Lensless hyperspectral imaging system and method based on wavelength phase dual coding

Abstract

The invention provides a lensless hyperspectral imaging system and method based on wavelength phase dual coding, relates to the technical field of hyperspectral imaging, and aims to solve the problems that an existing lens spectral imaging system is large in lens system size and complex in structure, compromises between spatial resolution and spectral resolution, and is low in incident light utilization rate. The system comprises an aluminum alloy shell, an optical diffraction element and a quantum dot spectrum chip, wherein a light limiting hole is formed in the aluminum alloy shell; the central positions of the light limiting hole, the optical diffraction element and the quantum dot spectrum chip are located on the same plane; the light limiting hole is used for limiting the diffraction angle of incident light after passing through the optical diffraction element; the optical diffraction element is used for encoding incident light; and the quantum dot spectrum chip is used for carrying out wavelength coding on the incident light coded by the optical diffraction element, converting the incident light into an electric signal and carrying out spectrum reconstruction on the electric signal. The problems of low utilization rate of incident light and the like are avoided, and improvement of spectral resolution and spatial resolution is ensured while the system structure is simplified.

Description

Technical Field

[0001] The present invention belongs to the technical field of hyperspectral imaging, and in particular relates to a lensless hyperspectral imaging system and method based on wavelength-phase dual encoding. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] Spectral imaging technology is a comprehensive, interdisciplinary cutting-edge technology that integrates optical imaging, spectral analysis, signal detection, computers, information processing, and digital image processing. The spectral imaging system can not only obtain spatial information of the target under test, but also reveal the material composition of the target under test. This spectral imaging technology that organically combines spectral measurement technology and imaging technology has extremely important application value in remote sensing, medical diagnosis, autonomous driving, face recognition and other fields.

[0004] Hyperspectral imaging technology combines imaging and spectral analysis technology. It can perform continuous spectral measurements on target objects within a specific wavelength range, obtain spatial and spectral information of the target at the same time, and obtain three-dimensional cube data of space and spectrum through two-dimensional detection. Hyperspectral imaging systems can usually obtain hundreds of spectral data, distinguish subtle spectral differences of objects, and realize material identification. In addition, hyperspectral imaging systems can not only identify the morphology and structure of objects, but also accurately analyze the spectral characteristics of each pixel point to achieve accurate positioning and analysis of the target to be measured. At the same time, it can realize non-destructive detection of the object to be measured.

[0005] Existing spectral imaging technologies are complex in structure, bulky in size, and cannot achieve in-situ real-time monitoring. However, with the rapid growth in demand for spectrum analysis, the need to reduce the physical size, cost, or power consumption of spectrometers takes precedence over the need for high performance. At the same time, the demand for portable or handheld spectral analysis equipment in consumer electronics also requires further simplification of the system structure. In order to further compress the system structure, the method of simply shrinking the desktop system is facing serious challenges, and the emergence of lensless spectral imaging systems based on computational spectral imaging technology provides new ideas for the development of the next generation of miniature spectral imaging technology.

[0006] Lensless imaging abandons the traditional lens point-to-point mapping mode and projects the points in the physical space into specific patterns in the image space. Different object points are superimposed and encoded on the image plane to form image information that cannot be recognized by the human eye but can be decoded and restored through algorithms. It has obvious advantages in compact structure. In recent years, the feasibility of achieving spatial and spectral resolution decoupling through lensless system design has also been verified. At the same time, the development of deep learning technology has further improved the imaging quality of lensless imaging technology.

[0007] Amplitude-type lensless imaging usually uses binary black and white patterns to modulate the incident light. The research focus is mainly on the design of the mask structure. The purpose of improving numerical stability is achieved through the optimization of the mask structure, and then the optimization of the point spread function is achieved to improve the quality of the spectrum reconstruction. However, the amplitude-type lensless imaging structure only allows up to 50% of the light to pass through, which will cause a part of the light intensity to be lost, resulting in low energy utilization.

[0008] Hyperspectral imaging technology combines imaging and spectral analysis technology. It can perform continuous spectral measurements on target objects within a specific wavelength range, obtain spatial and spectral information of the target at the same time, and obtain three-dimensional cube data of space and spectrum through two-dimensional detection. Hyperspectral imaging systems can usually obtain hundreds of spectral data, distinguish subtle spectral differences of objects, and realize material identification. In addition, hyperspectral imaging systems can not only identify the morphology and structure of objects, but also accurately analyze the spectral characteristics of each pixel point to achieve accurate positioning and analysis of the target to be measured. At the same time, it can realize non-destructive detection of the object to be measured.

[0009] Lensless imaging abandons the traditional lens point-to-point mapping mode, and projects the points in the physical space into specific patterns in the image space. Different object points are superimposed and encoded on the image plane, forming an image information that cannot be recognized by the human eye but can be decoded and restored by algorithms. The compact structure has obvious advantages. However, due to the limitations of existing technologies, the current lensless spectral imaging system is still in the laboratory research stage and cannot meet the needs of actual application scenarios. Among them, the spectral detection range, incident light utilization, and the number of spectral channels are all limiting factors.

[0010] At present, lensless spectral imaging technology, especially phase-type lensless spectral imaging technology, generally adopts the design of binary or ternary optical diffraction elements to encode the incident light, or adopts a lensless spectral imaging system that combines a unidimensional optical diffraction element with a narrow-band filter. The lensless spectral imaging system designed with a ternary optical diffraction element currently used in the laboratory consists only of an optical diffraction element and an image sensor. This technical method can achieve spectral imaging within the wavelength range of 400-600nm. At the same time, it has a high spatial resolution based on a high-performance image sensor, and does not require a lens structure. It is simple and can greatly compress the system structure, but the spectral detection range is narrow, and the number of spectral channels is limited, so wide spectrum detection cannot be achieved, and the usage scenarios are limited.

[0011] The spectral imaging system based on the binary optical diffraction element design consists of a broadband filter, a binary optical diffraction element, a Fresnel lens and an image sensor. This binary optical diffraction element design can replace the original dual-camera system of the single optical diffraction element, significantly simplifying the system structure and reducing system complexity. However, this specially designed optical diffraction element also faces the problem of a small number of spectral channels and a narrow spectral detection range.

[0012] A composite lensless spectral imaging system, this method combines a one-dimensional optical diffraction element with phase encoding characteristics with a commercial narrow-band filter array. The optical diffraction element maps point light sources into surface information, realizes multiplexing of incident light, and improves the utilization rate of incident light. The incident light encoded by the optical diffraction element is then wavelength-encoded by a narrow-band filter. The encoded incident light is collected by the image sensor and converted into an electrical signal for subsequent spectrum reconstruction. The system can achieve spectral detection within the wavelength range of 400-700nm, the number of spectral channels reaches 64, and the spatial resolution can reach 0.19 superpixels. However, the phase-encoded optical diffraction element used is a pseudo-random design and does not have clear optimization indicators. The commercial narrow-band filter used is first limited in the number of spectral channels, and the increase in the number of spectral channels will significantly increase the cost. In addition, the narrow-band filter will reduce the utilization rate of incident light. Summary of the invention

[0013] In order to overcome the shortcomings of the above-mentioned prior art, the present invention provides a lensless hyperspectral imaging system and method based on wavelength-phase dual encoding. By using a phase-type optical diffraction element, a point light source is mapped into a surface light source with a specific image, and then wavelength encoding is performed through a broadband quantum dot filter to improve the utilization rate of the incident light. At the same time, through the dual encoding of wavelength and phase and the decoding algorithm, the coupling problem of spectral resolution and spatial resolution is solved, thereby realizing the design of a hyperspectral imaging system with compact structure and excellent performance, and providing technical support for the research and development and engineering application of micro hyperspectral imaging systems.

[0014] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0015] The first aspect of the present invention provides a lensless hyperspectral imaging system based on wavelength-phase dual encoding, comprising:

[0016] Aluminum alloy housing, optical diffraction element and quantum dot spectrum chip;

[0017] The aluminum alloy housing is provided with a limited light hole;

[0018] The center positions of the light-limiting hole, the optical diffraction element and the quantum dot spectrum chip are on the same plane;

[0019] The light-limiting hole is used to limit the diffraction angle of the incident light after passing through the optical diffraction element, and the diffraction angle of the incident light is related to the size of the light-limiting hole;

[0020] The optical diffraction element is used to encode the incident light passing through to obtain the incident light encoded by the optical diffraction element;

[0021] The quantum dot spectrum chip is used to encode the wavelength of the incident light after the optical diffraction element is encoded, and at the same time convert the incident light into an electrical signal and reconstruct the signal spectrum.

[0022] As an embodiment, the optical diffraction element and the quantum dot spectrum chip are packaged inside an aluminum alloy housing.

[0023] As an embodiment, the quantum dot spectroscopy chip is a broadband quantum dot spectroscopy chip, including a broadband quantum dot filter array and an image sensor.

[0024] As an implementation mode, the broadband quantum dot filter array uses quantum dot materials with broad absorption peaks.

[0025] As an implementation mode, the broadband quantum dot filter array adopts a periodic structure, and the periodic structure is: the broadband quantum dot wave array is divided into 183×230 periodic units, and each periodic unit contains 3×3 quantum dot arrays.

[0026] As an implementation mode, the size of the broadband quantum dot filter array is 5.5 mm×6.9 mm; the number of broadband quantum dot filter arrays is 550×690; and the size of each quantum dot is 10 μm.

[0027] As an embodiment, the size of the image sensor is consistent with the size of the broadband quantum dot filter array.

[0028] The second aspect of the present invention provides a lensless hyperspectral imaging method based on wavelength-phase dual encoding, comprising:

[0029] Acquiring incident light to be imaged;

[0030] The incident light to be imaged is encoded by the optical diffraction element to obtain the encoded incident light;

[0031] The encoded incident light is wavelength-encoded through a broadband quantum dot spectral chip, and the incident light is converted into an electrical signal, and the electrical signal is reconstructed to obtain spectral imaging.

[0032] As an implementation method, the incident light to be imaged is encoded by an optical diffraction element to obtain the encoded incident light, and the formula is:

[0033]

[0034] Among them, v[x, y, λ] is the incident light, h[x, y] is the point spread function, crop represents the discrete two-dimensional linear convolution in the spatial dimension, and w[x, y, λ] is the encoded incident light.

[0035] As an implementation method, the encoded incident light is wavelength encoded through a broadband quantum dot spectrum chip, and the formula is:

[0036] L[x,y]=F λ [x,y]·w[x,y,λ];

[0037] Where L[x,y] is the incident light entering the image sensor after wavelength encoding, · represents the dot product,

[0038] F λ [x, y] is the filter function of the quantum dot array, and w[x, y, λ] is the incident light that enters the quantum dot filter array after being encoded by the optical diffraction element.

[0039] One or more of the above technical solutions have the following beneficial effects:

[0040] In this embodiment, the lensless spectral imaging system structure based on wavelength-phase dual encoding avoids the problems of bulky and complex structure of the lens system, compromise between spatial resolution and spectral resolution, and low utilization rate of incident light. The lensless system structure design avoids the problems of bulky and complex structure of the lens spectral imaging system; the incident light wavelength division multiplexing technology and broadband filtering technology avoid the problems of low utilization rate of incident light, and while simplifying the system structure, it ensures the improvement of spectral resolution and spatial resolution of the lensless spectral imaging system, making it more suitable for future intelligent and miniaturized spectrometer research and development and engineering applications.

[0041] In this embodiment, the broadband quantum dot spectral chip structure is designed, and a low-cost, mass-producible broadband quantum dot filter array is used to encode the incident light. Based on the compressed sensing theory, a small number of broadband filter arrays can be used to complete the design of a spectral chip with a large number of channels.

[0042] In this embodiment, the system structure combines a phase-coded optical diffraction element with a quantum dot hyperspectral chip, and utilizes the multiplexing characteristics of phase coding and the broadband coding characteristics of the quantum dot filter to improve light utilization and spectral coding efficiency, achieve decoupling of spectral resolution and spatial resolution, and compress the system size to complete the design of a high-performance, low-cost, compact lens-free hyperspectral imaging system.

[0043] In this embodiment, a broadband quantum dot hyperspectral chip design is adopted to broaden the spectral detection range and increase the number of spectral channels, thereby avoiding the low sensitivity problem caused by low incident light utilization rate often faced by narrow-band filters, and at the same time reducing costs to achieve large-scale mass production.

[0044] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0046] Figure 1 A schematic diagram of a lens-free spectral imaging system based on a ternary optical diffraction element in the prior art;

[0047] Figure 2 A schematic diagram of a spectral imaging system designed based on a binary optical diffraction element in the prior art;

[0048] Figure 3 A schematic diagram of a composite lensless spectral imaging system combining an optical diffraction element and a narrow-band filter in the prior art;

[0049] Figure 4 This is a schematic diagram of a lensless spectral imaging system according to Embodiment 1 of the present invention;

[0050] Figure 5 This is a schematic diagram of the structure of a broadband quantum dot spectrum chip according to Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0051] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0052] It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.

[0053] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0054] Embodiment 1

[0055] The first aspect of the present invention discloses a lensless hyperspectral imaging system based on wavelength-phase dual encoding, comprising:

[0056] Aluminum alloy housing, optical diffraction element and quantum dot spectrum chip;

[0057] The aluminum alloy housing is provided with a limited light hole;

[0058] The center positions of the light-limiting hole, the optical diffraction element and the quantum dot spectrum chip are on the same plane;

[0059] The light-limiting hole is used to limit the diffraction angle of the incident light after passing through the optical diffraction element, and the diffraction angle of the incident light is related to the size of the light-limiting hole;

[0060] The optical diffraction element is used to encode the incident light passing through to obtain the incident light encoded by the optical diffraction element;

[0061] The quantum dot spectrum chip is used to encode the wavelength of the incident light after the optical diffraction element is encoded, and at the same time convert the incident light into an electrical signal and reconstruct the signal spectrum.

[0062] The existing spectral imaging technology has certain shortcomings, such as Figure 1 As shown in the figure, the lensless spectral imaging system designed with a ternary optical diffraction element currently used in the laboratory is composed only of an optical diffraction element and an image sensor. The spectral detection range is narrow and the number of spectral channels is limited. It is unable to achieve wide spectrum detection and its usage scenarios are limited.

[0063] Another example Figure 2 As shown in the figure, a spectral imaging system designed based on a binary optical diffraction element is composed of a broadband filter, a binary optical diffraction element, a Fresnel lens and an image sensor. This specially designed optical diffraction element also faces the problem of a small number of spectral channels and a narrow spectral detection range.

[0064] For example Figure 3 As shown, a composite lensless spectral imaging system is shown. This method combines a unitary optical diffraction element with phase encoding characteristics with a commercial narrow-band filter array. However, the phase-encoded optical diffraction element used is a pseudo-random design and does not have clear optimization indicators. The commercial narrow-band filters used are firstly limited in the number of spectral channels, and the increase in the number of spectral channels will significantly increase the cost. In addition, the narrow-band filters will reduce the utilization rate of the incident light.

[0065] Based on the above problems, this embodiment provides a lensless hyperspectral imaging system based on wavelength-phase dual encoding, which improves the utilization rate of incident light, has a compact structure and excellent performance, and provides technical support for the research and development and engineering application of micro hyperspectral imaging systems.

[0066] like Figure 4As shown, in this embodiment, a lensless spectral imaging system is composed of a light-limiting hole, a spectral diffraction element and a quantum dot spectral chip. The centers of the three are on the same plane. The entire system is packaged in an aluminum alloy shell to reduce the influence of spatial stray light.

[0067] In this embodiment, a light-limiting hole is arranged at the opening of the aluminum alloy shell. The incident light passes through the light-limiting hole into the aluminum alloy shell. The diffraction angle of the incident light passing through the optical diffraction element is related to the size of the light-limiting hole. The light-limiting hole of the corresponding size can be selected as needed to ensure precise control of the diffraction angle of the incident light.

[0068] In this embodiment, the spectrum diffraction element adopts a phase-type optical diffraction element to improve the transmittance of incident light.

[0069] First, the high transmittance of the phase-type diffraction element can improve the utilization rate of the incident light and enhance the sensitivity of the system.

[0070] Secondly, the component can map a point light source into a specific pattern and use the principle of optical division multiplexing to reduce the loss of incident light information.

[0071] Finally, in order to objectively measure the amount of information collected in the lensless imaging system, a Fourier domain indicator based on the modulation transfer function is introduced: the modulation transfer function volume. The larger the value of this indicator, the more information the system encodes.

[0072] Based on this index, the Voronoi-Fresnel phase function is used. This lensless phase function similar to the compound eye of insects is composed of the Fresnel function of an ideal lens as the basic unit, which is closely arranged in the form of a Voronoi diagram in two-dimensional space. The generated point spread function has the characteristics of both sparseness and high contrast, and the polygonal aperture of each Voronoi unit represents the Fourier domain directional filtering, and has a more compact spatial distribution.

[0073] In addition, this method can optimize the distribution of these basic units to obtain a phase function with better frequency domain characteristics and optical performance, thereby improving system performance.

[0074] In this embodiment, the quantum dot spectrum chip uses a broadband quantum dot spectrum chip. The use of a broadband quantum dot spectrum chip can improve the utilization rate of incident light while increasing the number of spectrum channels and preventing the loss of effective information. At the same time, the spectrum chip converts the incident light into an electrical signal, and the detected signal is then uploaded to the host computer for spectrum reconstruction.

[0075] like Figure 5 As shown, in this embodiment, the broadband quantum dot spectrum chip includes a broadband quantum dot filter array and an image sensor.

[0076] In this embodiment, the spectral response range of the image sensor is one of the determining factors of the spectral detection range of the spectral imaging system. An image sensor with a wide spectral response of visible-near infrared (400nm-1000nm) is used to improve the spectral detection range of the spectral imaging system.

[0077] In this embodiment, the quantum dot filter array uses quantum dot materials with wide absorption peaks. Compared with the currently commercially available narrow-band filters, the broadband quantum dot filter can allow more incident light to pass through, thereby improving the utilization rate of incident light while reducing the probability of information loss.

[0078] The quantum dot filter array adopts a periodic structure. The size of the entire quantum array is consistent with the image sensor, 5.5mm×6.9mm. The size of each quantum dot is 10μm, and the number of quantum dot arrays is 550×690. The entire array is based on 3×3 quantum dot arrays as a periodic unit, and the entire array is divided into 183×230 periodic units. Each periodic unit can be used as an independent spectrometer for spectral analysis. At the same time, based on the reuse of quantum dot spectral chips, the entire quantum dot spectral chip can achieve 548×682 spatial spectral pixels.

[0079] At the same time, by combining phase-type optical diffraction elements with broadband quantum dot filters, it is possible to decouple spatial resolution from spectral resolution, further improve the spatial resolution of the lensless spectral imaging system, achieve sub-superpixel spatial resolution, and achieve high spectral resolution (spectral resolution <10nm).

[0080] In this embodiment, the incident light passes through the light-limiting hole and enters the aluminum alloy housing, reducing the influence of spatial stray light; then the incident light is encoded by the phase-type optical diffraction element to obtain the encoded incident light, thereby improving the transmittance of the incident light; and then the wavelength is encoded by the broadband quantum dot spectrum chip, which can improve the utilization rate of the incident light while increasing the number of spectral channels and preventing the loss of effective information. At the same time, the broadband quantum dot spectrum chip converts the incident light into an electrical signal, and the detected signal is then uploaded to the host computer for image reconstruction, thereby completing spectral imaging.

[0081] In summary, the lensless spectral imaging system that combines a phase-type lensless imaging system with a broadband quantum dot spectral chip can significantly compress the system structure, improve the utilization rate of incident light, enhance the system sensitivity, and at the same time achieve the decoupling of spectral resolution and spatial resolution, thereby achieving an overall improvement in system performance.

[0082] Embodiment 2

[0083] This embodiment provides a lensless hyperspectral imaging method based on wavelength-phase dual encoding, including:

[0084] S1, obtaining incident light to be imaged;

[0085] S2, encoding the incident light to be imaged through the optical diffraction element to obtain the encoded incident light;

[0086] S3. The encoded incident light is wavelength-encoded through a broadband quantum dot spectral chip, and the incident light is converted into an electrical signal, and the electrical signal is reconstructed to obtain spectral imaging.

[0087] In this embodiment, in step S1, incident light to be imaged is acquired.

[0088] By setting a light-limiting hole at the opening of the aluminum alloy shell, the incident light passes through the light-limiting hole into the aluminum alloy shell. The light-limiting hole is used to limit the diffraction angle of the incident light after passing through the optical diffraction element. The light-limiting hole of the corresponding size can be selected as needed to ensure that the incident light that meets the requirements is finally incident on the photosensitive area of ​​the image sensor.

[0089] In this embodiment, in step S2, the incident light to be imaged is encoded by the optical diffraction element to obtain the encoded incident light.

[0090] Specifically, the incident light to be imaged is encoded by the optical diffraction element, and the formula is:

[0091]

[0092] Among them, v[x, y, λ] is the incident light, h[x, y] is the point spread function, crop represents the discrete two-dimensional linear convolution in the spatial dimension, and w[x, y, λ] is the encoded incident light.

[0093] The incident light that passes through the light-limiting hole and enters the aluminum alloy housing is encoded by the optical diffraction element to achieve the first encoding, thereby obtaining the incident light encoded by the optical diffraction element.

[0094] Specifically, (1) a phase-type optical diffraction element is selected.

[0095] The high transmittance of the phase-type optical diffraction element can improve the utilization rate of the incident light and enhance the sensitivity of the system.

[0096] (2) Map the point light source into a specific pattern and use the principle of optical division multiplexing to reduce the loss of incident light information.

[0097] (3) Fourier domain index based on modulation transfer function: Modulation Transfer Function volume (MTFv), the formula is:

[0098]

[0099] Among them, f x ,f y is the Fourier frequency, Ω is the area of ​​the design space, and MTF is the Fourier transform of the spectral point spread function PSF.

[0100] The MTFv is introduced to objectively measure the amount of information collected in the lensless imaging system. The larger the value of this indicator, the more information the system encodes.

[0101] (4) Based on the modulation transfer function volume index, the Voronoi-Fresnel phase function is used as the basic unit, and the formula is:

[0102]

[0103] Where, ξ,η,i=1,2,3……K, K is the center position of the Voronoi-Fresnel unit, A i is the aperture function, defined by the vertices of the i-th sub-region.

[0104] They are closely arranged in the form of a Voronoi diagram in two-dimensional space, generating a point spread function that has both sparse and high contrast characteristics. Moreover, the polygonal aperture of each Voronoi unit represents a directional filter in the Fourier domain and has a more compact spatial distribution.

[0105] The Voronoi-Fresnel phase function is similar to the lensless phase function of an insect compound eye, which is derived from the Fresnel function of an ideal lens.

[0106] (5) By optimizing the distribution of these basic units, a phase function with better frequency domain characteristics and optical performance is obtained, thereby improving the system performance.

[0107] In this embodiment, in step S3, the encoded incident light is wavelength-encoded by a broadband quantum dot spectral chip, and the incident light is converted into an electrical signal, and the electrical signal is spectrally reconstructed to obtain spectral imaging.

[0108] The encoded incident light is wavelength encoded through a broadband quantum dot spectrum chip, and the formula is:

[0109] L[x,y]=F λ [x,y]·w[x,y,λ](4)

[0110] Where L[x,y] is the incident light entering the image sensor after wavelength encoding, · represents the dot product,

[0111] F λ [x, y] is the filter function of the quantum dot array, and w[x, y, λ] is the incident light that enters the quantum dot filter array after being encoded by the optical diffraction element.

[0112] The incident light encoded by the quantum dot filter array is detected by the image sensor, which converts the incident light into an electrical signal and reconstructs the electrical signal to obtain spectral imaging.

[0113] The encoded incident light is wavelength-encoded through a broadband quantum dot spectrum chip, which is double encoding. This can improve the utilization rate of incident light while increasing the number of spectral channels and prevent the loss of effective information.

[0114] At the same time, the spectral chip converts the incident light into electrical signals, and the detected signals are uploaded to the host computer for spectrum reconstruction.

[0115] The information encoded by the optical diffraction element and the quantum dot filter array is decoded by a decoding algorithm. The formula of the information encoded by the optical diffraction element and the quantum dot filter array is:

[0116]

[0117] Where b represents the information encoded by the optical diffraction element and the quantum dot filter array, and F λ [x, y] is the filter function of the quantum dot array, v[x, y, λ] is the incident light, h[x, y] is the point spread function, and crop represents the discrete two-dimensional linear convolution in the spatial dimension. By decoding formula (5), spectral imaging can be obtained.

[0118] In response to the design requirements of the hyperspectral imaging system for compact structure, broadband, high performance and low cost, this embodiment proposes a lensless hyperspectral imaging system design based on wavelength-phase dual encoding, which combines the phase-type lensless system with the broadband quantum dot spectrum chip, integrates the optical diffraction element, quantum dot filter and image sensor into the same spectrum imaging system, compresses the system structure, improves the utilization rate of the incident light, and based on the theory of compressed sensing, can realize the design of the number of multi-spectral channels with only a small number of filters. The entire system design uses a phase-type optical diffraction element to map a point light source into a surface light source with a specific image, and then uses a broadband quantum dot filter to encode the wavelength to improve the utilization rate of the incident light. At the same time, through the dual encoding of wavelength and phase and the decoding algorithm, the coupling problem of spectral resolution and spatial resolution is solved, thereby realizing the design of a hyperspectral imaging system with compact structure and excellent performance, and providing technical support for the research and development and engineering application of micro hyperspectral imaging systems.

[0119] The steps involved in the apparatus of the above embodiment correspond to the method embodiment 1, and the specific implementation method can refer to the relevant description part of embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode or carry an instruction set for execution by a processor and enable the processor to execute any method in the present invention.

[0120] Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0121] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A lensless hyperspectral imaging system based on wavelength-phase dual encoding, characterized in that: include: Aluminum alloy housing, optical diffraction element and quantum dot spectrum chip; The aluminum alloy housing is provided with a limited light hole; The center positions of the light-limiting hole, the optical diffraction element and the quantum dot spectrum chip are on the same plane; The light-limiting hole is used to limit the diffraction angle of the incident light after passing through the optical diffraction element, and the diffraction angle of the incident light is related to the size of the light-limiting hole; The optical diffraction element is used to encode the incident light passing through to obtain the incident light encoded by the optical diffraction element; The quantum dot spectrum chip is used to encode the wavelength of the incident light after the optical diffraction element is encoded, and at the same time convert the incident light into an electrical signal and reconstruct the signal spectrum.

2. The lensless hyperspectral imaging system based on wavelength-phase dual encoding according to claim 1, characterized in that: The optical diffraction element and the quantum dot spectrum chip are packaged inside the aluminum alloy shell.

3. The lensless hyperspectral imaging system based on wavelength-phase dual encoding according to claim 1, characterized in that: The quantum dot spectrum chip is a broadband quantum dot spectrum chip, which includes a broadband quantum dot filter array and an image sensor.

4. The lensless hyperspectral imaging system based on wavelength-phase dual encoding according to claim 1, characterized in that: The broadband quantum dot filter array adopts quantum dot materials with broad absorption peaks.

5. The lensless hyperspectral imaging system based on wavelength-phase dual encoding according to claim 1, characterized in that: The broadband quantum dot filter array adopts a periodic structure, and the periodic structure is: the broadband quantum dot wave array is divided into 183×230 periodic units, and each periodic unit contains 3×3 quantum dot arrays.

6. The lensless hyperspectral imaging system based on wavelength-phase dual encoding according to claim 1, characterized in that: The size of the broadband quantum dot filter array is 5.5 mm×6.9 mm; the number of broadband quantum dot filter arrays is 550×690; and the size of each quantum dot is 10 μm.

7. The lensless hyperspectral imaging system based on wavelength-phase dual encoding according to claim 1, characterized in that: The size of the image sensor is consistent with the size of the broadband quantum dot filter array.

8. A lensless hyperspectral imaging method based on wavelength-phase dual encoding, characterized in that: include: Acquiring incident light to be imaged; The incident light to be imaged is encoded by the optical diffraction element to obtain the encoded incident light; The encoded incident light is wavelength-encoded through a broadband quantum dot spectral chip, and the incident light is converted into an electrical signal, and the electrical signal is reconstructed to obtain spectral imaging.

9. The lensless hyperspectral imaging method based on wavelength-phase dual encoding as claimed in claim 8, characterized in that: The incident light to be imaged is encoded by the optical diffraction element to obtain the encoded incident light, and the formula is: Among them, v[x, y, λ] is the incident light, h[x, y] is the point spread function, crop represents the discrete two-dimensional linear convolution in the spatial dimension, and w[x, y, λ] is the encoded incident light.

10. The lensless hyperspectral imaging method based on wavelength-phase dual encoding according to claim 8, characterized in that: The encoded incident light is wavelength encoded through a broadband quantum dot spectrum chip, and the formula is: L[x,y]=F λ [x,y]·w[x,y,λ]: Where L[x,y] is the incident light entering the image sensor after wavelength encoding, · represents the dot product, F λ [x, y] is the filter function of the quantum dot array, and w[x, y, λ] is the incident light that enters the quantum dot filter array after being encoded by the optical diffraction element.

Images