0014] Preferably, the specific steps for reconstructing the spectral information of unknown light using compressed sensing algorithms are as follows: Using compressed sensing algorithms, based on the spectral response function matrix during the calibration process... And the spectral response values ​​of the filter array unit under unknown light irradiation corresponding to different ultraviolet light irradiation durations during the measurement process. An unknown incident light spectrum was obtained on a photochromic microcavity computational spectrometer. , is represented as: ; in, Quantum efficiency for CMOS image cameras; , These are the maximum wavelength and the minimum wavelength, respectively.

[0015] Preferably, the duration of ultraviolet irradiation is in three groups: 0s, 6s, and 15s.

[0016] Therefore, the present invention employs the above-mentioned computational spectrometer and spectral reconstruction method based on photochromic microcavity coupling, which has the following beneficial effects: (1) Non-contact dynamic control: The present invention uses ultraviolet light irradiation as a non-contact means to control the coupling strength between photochromic molecules and microcavities. It does not require the integration of complex electrode structures or temperature control devices, which is conducive to the further miniaturization of computational spectrometers and effectively solves the problems of existing electrochromic schemes relying on electrode structures and phase change material schemes relying on temperature control devices.

[0017] (2) Spectral channel number expansion: By applying ultraviolet light for different durations to the photochromic microcavity filter, each filter unit exhibits different spectral responses under different light durations, thereby expanding the number of spectral channels and effectively solving the trade-off between the number of spectral channels and physical size.

[0018] (3) Improved spectral resolution: This invention has achieved the reconstruction of unknown spectra in the visible light band of 400-700nm. Experimental results show that the mean square error of the reconstructed spectrum can be reduced from 1.5E-3 under a single set of illumination to 8.3E-4 under three sets of illumination, and the spectral resolution is significantly improved.

[0019] (4) Simple structure and low cost: The present invention uses pulsed DC reactive magnetron sputtering and spin coating to prepare photochromic microcavity filter arrays. The process is simple, compatible with existing CMOS image sensor manufacturing processes, and suitable for large-scale production.

[0020] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the photochromic microcavity device and filter array structure according to an embodiment of the present invention; Figure 2 This is a test optical path diagram of the photochromic computational spectrometer according to an embodiment of the present invention; Figure 3 This is a comparison of reconstructed spectra for different illumination groups according to an embodiment of the present invention; Figure 4 This describes the workflow of the computational spectrometer for photochromic microcavity coupling of the present invention. Detailed Implementation

[0022] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0024] Example 1: Fabrication of a photochromic microcavity filter array like Figure 1 As shown, the photochromic microcavity computational spectrometer of this embodiment consists of two parts: a photochromic Fabry-Perot microcavity filter array and a CMOS image camera. The photochromic Fabry-Perot microcavity filter array is positioned in front of the CMOS image camera, with each filter unit of the filter array corresponding one-to-one with a pixel area of ​​the CMOS image camera, enabling the light signal passing through the filter array to be received and recorded by the CMOS image camera.

[0025] The fabrication of a photochromic Fabry-Perot microcavity filter array includes the following steps: This embodiment provides a fabrication method for a photochromic microcavity filter array, such as... Figure 1 As shown, it includes the following steps: (1) A first silver film was prepared on a polyethylene terephthalate (PET) substrate by pulsed DC reactive magnetron sputtering. The sputtering power was 30 W and the sputtering time was 100 s, resulting in a silver film with a thickness of approximately 35 nm.

[0026] (2) A toluene solution containing the photochromic molecules spiropyran (SPI) and polymethyl methacrylate (PMMA) (SPI to PMMA weight ratio of 3:2) was spin-coated onto a silver film at different rotation speeds. By adjusting the spin-coating speed to a range of 1300 rpm, 1400 rpm, 1500 rpm, 1600 rpm, 1700 rpm, 1800 rpm, 1900 rpm, 2000 rpm, and 2100 rpm, photochromic molecular layer thicknesses of 115 nm, 121 nm, 127 nm, 137 nm, 141 nm, 151 nm, 154 nm, 165 nm, and 170 nm were obtained. Different filter units correspond to different molecular layer thicknesses, thereby achieving different microcavity resonant frequencies.

[0027] (3) A second silver film was deposited on the photochromic molecular layer using pulsed DC reactive magnetron sputtering. The sputtering power was 30W and the sputtering time was 100s, resulting in a silver film with a thickness of approximately 35nm.

[0028] The final photochromic microcavity filter unit structure is represented as: PET / silver / photochromic molecular layer / silver. Arranging filter units with different molecular layer thicknesses into an array yields a photochromic Fabry-Perot microcavity filter array.

[0029] (4) The photochromic Fabry-Perot microcavity filter array prepared in step (3) is directly attached to the photosensitive surface of the CMOS image camera, so that each filter unit in the filter array is spatially aligned with the pixel area of ​​the CMOS image camera, thus completing the fabrication of the photochromic microcavity computational spectrometer. The CMOS image camera is used to receive and record the intensity of the light signal transmitted through each unit of the filter array.

[0030] Example 2: Spectral Calibration like Figure 2 As shown, a xenon lamp is used as the light source, and a tunable monochromator generates monochromatic light with wavelengths tuned from 400 nm to 700 nm. The photochromic microcavity filter array prepared in Example 1 is placed in front of a CMOS image camera.

[0031] The filter array was irradiated with ultraviolet light (UV=365nm) for 0s, 6s, and 15s respectively to modulate the coupling strength between the photochromic molecule transition frequency and the microcavity mode frequency. The ultraviolet light was provided by a 365nm LED light source at a distance of 5cm and a power of 124.2mW. At each ultraviolet irradiation duration, the filter array was sequentially irradiated with monochromatic light ranging from 400nm to 700nm. The transmission intensity of each filter unit was read by a CMOS image camera. Duration of UV exposure ( ) and the Each filter unit ( Record its wavelength Transmission intensity at Therefore, the spectral response function matrix is ​​constructed. ,in, Indicates the first Duration of ultraviolet light exposure.

[0032] Example 3: Measurement and Reconstruction of Unknown Spectra A xenon lamp source generates the unknown light to be measured through a broadband filter, which is then incident on a photochromic microcavity filter array, with the optical path setup identical to that in Example 2. The photochromic microcavity filter array is irradiated with the unknown light, and ultraviolet light irradiation is applied to the filter array for 0s, 6s, and 15s respectively (irradiation conditions are exactly the same as in Example 2). At each ultraviolet irradiation duration, the transmission intensity of each filter unit under the unknown light irradiation is read by a CMOS image camera. ,in, This indicates the filter unit number.

[0033] The location spectrum is reconstructed using a compressed sensing algorithm. In this embodiment, the unknown spectrum is solved in reverse using the Tikhonov regularization method.

[0034] Specifically, based on the spectral response function matrix obtained during the calibration process and the response values ​​obtained during the measurement process Solve the following integral equation: ; ; in, For the unknown spectrum to be reconstructed, Quantum efficiency for CMOS image cameras; , These are the maximum wavelength and the minimum wavelength, respectively. It refers to the number of filter units.

[0035] First, a mathematical model is established to determine the response intensity of the spectrum received by the CMOS image camera during the unknown spectrum reconstruction process. With the unknown spectrum to be sought The following discretized linear equations are satisfied: Next, the regularization objective function is constructed and solved. The following Tikhonov regularization objective function is constructed: ; in, This is used to ensure the consistency between the reconstructed spectrum and the measured spectral response intensity. This is a regularization term used to suppress severe oscillations in the solution caused by noise.

[0036] Finally, here's a result example, measuring the difference between the reconstructed spectrum and the known spectrum using mean squared error. Measurements were performed using a standard light source with a known spectrum (from a commercial spectrometer) as the unknown light source, and the reconstructed spectrum was compared to the reference spectrum from the standard light source. For example... Figure 3 As shown, the experimental results indicate that the resolution of the reconstructed spectrum significantly improves with the increase in the number of UV illumination groups (from 1 group to 3 groups). When using three groups of UV illumination (0s, 6s, and 15s), the number of spectral channels is expanded to three times the number of filter units, and the mean square error of the reconstructed spectrum decreases from 1.5E-3 under single illumination to 8.3E-4 under three illumination groups.

[0037] like Figure 4 The diagram illustrates the complete signal processing flow from input light passing through a broadband filter, a photochromic microcavity filter array, and a CMOS detector array, to the output reconstructed spectrum from the compressed sensing reconstruction algorithm. Input light... First, an unknown incident spectrum is formed by passing the light through a broadband filter. It then enters the photochromic Fabry-Perot microcavity filter array. to Each filter unit in the filter array performs spectral encoding on the incident light under ultraviolet irradiation at 0s, 6s, and 15s, respectively. Each filter unit in the filter array corresponds one-to-one with a pixel area of ​​the CMOS image camera, ensuring that the light signal passing through the filter array can be received and recorded by the CMOS image camera. The encoded light signal is then processed by the corresponding pixel of the CMOS image camera. to Receive and output response values to Finally, based on the spectral response function matrix obtained during the calibration phase, the reconstructed spectrum is output by the compressed sensing reconstruction algorithm. .

[0038] Through the above Examples 1 to 3, the entire process of fabrication, calibration and spectral reconstruction of a computational spectrometer based on photochromic microcavity coupling was completed, verifying the feasibility of the present invention in expanding the number of spectral channels and improving spectral resolution through non-contact light modulation.

[0039] Therefore, this invention provides a computational spectrometer and its spectral reconstruction method based on photochromic microcavity coupling. Using 365nm ultraviolet irradiation time as a non-contact control method, it precisely drives the reversible transition of spiropyran molecules between the transparent and absorption states, thereby achieving optical control of the coupling strength between molecular transition frequencies and microcavity modes. Furthermore, by applying three sets of ultraviolet irradiation times (0s, 6s, and 15s) to each filter unit, the same physical unit exhibits independent spectral response characteristics under different illumination conditions, expanding the effective spectral channel count to three times the number of filter units. Without adding any electrode structures, voltage-driven circuits, or temperature control devices, it achieves spatial gain through multiple measurements in the time dimension. The increased number of channels in the dimension completely breaks through the rigid constraint of chip area on spectral resolution. In addition, through the fusion reconstruction of compressed sensing algorithm, the mean square error of the reconstructed spectrum is significantly reduced from 1.5E-3 under single-set illumination to 8.3E-4 under three-set illumination, and the spectral resolution is improved to 3 times the original level. This invention uses the layered structure of photochromic microcavity as the physical platform, ultraviolet irradiation time as the control means, and compressed sensing algorithm as the reconstruction engine. The technical features of the three levels work together and are indispensable to achieve the overall technical effect of simplifying device structure, expanding spectral channels and improving resolution. It has broad application prospects in portable spectral detection, environmental monitoring, biomedical sensing, food safety detection and other fields.

[0040] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A computational spectrometer based on photochromic microcavity coupling, characterized in that: It includes a photochromic Fabry-Perot microcavity filter array and a CMOS image camera; the photochromic Fabry-Perot microcavity filter array is composed of multiple filter units, and the structure of each filter unit from bottom to top is as follows: a polyethylene terephthalate substrate, a first silver film, a photochromic molecular layer, and a second silver film; Among them, the photochromic molecular layer consists of spiropyran molecules and polymethyl methacrylate. Different filter units adjust the microcavity resonance peak position by changing the thickness of the photochromic molecular layer, thus forming a filter array with different spectral response characteristics. A CMOS image camera is positioned on the light-emitting side of the photochromic Fabry-Perot microcavity filter array to receive and record the intensity of the light signal transmitted through the filter array.

2. A computational spectrometer based on photochromic microcavity coupling according to claim 1, characterized in that: The thickness of the first silver film is 30±10 nm; the thickness of the second silver film is 30±10 nm; both the first and second silver films are prepared by pulsed DC reactive magnetron sputtering.

3. A computational spectrometer based on photochromic microcavity coupling according to claim 1, characterized in that: In the photochromic molecular layer, the weight ratio of spiropyran molecules to polymethyl methacrylate is 3:

2.

4. A computational spectrometer based on photochromic microcavity coupling according to claim 1, characterized in that: In the photochromic molecular layer, the thickness of the photochromic molecular layer is 115nm~170nm by changing the spin coating speed, and the spin coating speed is 1300rpm~2100rpm.

5. A method for spectral reconstruction of a computational spectrometer based on photochromic microcavity coupling, characterized in that, Using the computational spectrometer as described in any one of claims 1-4, the method includes the following steps: Step 1: Spectral Calibration: The photochromic Fabry-Perot microcavity filter array was irradiated with monochromatic light in the wavelength range of 400nm~700nm. Different durations of ultraviolet light irradiation were applied to the filter array to modulate the coupling strength between the photochromic molecular transition frequencies and the microcavity mode frequencies. The transmission intensity of each filter unit under monochromatic light irradiation for different durations of ultraviolet light irradiation was read by a CMOS image camera to obtain the spectral response function matrix of the filter array. ,in, Indicates the first Duration of ultraviolet light exposure, Indicates wavelength; Step 2, Unknown Spectrum Measurement and Reconstruction: The photochromic Fabry-Perot microcavity filter array is irradiated with unknown light. Ultraviolet light irradiation for the same duration as in Step 1 is applied to the filter array. The spectral response values ​​of each filter unit under unknown light irradiation for different ultraviolet light irradiation durations are read by a CMOS image camera. , This indicates the filter unit number; the spectral information of the unknown light is reconstructed using a compressed sensing algorithm.

6. The spectral reconstruction method according to claim 5, characterized in that: In step 1, the fabrication method of the photochromic Fabry-Perot microcavity filter array is as follows: 1) The first silver film was prepared on a polyethylene terephthalate (PET) substrate by pulsed DC reactive magnetron sputtering. 2) Then, a toluene solution containing the photochromic molecules spiropyran and polymethyl methacrylate is spin-coated onto a silver film at different rotation speeds to obtain a photochromic molecular layer; 3) Finally, a silver film is sputtered onto the photochromic molecular layer to form a photochromic microcavity filter, the structure of which is represented as PET / silver / photochromic molecular layer / silver.

7. The spectral reconstruction method according to claim 5, characterized in that: In step 1, the ultraviolet irradiation of different durations includes three durations: 0s, 6s, and 15s.

8. The spectral reconstruction method according to claim 7, characterized in that: The specific steps for reconstructing the spectral information of unknown light using compressed sensing algorithms are as follows: Based on the spectral response function matrix during the calibration process And the spectral response values ​​of the filter array unit under unknown light irradiation corresponding to different ultraviolet light irradiation durations during the measurement process. An unknown incident light spectrum was obtained on a photochromic microcavity computational spectrometer. , represented as: ； in, Quantum efficiency for CMOS image cameras; , These are the maximum wavelength and the minimum wavelength, respectively.

Images