Minimum calculation imaging spectrometer based on cephalopod bionics

Abstract

The invention discloses a minimum calculation imaging spectrometer based on cephalopod bionics, which belongs to the technical field of optical imaging and comprises a mask, a dispersion lens and a photoelectric detector which are sequentially and coaxially arranged along an optical axis of a system. A through hole is formed in the mask, and the through hole is in a non-centrosymmetric shape by taking a system optical axis as a reference; a displacement table is arranged on the photoelectric detector and is used for driving the photoelectric detector to move along the optical axis of the system; the photoelectric detector is connected with a controller, and the controller is used for acquiring an out-of-focus image from the photoelectric detector and reconstructing spectral data from the out-of-focus image through a compressed sensing algorithm; the out-of-focus image is obtained by refracting the spectral signal to the photoelectric detector through the dispersion lens after passing through the through hole; the position of the photoelectric detector is adjusted through the displacement table, and the defocusing amount is changed, so that defocusing images with different defocusing amounts are obtained. The method has the characteristics of extremely simple structure and high coding efficiency, and the spectrum reconstruction precision can be remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of optical imaging technology, and in particular to a minimalist computational imaging spectrometer based on cephalopod bionics. Background Art

[0002] Computational spectral imaging provides a revolutionary tool for material identification and composition analysis by acquiring a spatial-spectral three-dimensional data cube of the target scene. It is widely used in remote sensing mapping, biomedicine, food safety and other fields. Traditional spectral imaging relies on independent spectroscopic elements such as prisms and gratings to collect data wavelength by wavelength or pixel by pixel, resulting in a bulky and expensive system and difficulty in balancing spatial resolution, spectral resolution and temporal resolution. In recent years, computational spectral imaging technology based on compressed sensing theory has emerged. It uses an optical system to encode and compress the light field of the three-dimensional data cube, and then reconstructs the spectral information through an algorithm, achieving a lightweight and efficient system. However, existing methods still face key challenges in the coding mechanism - how to achieve high-resolution spectral information modulation in a compact architecture has become a core issue restricting its practical application. Summary of the Invention

[0003] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide a minimalist computational imaging spectrometer based on cephalopod bionics. By imitating the asymmetric pupil structure and utilizing the chromatic aberration of the optical system, high-resolution PSF encoding is achieved, the effectiveness of the spectral response matrix is ​​enhanced, and the spectral reconstruction accuracy is improved while simplifying the system structure.

[0004] To achieve the above object, the present invention is implemented by adopting the following technical solutions:

[0005] The present invention provides a minimalist computational imaging spectrometer based on cephalopod bionics, comprising a mask, a dispersion lens, and a photodetector coaxially arranged along the optical axis of the system;

[0006] The mask is provided with a through hole, which is non-centrosymmetrical with respect to the optical axis of the system, and preferentially collects off-axis light in a specific direction to achieve wavelength-sensitive PSF modulation;

[0007] The photoelectric detector is provided with a displacement stage, and the displacement stage is used to drive the photoelectric detector to move along the direction of the system optical axis;

[0008] The photodetector is connected to a controller, and the controller is used to obtain a defocused image from the photodetector and reconstruct spectral data from the defocused image using a compressed sensing algorithm;

[0009] The spectral signal of the object passes through the through hole and is refracted by the dispersive lens to the photodetector to generate the defocused image; the position of the photodetector is adjusted by the translation stage to change the defocus amount, thereby obtaining defocused images with different defocus amounts.

[0010] Optionally, the mask is integrated with the dispersion lens.

[0011] Optionally, the through hole is semi-circular, strip-shaped or W-shaped.

[0012] Optionally, the dispersion lens is a lens made of a high-refractive-index, low-Abbe-number material, and the high-refractive-index, low-Abbe-number material is lanthanide optical glass, heavy flint glass, or titanium oxide.

[0013] Optionally, the dispersive lens is a diffractive optical element.

[0014] Optionally, the photoelectric detector is a planar array detector.

[0015] Optionally, the translation stage is driven by electricity, and its driving signal is provided by the controller.

[0016] Optionally, reconstructing spectral data from the defocused image using a compressed sensing algorithm includes:

[0017] Construct a multi-focal plane light intensity matrix based on defocused images with different defocus amounts and the measurement matrix :

[0018]

[0019]

[0020] Where, , is the number of groups of defocus amount; For the The original image under wavelengths of light, For the The defocus amount of the group Defocused image under wavelengths of light, is the pixel coordinate, is the noise parameter, is the number of wavelengths;

[0021] Based on multi-focal plane intensity matrix , converting the reconstruction problem into a regularized optimization problem:

[0022]

[0023] Where, The reconstructed original image , is the strength coefficient, is a sparse or smooth regularization term;

[0024] An iterative shrinkage threshold algorithm and a deep unfolding network are used to solve the regularized optimization problem.

[0025] Optionally, before reconstructing spectral data from the defocused image using a compressed sensing algorithm, the defocused image is also preprocessed, and the preprocessing includes dark current correction and flat field correction.

[0026] Optionally, the defocused image is subjected to field correction according to the defocus amount, and the expression is:

[0027]

[0028] Where, is the field of view size before and after field correction, is the distance between the object and the image plane on the photodetector, is the focal length of the dispersive lens, is the defocus amount.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] The present invention provides a minimalist computational imaging spectrometer based on cephalopod bionics. It uses an asymmetric through-hole to imitate the pupil of a cephalopod, significantly enhancing the PSF difference of light of different wavelengths and improving the effectiveness of the spectral response matrix. It does not require independent splitting elements and only consists of an asymmetric aperture, a lens and a detector. It is small in size and simple in structure. Multi-focal plane image acquisition is combined with a compressed sensing algorithm to effectively improve spectral resolution and noise resistance, making it suitable for high-precision spectral detection in complex scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 Schematic diagram of the structure of a minimalist computational imaging spectrometer based on cephalopod bionics provided by an embodiment of the present invention;

[0032] Figure 2 1 is a schematic structural diagram of a mask and a through hole thereon provided by an embodiment of the present invention;

[0033] Figure 3 This is a schematic diagram of the defocus measurement principle of the minimalist computational imaging spectrometer provided by an embodiment of the present invention;

[0034] Figure 4 is a schematic diagram of defocused images of different wavelengths provided by an embodiment of the present invention;

[0035] The following are marked in the figure:

[0036] 1-Mask; 2-Dispersion lens; 3-Photodetector; 4-System optical axis. DETAILED DESCRIPTION

[0037] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention.

[0038] Example 1:

[0039] like Figure 1 As shown, an embodiment of the present invention provides a minimalist computational imaging spectrometer based on cephalopod bionics, comprising a mask 1, a dispersion lens 2, and a photodetector 3 coaxially arranged along a system optical axis 4.

[0040] The mask 1 is provided with a through hole, which is non-centrosymmetrical with respect to the optical axis 4 of the system. Figure 2 As shown, the through hole can be set to a semi-circular shape, a long strip shape, a W shape or any other shape that meets the requirements.

[0041] Semi-circular: inner diameter , outer diameter , light-transmitting in the horizontal direction and blocking in the vertical direction; long strip: width ,length , , giving priority to collecting horizontal off-axis light.

[0042] The transmittance function of the through hole satisfies the asymmetric distribution. For example, the transmittance function of the semi-annular through hole is:

[0043]

[0044] Where, is the radius angle The pass rate of the department.

[0045] Inspired by the asymmetric pupil structure of cephalopods (e.g., a semi-annular pupil preferentially captures horizontal off-axis light), this approach selectively enhances off-axis light transmission through a semi-annular or slit-like aperture design that mimics the cephalopod pupil. Combined with the lens's inherent chromatic aberration (different focal lengths at different wavelengths), this approach allows light of different wavelengths to form significantly different PSFs at the image plane. For example, due to their different focal lengths, the difference in horizontal PSF spread between blue and red light is amplified under semi-annular aperture modulation, resulting in a unique "wavelength-to-blur pattern" mapping. This biomimetic encoding mechanism eliminates the need for independent dispersive elements and, through the synergistic effect of aperture shape and chromatic aberration, constructs a high-rank spectral response matrix, significantly reducing the underdetermination of spectral reconstruction.

[0046] The dispersion lens 2 is a lens or diffractive optical element made of a material with a high refractive index and a low Abbe number (ie, high dispersion), such as lanthanide optical glass, heavy flint glass, or titanium oxide.

[0047] If the dispersion lens 2 is a lens, it can be integrated with the mask 1, that is, the mask 1 is attached to one side of the dispersion lens 2. The lens material has a wavelength-dependent refractive index, and its focal length satisfies the thin lens formula:

[0048]

[0049] Where, wavelength The refractive index and focal length under is the radius of curvature of the lens surface.

[0050] The photodetector 3 uses a high-sensitivity array detector to collect defocused images. In order to change the defocus amount, a displacement stage is usually set on the photodetector 3, such as Figure 3 As shown, the translation stage is used to drive the photodetector 3 to move along the direction of the system optical axis 4, thereby changing the defocus amount.

[0051] Defocus , , is the image distance and object distance.

[0052] The photodetector 3 is connected to a controller, which is used to obtain a defocused image from the photodetector 3 and reconstruct spectral data from the defocused image through a compressed sensing algorithm; the defocused image is obtained by the spectral signal passing through the through hole and refracting onto the photodetector through a dispersive lens; the position of the photodetector 3 is adjusted by a displacement stage to change the defocus amount, thereby obtaining defocused images with different defocus amounts.

[0053] To facilitate automated operation, the translation stage is electrically driven, and its driving signal is provided by a controller. It is controlled and adjusted according to the step length within a certain range, such as collecting a set of data with a step length of 100um.

[0054] Specifically in this embodiment, reconstructing spectral data from a defocused image using a compressed sensing algorithm includes the following steps:

[0055] Step S1: preprocessing the defocused image, including dark current correction and flat field correction, to eliminate noise and non-uniform response.

[0056] Step S2: Construct a multi-focal plane light intensity matrix based on defocused images with different defocus amounts and the measurement matrix :

[0057]

[0058]

[0059] Where, , is the number of groups of defocus amount; For the The original image under wavelengths of light, For the The defocus amount of the group Defocused image under wavelengths of light, is the pixel coordinate, is the noise parameter, is the number of wavelengths;

[0060] The asymmetric through-hole amplifies the difference in off-axis light transmission paths for light of different wavelengths, and the defocused image satisfies:

[0061]

[0062] The asymmetric structure causes the diffusion degree of the PSF in the opening direction (such as the horizontal direction) to vary significantly with the wavelength, forming a unique wavelength coding feature. Figure 4 As shown in Figure 3, the wavelength sensitivity of PSF at the same defocus position is different. The PSFs at different wavelengths have significant differences, which form the basis of spectral recovery.

[0063] Step S3: Based on the multi-focal plane light intensity matrix , converting the reconstruction problem into a regularized optimization problem:

[0064]

[0065] Where, The reconstructed original image , is the strength coefficient, is a sparse or smooth regularization term;

[0066] Step S4: using an iterative shrinkage threshold algorithm and a deep expansion network to solve the regularization optimization problem.

[0067] Because the focal plane displacement will change the field of view, it needs to be corrected in advance. The geometric optical model is used to calculate the change in depth of focus when the object position and the image plane position are fixed. The field of view size closest to the image plane is used as the standard field of view size. The calculated or collected PSF library, that is, the measurement matrix, is cropped to the same field of view range and the pixel number interpolation correction is performed. The same operation is performed on the subsequent collected images. The field of view of the defocused image is corrected according to the defocus amount, and its expression is:

[0068]

[0069] Where, is the field of view size before and after field correction, is the distance between the object and the image plane on the photodetector, is the focal length of the dispersive lens, is the defocus amount.

[0070] The minimalist computational imaging spectrometer based on cephalopod bionics provided by the present invention has the following advantages:

[0071] 1. Minimalist architecture: Eliminates complex spectroscopic components such as prisms and gratings, and consists solely of a bionic asymmetric aperture lens and detector, resulting in a smaller size and lower cost.

[0072] 2. Efficient Encoding: Utilizing the pupil shape optimized by biological evolution and naturally adapting to chromatic aberration characteristics, the wavelength sensitivity of the PSF is significantly improved compared to traditional symmetrical apertures, thereby improving spectral resolution.

[0073] 3. Robustness: By simulating the dynamic focusing mechanism of cephalopods (adjusting the image distance and scanning focus position) and combining it with a spectral reconstruction algorithm, the reconstruction error in low signal-to-noise ratio scenarios is improved.

[0074] This design provides a new path for miniaturized spectral imaging equipment and has significant application potential in environmental emergency monitoring, portable medical testing and other fields.

[0075] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, features defined as "first", "second", etc. may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0076] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0077] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A minimalist computational imaging spectrometer based on cephalopod biomimetic, characterized by: It includes a mask, a dispersion lens and a photodetector which are coaxially arranged along the optical axis of the system; The mask is provided with a through hole, which is non-centrosymmetrical with respect to the optical axis of the system, and preferentially collects off-axis light in a specific direction to achieve wavelength-sensitive PSF modulation; The photoelectric detector is provided with a displacement stage, and the displacement stage is used to drive the photoelectric detector to move along the direction of the system optical axis; The photodetector is connected to a controller, and the controller is used to obtain a defocused image from the photodetector and reconstruct spectral data from the defocused image using a compressed sensing algorithm; The spectral signal of the object passes through the through hole and is refracted by the dispersive lens to the photodetector to generate the defocused image; the position of the photodetector is adjusted by the translation stage to change the defocus amount, thereby obtaining defocused images with different defocus amounts.

2. The minimalist computational imaging spectrometer based on cephalopod bionics according to claim 1, characterized in that: The mask is integrated with the dispersion lens.

3. The minimalist computational imaging spectrometer based on cephalopod bionics according to claim 1, characterized in that: The through hole is in a semi-circular shape, a long strip shape or a W shape.

4. The minimalist computational imaging spectrometer based on cephalopod bionics according to claim 1, characterized in that: The dispersion lens is a lens made of a material with a high refractive index and a low Abbe number. The material with a high refractive index and a low Abbe number is lanthanide optical glass, heavy flint glass or titanium oxide.

5. The minimalist computational imaging spectrometer based on cephalopod biomimetics according to claim 1, characterized in that: The dispersion lens is a diffractive optical element.

6. The minimalist computational imaging spectrometer based on cephalopod bionics according to claim 1, characterized in that: The photoelectric detector is a planar array detector.

7. The minimalist computational imaging spectrometer based on cephalopod biomimetics according to claim 1, characterized in that: The translation stage is driven by electricity, and its driving signal is provided by the controller.

8. The minimalist computational imaging spectrometer based on cephalopod bionics according to claim 1, characterized in that: The reconstructing spectral data from the defocused image by using a compressed sensing algorithm comprises: Construct a multi-focal plane light intensity matrix based on defocused images with different defocus amounts and the measurement matrix : Where, , is the number of groups of defocus amount; For the The original image under wavelength light, For the The defocus amount of the group Defocused image under wavelengths of light, is the pixel coordinate, is the noise parameter, is the number of wavelengths; Based on multi-focal plane intensity matrix , converting the reconstruction problem into a regularized optimization problem: Where, The reconstructed original image , is the strength coefficient, is a sparse or smooth regularization term; An iterative shrinkage threshold algorithm and a deep unfolding network are used to solve the regularized optimization problem.

9. The minimalist computational imaging spectrometer based on cephalopod biomimetics according to claim 1, characterized in that: Before reconstructing spectral data from the defocused image using a compressed sensing algorithm, the defocused image is preprocessed, and the preprocessing includes dark current correction and flat field correction.

10. The minimalist computational imaging spectrometer based on cephalopod bionics according to claim 1, characterized in that: The defocused image is corrected for field of view according to the defocus amount, and the expression is: Where, is the field of view size before and after field correction, is the distance between the object and the image plane on the photodetector, is the focal length of the dispersive lens, is the defocus amount.

Images