Long-wave infrared tunable optical analog computing device and spatial frequency regulation method thereof

Abstract

The application discloses a long-wave infrared tunable optical analog computing device and a spatial frequency regulation method thereof, and belongs to the technical field of long-wave infrared optical image processing. The long-wave infrared tunable optical analog computing device comprises, in sequence along an optical path, an upper distributed Bragg reflector, a liquid crystal cavity layer and a lower distributed Bragg reflector, and forms a tunable Fabry-Perot resonant cavity. The effective refractive index of the liquid crystal cavity layer is adjusted by an external electric field, the optical thickness and the phase matching condition of the resonant cavity are changed, the transmission peak or the stop band is dynamically moved in the spatial frequency domain, and thus switchable spatial frequency transfer functions such as low-pass, high-pass, band-pass or band-stop are obtained. The application utilizes the mapping relationship between the angle transmission spectrum and the spatial frequency, directly realizes the processing of long-wave infrared image in the process of light transmission, and has the advantages of compact structure, tunability, multi-mode switching and easy integration in a transmission type infrared imaging system.

Description

Technical Field

[0001] This invention belongs to the field of long-wave infrared optical image processing technology, specifically relating to long-wave infrared tunable optical analog computing devices and their spatial frequency modulation methods. Background Technology

[0002] Long-wave infrared imaging technology has important applications in night vision detection, thermal target identification, remote sensing, industrial inspection and intelligent transportation. It mainly acquires the thermal radiation information of objects themselves and can work under conditions such as night, low light, smoke and complex weather, with all-weather detection capability.

[0003] However, infrared images reflect the thermal radiation distribution of the target and background, not the reflective information of the object's surface. Affected by factors such as thermal diffusion, detector resolution and noise, and background thermal radiation, long-wave infrared images typically suffer from low contrast, blurred edges, insufficient texture detail, and significant noise interference. These problems reduce the distinguishability between the target and the background, thus affecting the accuracy of subsequent target detection, recognition, and image understanding.

[0004] Existing long-wave infrared image processing methods primarily rely on back-end electronic algorithms for noise reduction and edge detection. These methods typically involve photoelectric conversion, digital sampling, data transmission, and computational processing, making them susceptible to limitations in processing latency, data throughput, and power consumption in real-time sensing systems. In contrast, analog optical computing can utilize physical processes such as light propagation, interference, diffraction, and structural dispersion to directly manipulate the spatial frequency information of the image during light transmission, offering inherent potential for parallelism, high speed, and low power consumption.

[0005] Traditional optical image processing systems often employ 4f systems and Fourier surface filters to achieve spatial frequency modulation. However, these systems typically require multiple lenses and precise Fourier surface adjustments, hindering compactness and integration. In recent years, metasurfaces, multilayer thin films, and angle-sensitive photonic structures have offered new solutions for miniaturized optical image processing. Among these, angle-sensitive filters can establish a correspondence between the incident angle and the spatial frequency of the image. For incident light in air, the normalized transverse wave vector satisfies k... x Since / k0=sinθ, the angular transmission spectrum of the device can be mapped to the optical transfer function in the spatial frequency domain, thereby achieving selective transmission or suppression of different spatial frequency components.

[0006] Despite progress in optical spatial differentiators and edge detection devices, existing technologies still have the following shortcomings: First, most devices operate primarily in the visible, near-infrared, or communication bands, and compact devices suitable for long-wave infrared image preprocessing remain relatively limited; second, some devices rely on reflective structures or specific polarization states, making them difficult to integrate directly into conventional transmissive infrared imaging systems; third, many devices have relatively simple functions, typically only capable of fixed edge detection, high-pass filtering, or specific convolution operations, making it difficult to simultaneously meet requirements such as noise suppression, target edge detection, and multi-scale feature selection.

[0007] Therefore, it is necessary to propose an optical front-end preprocessor and its control method that is oriented towards the long-wave infrared band, has a compact transmissive structure, and supports multi-functional switching. Summary of the Invention

[0008] To address the shortcomings of existing technologies, this invention provides a long-wave infrared tunable optical analog computing device and its spatial frequency modulation method. This method constructs a multilayer resonant structure using upper and lower distributed Bragg mirrors (DBRs) and a central liquid crystal cavity. The DBRs provide angle-dependent reflection phase and Fabry-Perot resonance feedback, and the effective refractive index variation of the liquid crystal layer under different orientation states enables the tunable design of the spatial frequency transfer function. By presetting the DBR structural parameters, the basic position, bandwidth, and shape of the transmission peak or stopband in k-space are determined. The effective refractive index of the liquid crystal cavity layer is changed by controlling the liquid crystal orientation state through an electric field, allowing the transmission peak or stopband to move or reconstruct within a preset range, thereby achieving tunable switching between spatial frequency response modes such as low-pass, high-pass, band-pass, and band-stop.

[0009] This invention is achieved through the following technical solution:

[0010] In a first aspect, the present invention provides a method for spatial frequency modulation of a long-wave infrared tunable optical analog computing device, comprising:

[0011] A long-wave infrared optical simulation computing device is established; the device includes an upper reflective layer 1, a liquid crystal cavity layer 2 and a lower reflective layer 3 in sequence along the incident light propagation direction; the upper reflective layer 1 and the lower reflective layer 3 are both distributed Bragg mirrors, each formed by a periodic alternation of a first dielectric layer 4 and a second dielectric layer 5; the liquid crystal cavity layer 2 is disposed between the upper reflective layer 1 and the lower reflective layer 3 to form a tunable Fabry-Perot type resonant cavity.

[0012] By adjusting the effective refractive index of the liquid crystal cavity layer 2 with an external electric field, the optical thickness and phase matching conditions of the tunable Fabry-Perot type resonant cavity are changed, causing the position of the transmission peak or stopband in k-space to move dynamically, thereby obtaining different spatial frequency transfer functions. Using different spatial frequency transfer functions, different spatial frequency components in long-wave infrared images are selectively transmitted, suppressed, or enhanced.

[0013] Furthermore, the liquid crystal cavity layer 2 is composed of a liquid crystal material with electrically controllable refractive index tuning capability. The liquid crystal material is E7 liquid crystal, and its effective refractive index is n=1.499 in the unenergized state. In the energized state, the orientation of the liquid crystal molecules changes, and the effective refractive index becomes n=1.693.

[0014] Furthermore, the first dielectric layer 4 is a low refractive index dielectric layer, including SiO2, MgF2 or Al2O3; the second dielectric layer 5 is a high refractive index dielectric layer, including TiO2, Si, Ge, ZnS or ZnSe; at the target operating wavelength, the refractive index difference between the second dielectric layer 5 and the first dielectric layer 4 is not less than 0.2.

[0015] Furthermore, the thickness of the first dielectric layer 4 and the second dielectric layer 5 is both 0.05 μm to 10 μm; the number of periods of the upper reflective layer 1 and the lower reflective layer 3 is both 1 to 15; and the thickness of the liquid crystal cavity layer 2 is 1 μm to 50 μm.

[0016] Furthermore, the spatial frequency transfer function includes a low-pass filter transfer function, a high-pass filter transfer function, a band-stop filter transfer function, or a band-pass filter transfer function, which are used to achieve denoising processing, edge enhancement processing, noise suppression processing of specific frequency bands, or target enhancement processing of specific scales in long-wave infrared images, respectively.

[0017] Secondly, the present invention provides a design method for the long-wave infrared optical analog computing device, comprising:

[0018] Step 1: Establish a multilayer structure model of the distributed Bragg mirror-liquid crystal cavity (DBR-LC), and set the materials and thicknesses of the first dielectric layer 4 and the second dielectric layer 5, the number of periods of the upper reflective layer 1 and the lower reflective layer 3, the thickness of the liquid crystal cavity layer 2, and the normalized transverse wave vector k. x The calculation range of / k0; where k0 is the wavenumber of the incident light in vacuum, k x This represents the transverse component of the incident light wave vector along the direction of the device surface;

[0019] Step 2: Calculate the effective refractive index of liquid crystal cavity layer 2 under different electric field control states, substitute it into the distributed Bragg reflector-liquid crystal cavity multilayer structure model, and use the transfer matrix method to calculate the transmittance under different wavelengths, different incident angles and different polarization conditions to obtain the angular transmission spectrum of the distributed Bragg reflector-liquid crystal cavity multilayer structure under different liquid crystal electronic control states.

[0020] Step 3: Based on the wave vector mapping relationship k between the normalized transverse wave vector and the incident angle x / k0=sinθ, mapping the angular transmission spectrum to a spatial frequency transfer function, obtaining the current position of the transmission peak or stopband in the spatial frequency domain; where θ is the incident angle of the incident light relative to the device normal direction. By adjusting the layer thickness of the first dielectric layer 4 and the second dielectric layer 5, the number of periods of the upper reflective layer 1 and the lower reflective layer 3, and the effective refractive index of the liquid crystal cavity layer 2, the position, bandwidth, and shape of the transmission peak or stopband in the spatial frequency domain are adjusted until the spatial frequency response meets the design requirements, thereby forming a low-pass, high-pass, band-pass, or band-stop spatial frequency transfer response;

[0021] Step 4: Rotate and expand the obtained spatial frequency transfer function along the azimuth direction to generate a two-dimensional spatial frequency transfer function. Multiply it with the Fourier spectrum of the input image and then perform an inverse Fourier transform to obtain the processed image. If the image denoising, edge enhancement, or scale selection functions do not achieve the expected results, return to Step 2 or Step 3 to readjust the parameters until the verification is qualified.

[0022] Step 5: Prepare a long-wave infrared tunable optical analog computing device according to the verified parameters.

[0023] Thirdly, the present invention provides a long-wave infrared tunable optical analog computing device, which includes an upper reflective layer 1, a liquid crystal cavity layer 2, and a lower reflective layer 3 sequentially along the incident light propagation direction; the upper reflective layer 1 is an upper distributed Bragg mirror, and the lower reflective layer 3 is a lower distributed Bragg mirror; the liquid crystal cavity layer 2 is disposed between the upper reflective layer 1 and the lower reflective layer 3 to form a tunable Fabry-Perot resonant cavity; the upper reflective layer 1 and the lower reflective layer 3 are both formed by periodically alternating arrangement of a first dielectric layer 4 and a second dielectric layer 5.

[0024] Furthermore, the lower reflective layer 3 is disposed on a fused silica substrate; the first dielectric layer 4 is a low refractive index dielectric layer, and the second dielectric layer 5 is a high refractive index dielectric layer.

[0025] Furthermore, the target operating wavelength of the device is 10.6 μm, the first dielectric layer 4 is a SiO2 layer, the second dielectric layer 5 is a TiO2 layer, the upper reflective layer 1 and the lower reflective layer 3 each have 8 pairs of periods, the TiO2 layer thickness is 1100 nm, the liquid crystal cavity layer 2 thickness is 10 μm, and the SiO2 layer thickness is 1550 nm, 1580 nm, or 1600 nm. By electronically adjusting the effective refractive index of the liquid crystal cavity layer 2, the device switches between low-pass noise reduction mode, high-pass edge extraction mode, band-stop suppression mode, or band-pass scale selection mode.

[0026] The main principle of this invention is as follows: Different spatial frequency components in long-wave infrared images are equivalent to light wave components with different incident angles. The transmittance of the multilayer structure under different wavelengths, incident angles, and polarization conditions is calculated using the transfer matrix method, and based on k... x / k0=sinθ maps the angular transmission spectrum to a spatial frequency transfer function. Furthermore, a two-dimensional spatial frequency filter is generated by rotating the one-dimensional angular transmission curve around the frequency domain center and applied to the Fourier spectrum of the image, thereby verifying the function of this structure in image denoising, edge enhancement, and scale selection. To further illustrate the correspondence between the angular transmission spectrum and the image spatial frequency, the input image light field can be represented as the superposition of different transverse wave vector components. Input light field The frequency domain form can be expressed as:

[0027]

[0028] in, Let x and y be the spectral distribution of the input image in the spatial domain, and k be the spatial coordinates. x and k y E represents the transverse wave vector components along the x and y directions, respectively. in (x, y) represents the optical field distribution of the input light field in the spatial frequency domain, where x and y are spatial coordinates. When this light field passes through an angle-selective DBR-LC multilayer structure, different spatial frequency components will be transmitted or suppressed to varying degrees. Therefore, the output spectrum can be expressed as:

[0029]

[0030] in, To define the spectral distribution of the output light field in the spatial frequency domain, H(k x ,k y ) is the optical spatial frequency transfer function corresponding to the DBR-LC structure, used to characterize the transmission characteristics of different spatial frequency components in the structure.

[0031] Unlike non-volatile optical computing devices based on phase change materials, this invention achieves continuous and adjustable spatial frequency response through liquid crystal electronic modulation, without relying on the material phase change process. At the same time, this invention is aimed at long-wave infrared imaging front-end systems, realizing multi-scale spatial frequency selective modulation, rather than fixed-state optical computing function switching.

[0032] Compared with the prior art, the advantages of the present invention are as follows:

[0033] 1. This invention employs a distributed Bragg mirror-liquid crystal cavity (DBR-LC) structure. Through the synergistic effect of DBR layer thickness design and electrically controlled refractive index adjustment of the liquid crystal cavity layer, the long-wave infrared spatial frequency transfer function can be designed with adjustable precision. Specifically, the DBR dielectric layer thickness, number of periods, and material refractive index are used to adjust the angle-dependent reflection phase of the upper and lower reflective layers, thereby pre-setting the position, bandwidth, and shape of the k-space transmission window. The liquid crystal cavity layer dynamically adjusts the transmission window by changing its effective refractive index through an applied electric field. Therefore, this invention does not simply rely on material phase changes to achieve fixed two-state filtering, but rather achieves adjustable control of the spatial frequency response through a combination of pre-set structural parameters and electrically controlled refractive index adjustment.

[0034] 2. This invention enables multi-mode spatial frequency modulation of long-wave infrared images within the same DBR-LC structural framework. Different basic spatial frequency responses can be obtained through different DBR layer thickness designs; further, the conversion between low-pass, high-pass, band-stop, and band-pass transmission responses can be achieved through electronic control adjustment between the unpowered and powered states of the liquid crystal cavity layer. Therefore, this invention not only achieves image denoising and edge enhancement, but also enables noise suppression in specific frequency bands, target structure enhancement, and scale-selective extraction of structures from hundreds of micrometers to sub-millimeter scales, making it suitable for multi-scale front-end preprocessing of long-wave infrared images.

[0035] 3. This invention is designed for long-wave infrared image processing scenarios, preferably operating in the long-wave infrared band, and can be directly used for optical front-end preprocessing in infrared imaging systems. Compared to traditional infrared image processing methods that rely on back-end electronic algorithms, this invention can complete spatial frequency modulation during light propagation, thereby reducing the subsequent electronic computational burden. Compared to traditional 4f optical systems, this invention does not require multiple lenses and independent Fourier surface filters, resulting in a more compact structure and easier integration with transmissive infrared imaging systems. Attached Figure Description

[0036] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0037] Figure 1 This is a schematic diagram of the structure of the long-wave infrared tunable optical analog computing device described in this invention;

[0038] Figure 2 This is a schematic diagram of the structure of the distributed Bragg mirror-liquid crystal cavity (DBR-LC) structure under different electrical control states described in this invention;

[0039] Wherein, (a) is a schematic diagram of the DBR-LC structure in the liquid crystal unpowered state; (b) is a schematic diagram of the DBR-LC structure in the liquid crystal powered state;

[0040] Figure 3 This invention presents the two-dimensional k-space response diagram and spatial frequency transfer function diagram of the DBR-LC structure under s-polarization and p-polarization incident conditions when the SiO2 layer thickness is set to 1550 nm, the liquid crystal is in an unenergized state, and the effective refractive index n=1.499.

[0041] Among them, (a) is the two-dimensional k-space response diagram under s-polarization condition; (b) is the two-dimensional k-space response diagram under p-polarization condition; (c) is the spatial frequency transfer function diagram, with the blue curve representing s-polarization and the red curve representing p-polarization.

[0042] Figure 4 This is an image denoising result of the DBR-LC structure when the SiO2 layer thickness is set to 1550 nm, the liquid crystal is in an unpowered state, and the effective refractive index n=1.499.

[0043] Wherein, (a) is the original image, and (b) is the image after denoising;

[0044] Figure 5 This invention presents the two-dimensional k-space response diagram and spatial frequency transfer function diagram of the DBR-LC structure under s-polarization and p-polarization incident conditions when the SiO2 layer thickness is set to 1550 nm, the liquid crystal is in an energized state, and the effective refractive index n=1.693.

[0045] Among them, (a) is the two-dimensional k-space response diagram under s-polarization condition; (b) is the two-dimensional k-space response diagram under p-polarization condition; (c) is the spatial frequency transfer function diagram, with the blue curve representing s-polarization and the red curve representing p-polarization.

[0046] Figure 6 This is an image edge extraction result of the DBR-LC structure when the SiO2 layer thickness is set to 1550 nm, the liquid crystal is in an energized state, and the effective refractive index n=1.693.

[0047] Wherein, (a) is the original image; (b) is the image edge extraction result;

[0048] Figure 7 This invention presents the two-dimensional k-space response diagram and spatial frequency transfer function diagram of the DBR-LC structure under s-polarization and p-polarization incident conditions when the SiO2 layer thickness is set to 1580 nm, the liquid crystal is in an unenergized state, and the effective refractive index n=1.499.

[0049] Among them, (a) is the two-dimensional k-space response diagram under s-polarization condition; (b) is the two-dimensional k-space response diagram under p-polarization condition; (c) is the spatial frequency transfer function diagram, with the blue curve representing s-polarization and the red curve representing p-polarization.

[0050] Figure 8 The images show the band-stop denoising results of the DBR-LC structure when the SiO2 layer thickness is set to 1580 nm, the liquid crystal is in an unpowered state, and the effective refractive index n=1.499, as well as a comparison of the low-pass denoising and band-stop denoising effects under small-scale image conditions.

[0051] Among them, (ab) is the result of band-stop denoising for simple noisy targets; (cd) is the result of band-stop denoising for complex grayscale images; (eg) is the comparison result of band-stop denoising and low-pass denoising.

[0052] Figure 9 This invention presents the two-dimensional k-space response diagram and spatial frequency transfer function diagram of the DBR-LC structure under s-polarization and p-polarization incident conditions, with the SiO2 layer thickness set to 1580 nm, the liquid crystal in an energized state, and the effective refractive index n=1.693.

[0053] Among them, (a) is the two-dimensional k-space response diagram under s-polarization condition; (b) is the two-dimensional k-space response diagram under p-polarization condition; (c) is the spatial frequency transfer function diagram, with the blue curve representing s-polarization and the red curve representing p-polarization.

[0054] Figure 10 This is an image of the edge extraction result of the DBR-LC structure under the conditions of SiO2 layer thickness set to 1580nm, liquid crystal in the energized state and effective refractive index n=1.693.

[0055] Wherein, (a) is the original image; (b) is the image edge extraction result;

[0056] Figure 11 This invention presents the two-dimensional k-space response diagram and spatial frequency transfer function diagram of the DBR-LC structure under s-polarization and p-polarization incident conditions, with the SiO2 layer thickness set to 1600 nm, the liquid crystal in an unenergized state, and the effective refractive index n=1.499.

[0057] Among them, (a) is the two-dimensional k-space response diagram under s-polarization condition; (b) is the two-dimensional k-space response diagram under p-polarization condition; (c) is the spatial frequency transfer function diagram, with the blue curve representing s-polarization and the red curve representing p-polarization.

[0058] Figure 12 This invention presents the two-dimensional k-space response diagram and spatial frequency transfer function diagram of the DBR-LC structure under s-polarization and p-polarization incident conditions when the SiO2 layer thickness is set to 1600 nm, the liquid crystal is in an energized state, and the effective refractive index n=1.693.

[0059] Among them, (a) is the two-dimensional k-space response diagram under s-polarization condition; (b) is the two-dimensional k-space response diagram under p-polarization condition; (c) is the spatial frequency transfer function diagram, with the blue curve representing s-polarization and the red curve representing p-polarization.

[0060] Figure 13 This is a comparison of the image edge extraction effect of the liquid crystal in the unpowered state and the powered state when the SiO2 layer thickness is set to 1600nm.

[0061] Wherein, (ac) is the low-frequency and high-frequency bandpass edge extraction result image of the resolution target; (df) is the processing result image of the circular resolution test image under different bandpass responses; different radius positions correspond to different spatial frequencies, and the two bandpass modes select different frequency band ring texture information respectively; the low-frequency bandpass is in the unpowered state, and the high-frequency bandpass is in the powered state. Detailed Implementation

[0062] To clearly and completely describe the technical solution and its specific working process of the present invention, the specific embodiments of the present invention are as follows, in conjunction with the accompanying drawings:

[0063] Example 1

[0064] This embodiment is used to illustrate the structure of the long-wave infrared optical analog computing device and its spatial frequency modulation method according to the present invention.

[0065] like Figure 1 As shown, this embodiment provides a long-wave infrared optical analog computing device, which includes an upper reflective layer 1, a liquid crystal cavity layer 2, and a lower reflective layer 3 sequentially along the incident light propagation direction. The lower reflective layer 3 is disposed on a fused silica substrate; both the upper reflective layer 1 and the lower reflective layer 3 are distributed Bragg mirrors (DBRs), each independently formed by a periodically alternating arrangement of a first dielectric layer 4 and a second dielectric layer 5; the first dielectric layer 4 is a low refractive index dielectric layer, and the second dielectric layer 5 is a high refractive index dielectric layer; the liquid crystal cavity layer 2 is disposed between the upper reflective layer 1 and the lower reflective layer 3 to form a tunable Fabry-Perot type resonant cavity.

[0066] In this embodiment, the target operating wavelength is 10.6 μm. Both the upper reflective layer 1 and the lower reflective layer 3 have 8 pairs of periods. The first dielectric layer 4 is a SiO2 layer, and the second dielectric layer 5 is a TiO2 layer with a thickness of 1100 nm and a SiO2 layer thickness of 1550 nm. The liquid crystal cavity layer 2 is an E7 liquid crystal layer with a thickness of 10 μm.

[0067] The spatial frequency modulation method provided in this embodiment includes: providing the above-mentioned long-wave infrared optical simulation computing device; adjusting the effective refractive index of the liquid crystal cavity layer 2 by applying an external electric field, changing the optical thickness and phase matching conditions of the tunable Fabry-Perot type resonant cavity, so that the transmission window moves dynamically in k-space, obtaining different spatial frequency transfer functions, and realizing selective transmission, suppression or enhancement of different spatial frequency components.

[0068] The effective refractive index of the E7 liquid crystal layer material is n=1.499 in the unenergized state; when energized, the orientation of the liquid crystal molecules changes, and the effective refractive index becomes n=1.693. By controlling the orientation of the liquid crystal molecules with an external electric field, the effective refractive index of the liquid crystal cavity layer 2 can be varied between different electrically controlled orientation states.

[0069] like Figure 2 As shown, when the liquid crystal is in an unpowered state (n=1.499), the DBR-LC structure mainly retains low spatial frequency components and suppresses some high spatial frequency components, exhibiting a low-pass filter response, which can be used for long-wave infrared image denoising. When the liquid crystal is powered on (n=1.693), the change in the effective refractive index of the liquid crystal causes the resonant transmission region of the DBR-LC cavity to shift towards a larger k... x The / k0 region shifts, low spatial frequency regions are suppressed, and higher spatial frequency components are enhanced, exhibiting a high-pass or edge-enhancing filter response, which can be used to extract edge information in images.

[0070] By changing the intensity of the applied electric field, the effective refractive index of the liquid crystal cavity layer 2 can be continuously varied between n=1.499 and n=1.693. The spatial frequency transfer function can be continuously adjusted with the change of the effective refractive index, realizing arbitrary transition response between low-pass filtering and high-pass filtering.

[0071] The above-mentioned control process can be completed without changing the incident light wavelength and incident light polarization state, and the switching between different spatial frequency transfer functions can be achieved solely through electronic control.

[0072] The image processing mechanism of the device described in this embodiment is as follows: different spatial frequency components in the long-wave infrared image are equivalent to light wave components with different incident angles, and the normalized transverse wave vector is used to process the image based on the relationship k between the incident angle and the normalized transverse wave vector. x / k0=sinθ, mapping the angular transmission spectrum of the device to a spatial frequency transfer function; after the spatial frequency transfer function is applied to the Fourier spectrum of the input image, the processed long-wave infrared image is obtained through inverse Fourier transform, thereby realizing the front-end modulation of spatial frequency information at different scales.

[0073] The operating band of the device described in this embodiment includes the mid-infrared to long-wave infrared band, preferably the long-wave infrared band. The spatial frequency response is determined by normalizing the transverse wave vector k. x / k0 represents the character. Where k x / k0 and the incident angle θ satisfy k x / k0=sinθ. The angular transmission spectrum is obtained by calculation using the transfer matrix method, taking into account the transmittance under s-polarization and p-polarization respectively during the calculation.

[0074] The spatial frequency selectivity of the DBR-LC structure described in this embodiment originates from the angle-dependent phase matching of the multilayer resonant cavity. For a Fabry-Perot type resonant cavity composed of an upper reflective layer 1, a liquid crystal cavity layer 2, and a lower reflective layer 3, its phase matching condition can be approximately expressed as:

[0075]

[0076] Where λ is the incident light wavelength, n is the effective refractive index of the liquid crystal cavity layer, d is the thickness of the liquid crystal cavity layer, and θ cav The angle of light propagation within the liquid crystal cavity layer 2. and The reflection phases introduced by the upper reflective layer 1 and the lower reflective layer 3 are respectively. m is the Fabry-Perot resonance order, and Z represents the set of integers, i.e., m is an integer. From this equation, it can be seen that the thickness, number of periods, and refractive index of the first dielectric layer 4 and the second dielectric layer 5 affect the reflection phases of the upper reflective layer 1 and the lower reflective layer 3, thereby determining the initial position and bandwidth of the cavity transmission window in k-space. The effective refractive index change of the liquid crystal cavity layer 2 alters the intracavity optical path, causing the transmission window to dynamically move or reconstruct in k-space, obtaining different types of spatial frequency transfer functions.

[0077] The device described in this embodiment is used for optical front-end preprocessing of long-wave infrared images. By combining the DBR layer thickness design with the electrically controlled refractive index adjustment of the liquid crystal cavity layer 2, it can form transmission responses such as low spatial frequency preservation, high spatial frequency enhancement, specific spatial frequency band suppression, or specific spatial frequency band transmission, thereby realizing various image processing modes such as low-pass noise reduction, high-pass edge enhancement, band-stop noise suppression, and band-pass scale selection. In one specific embodiment, when the low refractive index dielectric layer is SiO2 and its layer thickness varies in the range of 1550nm to 1600nm, the position and shape of the k-space transmission window of the DBR-LC structure change accordingly; by electrically controlling the liquid crystal cavity layer 2 between the unpowered and powered states, the conversion between different spatial frequency response modes can be further realized. The device can selectively suppress or enhance noise, edges, or target structures in long-wave infrared images at the scale of approximately hundreds of micrometers to sub-millimeters; the typical processing scale can cover the range of approximately 80μm to 230μm, which can respectively correspond to the extraction of fine edge structures, enhancement of targets at specific scales, and suppression of noise or texture interference at the scale of approximately 0.2mm.

[0078] Example 2

[0079] This embodiment provides a design method for the long-wave infrared optical analog computing device, including the following steps:

[0080] Step 1: Establish a distributed Bragg reflector-liquid crystal cavity multilayer structure model. Based on the target operating wavelength of 10.6μm and the target spatial frequency transfer function, set the materials and thicknesses of the first dielectric layer 4 and the second dielectric layer 5, the number of periods of the upper reflective layer 1 and the lower reflective layer 3, as well as the thickness of the liquid crystal cavity layer 2 and the spatial frequency calculation range.

[0081] In this embodiment, the first dielectric layer 4 is a SiO2 layer, the second dielectric layer 5 is a TiO2 layer, the upper reflective layer 1 and the lower reflective layer 3 both have 8 pairs of periods, the TiO2 layer thickness is 1100nm, and the liquid crystal cavity layer 2 thickness is 10μm. The SiO2 layer thickness is set to three schemes: 1550nm, 1580nm, and 1600nm.

[0082] Step 2: Calculate the angular transmission spectrum of the DBR-LC multilayer structure under different electrical control states. For the high and low refractive index dielectric materials in the DBR, their complex refractive index at the target wavelength can be read from a material database. For example... Figure 2 As shown, for liquid crystal cavity layer 2, its effective refractive index under different orientation states is calculated using a liquid crystal refractive index model. The liquid crystal orientation state refers to the orientation relationship between the long axis of the liquid crystal molecules and the device normal direction and the polarization direction of the incident light.

[0083] The effective refractive index of liquid crystal cavity layer 2 under different electric field modulation states was calculated. In this embodiment, liquid crystal cavity layer 2 uses E7 liquid crystal material. The effective refractive index of E7 liquid crystal material is n=1.499 in the unenergized state and n=1.693 in the energized state. Substituting the effective refractive index into the multilayer structure model, the transmittance under different wavelengths, different incident angles, and different polarization conditions (s-polarization and p-polarization) was calculated using the transfer matrix method (implemented on the MATLAB platform), obtaining the angular transmission spectrum of the DBR-LC multilayer structure under different liquid crystal states.

[0084] Step 3: Map the angular transmission spectrum to a spatial frequency transfer function: based on the mapping relationship between the normalized transverse wave vector and the incident angle k x / k0=sinθ, mapping the angular transmission spectrum obtained in step two to the spatial frequency transfer function in the spatial frequency domain. Extracting the cross-section corresponding to the central working wavelength from the two-dimensional transmission spectrum, we obtain the transmittance as a function of k. x The curve representing the variation of / k0 is the one-dimensional spatial frequency transfer function of the device at that operating wavelength. Under air incidence conditions, k x / k0=sinθ represents different normalized lateral spatial frequencies corresponding to different incident angles. Therefore, the angular transmission response of the device can be used to characterize its selective transmission, suppression, or enhancement of image spatial frequency components. By comparing the spatial frequency transfer function under different SiO2 layer thicknesses and different liquid crystal control states, the position, bandwidth, and shape of the transmission peak or stopband in the spatial frequency domain can be obtained. By adjusting or optimizing the layer thickness of the first dielectric layer 4 and the second dielectric layer 5, the number of periods of the upper reflective layer 1 and the lower reflective layer 3, and combining the effective refractive index adjustment of the liquid crystal cavity layer 2, the spatial frequency response can meet the design requirements, thereby determining the operating state of the device in low-pass, high-pass, band-pass, or band-stop modes.

[0085] Step 4: Construct a two-dimensional frequency domain filter and perform image processing:

[0086] The obtained one-dimensional spatial frequency transfer function is used as a radial filter function. It is then rotated and expanded around the origin of the image frequency domain in all azimuth directions to generate a two-dimensional rotationally symmetric spatial frequency transfer function (k). x / k yThe process involves reading the input image and performing a two-dimensional Fourier transform to obtain the image spectrum. The image spectrum is then multiplied by a two-dimensional spatial frequency filter to selectively transmit or suppress different spatial frequency components. Finally, an inverse Fourier transform is performed on the filtered spectrum to obtain the processed image. If the processed image does not achieve the expected results in denoising, edge enhancement, or scale selection, the process returns to step two or three to readjust the liquid crystal control state or structural parameters until the image processing result meets the expected requirements. Through these steps, the application effect of the described DBR-LC structure in long-wave infrared image denoising, edge extraction, and multi-scale feature selection can be verified.

[0087] Step 5: Prepare the long-wave infrared tunable optical analog computing device according to the parameters that have been verified above.

[0088] Example 3

[0089] This embodiment illustrates how the DBR-LC structure switches between low-pass noise reduction and high-pass edge extraction functions under different liquid crystal states when the SiO2 layer thickness is 1550nm.

[0090] like Figure 3 As shown, when the liquid crystal is not powered on, the DBR-LC structure mainly retains low spatial frequency components and suppresses some high spatial frequency components, exhibiting a low-pass filter response. Figure 4 The image processing results show that after low-pass filtering, the small noise in the image is significantly smoothed, while the overall outline of the main target is still maintained. The maximum suppressable noise size is about 0.23 mm, indicating that this state can be used for denoising long-wave infrared images.

[0091] like Figure 5 As shown, when the liquid crystal is energized, the change in the effective refractive index of the liquid crystal causes the resonant transmission region of the DBR-LC cavity to shift towards a larger k-value. x The / k0 region shifts, suppressing low spatial frequency components and enhancing higher spatial frequency components. Figure 6 It can be seen that this state corresponds to a high-pass or edge-enhancing filter response, which can extract edge information in the image, with an edge extraction size of approximately 0.085mm to 0.102mm.

[0092] Depend on Figure 3 and Figure 5 It can be seen that when the SiO2 layer thickness is 1550nm, by switching the liquid crystal between the unpowered and powered states, the same DBR-LC structure can switch between low-pass noise reduction and high-pass edge extraction functions.

[0093] Example 4

[0094] This embodiment illustrates how the DBR-LC structure switches between bandstop noise reduction and bandpass edge extraction functions under different liquid crystal states when the SiO2 layer thickness is 1580nm.

[0095] In this embodiment, except that the SiO2 layer thickness is set to 1580 nm, the other structural parameters are the same as in Embodiment 2. In the unpowered state, the effective refractive index of the liquid crystal cavity layer 2 is (n=1.499); in the powered state, the effective refractive index of the liquid crystal cavity layer 2 is (n=1.693).

[0096] like Figure 7 As shown, when the liquid crystal is not powered on, the DBR-LC structure forms a suppression region within a specific spatial frequency range, exhibiting a band-stop filter response. Figure 8 The image processing results show that this state can selectively weaken noise or local textures falling within the stopband range, rather than smoothing the image as a whole; the main suppressed noise size is about 0.22mm to 0.23mm, indicating that this mode is suitable for selective suppression of noise or texture interference in specific frequency bands.

[0097] like Figure 9 As shown, when the liquid crystal is in an energized state, the transmittance is in the middle k. x The / k0 region forms a distinct passband, corresponding to a ring passband response in two-dimensional k-space. Figure 10 The image processing results show that this state can selectively extract edge structures that match the passband, and the feature size corresponding to the strong response is about 101.6μm to 194.7μm, indicating that this mode has spatial scale selection capability.

[0098] Depend on Figure 7 and Figure 9 It can be seen that when the SiO2 layer thickness is 1580nm, the DBR-LC structure can switch between band-stop noise reduction and band-pass edge extraction functions through liquid crystal state switching. Compared with Example 3, this example shows that changing the SiO2 layer thickness in the DBR can change the basic spatial frequency response, thereby obtaining a more refined frequency band selection function.

[0099] Example 5

[0100] This embodiment illustrates how the DBR-LC structure achieves edge extraction within different spatial frequency ranges under different liquid crystal states when the SiO2 layer thickness is 1600nm.

[0101] In this embodiment, except that the SiO2 layer thickness is set to 1600 nm, the other structural parameters are the same as in Embodiment 2. In the unpowered state, the effective refractive index of the liquid crystal cavity layer 2 is (n=1.499); in the powered state, the effective refractive index of the liquid crystal cavity layer 2 is (n=1.693).

[0102] like Figure 11 As shown, when the liquid crystal is not powered on, the DBR-LC structure forms a bandpass spatial frequency transfer function, which can selectively transmit image components within a certain spatial frequency range, thus enabling the extraction of edge or structural information matching that passband. For example... Figure 12 As shown, when the liquid crystal is energized, the change in the effective refractive index of the liquid crystal causes the resonant transmission channel of the DBR-LC cavity to shift, and the passband position of the two-dimensional k-space filter changes accordingly. Therefore, Figure 11 The corresponding edge extraction results and Figure 12 Unlike other methods, it can select edge information within another spatial frequency range.

[0103] Figure 13 Comparison results of scale selection under two liquid crystal states are presented. Figure 13 It can be seen that when the passband is located at a higher k x In the / k0 region, the response is more pronounced for finer line groups and small-sized edges, with the optimal extraction size being approximately 84.67 μm to 135.5 μm; when the passband is located at a lower k... x In the / k0 region, coarser structures and larger edges are more easily preserved, and the feature size that can be extracted relatively stably is about 169.34 μm.

[0104] Depend on Figures 11 to 12 It is known that when the SiO2 layer thickness is 1600 nm, the DBR-LC structure can change the position of the bandpass region in k-space through liquid crystal state switching, thereby achieving edge extraction within different spatial frequency ranges. This embodiment illustrates that the present invention can achieve scale-selective image processing through DBR structure parameter adjustment and liquid crystal electro-controlled refractive index adjustment.

[0105] The preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the specific details of the above embodiments. Within the scope of the technical concept of the present invention, various simple modifications can be made to the technical solution of the present invention, and these simple modifications all fall within the protection scope of the present invention.

[0106] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present invention will not describe the various possible combinations separately.

[0107] Furthermore, various different embodiments of the present invention can be combined in any way, as long as they do not violate the spirit of the present invention, they should also be regarded as the content disclosed by the present invention.

Claims

1. A method for spatial frequency modulation of a long-wave infrared tunable optical analog computing device, characterized in that, include: Establish a long-wave infrared optical simulation computing device; The device comprises an upper reflective layer (1), a liquid crystal cavity layer (2), and a lower reflective layer (3) in sequence along the direction of incident light propagation. The upper reflective layer (1) and the lower reflective layer (3) are both distributed Bragg mirrors, which are formed by periodically alternating arrangement of a first dielectric layer (4) and a second dielectric layer (5). The liquid crystal cavity layer (2) is disposed between the upper reflective layer (1) and the lower reflective layer (3) to form a tunable Fabry-Perot type resonant cavity. By adjusting the effective refractive index of the liquid crystal cavity layer (2) by applying an external electric field, the optical thickness and phase matching conditions of the tunable Fabry-Perot type resonant cavity are changed, causing the position of the transmission peak or stopband in k space to move dynamically, thereby obtaining different spatial frequency transfer functions; using different spatial frequency transfer functions, different spatial frequency components in long-wave infrared images are selectively transmitted, suppressed or enhanced.

2. The spatial frequency modulation method for the long-wave infrared tunable optical analog computing device as described in claim 1, characterized in that, The liquid crystal cavity layer (2) is composed of a liquid crystal material with electrically controllable refractive index tuning capability. The liquid crystal material is E7 liquid crystal material. The effective refractive index of the E7 liquid crystal is n=1.499 when it is not energized. When it is energized, the orientation of the liquid crystal molecules changes and the effective refractive index becomes n=1.

693.

3. The spatial frequency modulation method for the long-wave infrared tunable optical analog computing device as described in claim 1, characterized in that, The first dielectric layer (4) is a low refractive index dielectric layer, including SiO2, MgF2 or Al2O3; the second dielectric layer (5) is a high refractive index dielectric layer, including TiO2, Si, Ge, ZnS or ZnSe; at the target working wavelength, the refractive index difference between the second dielectric layer (5) and the first dielectric layer (4) is not less than 0.

2.

4. The spatial frequency modulation method for the long-wave infrared tunable optical analog computing device as described in claim 1, characterized in that, The thickness of the first dielectric layer (4) and the second dielectric layer (5) is 0.05 μm to 10 μm; the number of periods of the upper reflective layer (1) and the lower reflective layer (3) is 1 to 15; and the thickness of the liquid crystal cavity layer (2) is 1 μm to 50 μm.

5. The spatial frequency modulation method for the long-wave infrared tunable optical analog computing device as described in claim 1, characterized in that, The spatial frequency transfer function includes a low-pass filter transfer function, a high-pass filter transfer function, a band-stop filter transfer function, or a band-pass filter transfer function, which are used to achieve denoising processing, edge enhancement processing, noise suppression processing of specific frequency bands, or target enhancement processing of specific scales in long-wave infrared images, respectively.

6. The spatial frequency modulation method for the long-wave infrared tunable optical analog computing device as described in claim 1, characterized in that, The long-wave infrared tunable optical analog computing device is obtained through a design method, including: Step 1: Establish a distributed Bragg reflector-liquid crystal cavity multilayer structure model, and set the material and thickness of the first dielectric layer (4) and the second dielectric layer (5), the number of periods of the upper reflective layer (1) and the lower reflective layer (3), the thickness of the liquid crystal cavity layer 2, and the normalized transverse wave vector k. x The calculation range of / k0; where k0 is the wavenumber of the incident light in vacuum, k x This represents the transverse component of the incident light wave vector along the direction of the device surface; Step 2: Calculate the effective refractive index of the liquid crystal cavity layer (2) under different electric field control states, substitute it into the distributed Bragg reflector-liquid crystal cavity multilayer structure model, and use the transfer matrix method to calculate the transmittance under different wavelengths, different incident angles and different polarization conditions, so as to obtain the angular transmission spectrum of the distributed Bragg reflector-liquid crystal cavity multilayer structure under different liquid crystal electric control states. Step 3: Based on the wave vector mapping relationship k between the normalized transverse wave vector and the incident angle x / k0=sinθ, the angle transmission spectrum is mapped to a spatial frequency transfer function to obtain the current position of the transmission peak or stopband in the spatial frequency domain; where θ is the incident angle of the incident light relative to the normal direction of the device. By adjusting the layer thickness of the first dielectric layer (4) and the second dielectric layer (5), the number of periods of the upper reflective layer (1) and the lower reflective layer (3) and the effective refractive index of the liquid crystal cavity layer (2), the position, bandwidth and shape of the transmission peak or stopband in the spatial frequency domain are adjusted until the spatial frequency response meets the design requirements, thereby forming a low-pass, high-pass, band-pass or band-stop spatial frequency transfer response; Step 4: Rotate and expand the obtained spatial frequency transfer function along the azimuth direction to generate a two-dimensional spatial frequency transfer function. Multiply it with the Fourier spectrum of the input image and then perform an inverse Fourier transform to obtain the processed image. If the image denoising, edge enhancement, or scale selection functions do not achieve the expected results, return to Step 2 or Step 3 to readjust the parameters until the verification is qualified. Step 5: Prepare a long-wave infrared tunable optical analog computing device according to the verified parameters.

7. A long-wave infrared tunable optical analog computing device, characterized in that, Along the direction of incident light propagation, the structure includes an upper reflective layer (1), a liquid crystal cavity layer (2), and a lower reflective layer (3); the upper reflective layer (1) is an upper distributed Bragg mirror, and the lower reflective layer (3) is a lower distributed Bragg mirror; the liquid crystal cavity layer (2) is disposed between the upper reflective layer (1) and the lower reflective layer (3) to form a tunable Fabry-Perot resonant cavity; both the upper reflective layer (1) and the lower reflective layer (3) are formed by periodically alternating arrangement of a first dielectric layer (4) and a second dielectric layer (5).

8. The long-wave infrared tunable optical analog computing device as described in claim 7, characterized in that, The lower reflective layer (3) is disposed on the fused silica substrate; the first medium layer (4) is a low refractive index medium layer, and the second medium layer (5) is a high refractive index medium layer.

9. The long-wave infrared tunable optical analog computing device as described in claim 7, characterized in that, The target operating wavelength of the device is 10.6 μm. The first dielectric layer (4) is a SiO2 layer, the second dielectric layer (5) is a TiO2 layer, the upper reflective layer (1) and the lower reflective layer (3) each have 8 pairs of periods, the TiO2 layer has a thickness of 1100 nm, the liquid crystal cavity layer (2) has a thickness of 10 μm, and the SiO2 layer has a thickness of 1550 nm, 1580 nm or 1600 nm. By electronically adjusting the effective refractive index of the liquid crystal cavity layer (2), the device can switch between low-pass noise reduction mode, high-pass edge extraction mode, band-stop suppression mode or band-pass scale selection mode.

Images