Single-exposure hyperspectral imaging device based on orthogonal mask

Abstract

The utility model discloses a single-exposure hyperspectral imaging device based on an orthogonal mask, and the used orthogonal mask can significantly improve the final pattern quality of an imaging system, including peak signal-to-noise ratio, structural similarity and spectral accuracy. Compared with a newly disclosed single-exposure spectral imaging technology using a color mask in recent years, the orthogonal mask used by the utility model is a binary spatial modulation pattern, the obtained modulation pattern is a binary pattern, the manufacturing process is simple and mature, the cost is low, spectrum modulation with the same efficiency or even higher efficiency can be realized, and the imaging quality is improved. Therefore, the method is not limited by cost and operation complexity.

Description

Technical Field

[0001] This utility model relates to the field of hyperspectral imaging technology, and more specifically, to a single-exposure hyperspectral imaging device based on orthogonal masks. Background Technology

[0002] Traditional color imaging techniques can only obtain information from three spectral bands: red, green, and blue. In contrast, hyperspectral imaging can provide information from up to dozens of spectral channels, and therefore has been widely used in many fields, such as biomedicine, space remote sensing, environmental monitoring, industrial inspection, and archaeological identification.

[0003] Conventional hyperspectral imaging techniques typically rely on scanning, measuring hyperspectral data of objects point-by-point, line-by-line, or surface-by-surface, resulting in slow imaging speeds. The theory of compressed sensing, proposed in 2006, proved the feasibility of reconstructing the original information of an object through multiple samplings. Subsequently, in 2007-2008, researchers developed single-exposure hyperspectral imaging based on this theory. This technique can obtain complete hyperspectral information of an object using only a single exposure with a grayscale camera. The technique first encodes the spatial and spectral dimensions of the object using an amplitude mask and a dispersive prism, then performs compressed measurements using a single exposure with a grayscale camera, and finally reconstructs the hyperspectral pattern of the object from the compressed measurements using a reconstruction algorithm.

[0004] In early single-exposure hyperspectral imaging techniques, the spatial distribution of the amplitude mask used was typically random, resulting in poor quality of the final reconstructed pattern. Subsequent techniques introduced colored random masks, improving the mask's modulation efficiency of the object's spectral information and thus enhancing the imaging results to some extent.

[0005] However, such technologies also bring new challenges: specifically, they make the system more complex and expensive. This complexity is reflected not only in the design and manufacture of the mask, but also in the calibration and maintenance of the entire imaging system. Therefore, although single-exposure hyperspectral imaging technology has great potential in theory, its practical application is still limited by cost and operational complexity. Utility Model Content

[0006] The technical problem this invention aims to solve is how to overcome the limitations of existing single-exposure hyperspectral imaging technology due to cost and operational complexity. To overcome these limitations, this invention provides a single-exposure hyperspectral imaging device based on orthogonal masks.

[0007] This invention provides a single-exposure hyperspectral imaging device based on orthogonal masks, the single-exposure hyperspectral imaging device comprising the following components arranged sequentially along the direction of hyperspectral imaging information propagation formed by the object:

[0008] An object-space dispersive prism is used to disperse the hyperspectral imaging information formed by an object to obtain an object-space dispersive pattern.

[0009] An orthogonal mask is used to perform binary space modulation on the object-side dispersion pattern to obtain a modulation pattern;

[0010] An image-square dispersive prism, whose prism orientation is opposite to that of the object-square dispersive prism, is used to disperse the modulation pattern to obtain an image-square dispersive pattern.

[0011] A grayscale pattern sensor is used to perform exposure measurement on the image dispersion pattern and obtain the measurement result.

[0012] This invention discloses a single-exposure hyperspectral imaging device based on orthogonal masks. Compared with existing technologies, it has the advantage of simple structure. The orthogonal mask used can significantly improve the final pattern quality of the imaging system, including peak signal-to-noise ratio, structural similarity, and spectral accuracy. Compared with the single-exposure spectral imaging technology using color masks that has been disclosed in recent years, the orthogonal mask used in this invention modulates a binary (i.e., black and white) spatial pattern. The resulting modulation pattern is a binary pattern, which has a simple and mature manufacturing process, low cost, and can achieve spectral modulation with equal or even higher efficiency, thus not being limited by cost and operational complexity.

[0013] In one possible implementation, the orthogonal mask is a matrix composed of multiple light-transmitting units and light-blocking units, and in each row of the matrix, the same number of light-blocking units are provided between two adjacent light-transmitting units; thereby enabling binary space modulation of the object-side dispersion pattern and possessing orthogonal characteristics.

[0014] In one possible implementation, a first relay lens is provided between the object and the object-side dispersive prism along the direction of light signal propagation, and a second relay lens is provided between the object-side dispersive prism and the orthogonal mask; thereby enabling the hyperspectral imaging information formed by the object to be relayed to the orthogonal mask, and also enabling control of the dispersion amplitude to improve imaging quality.

[0015] In one possible implementation, a third relay lens is provided between the orthogonal mask and the image-side dispersive prism along the direction of light signal propagation, and a fourth relay lens is provided between the image-side dispersive prism and the grayscale pattern sensor. This scheme can not only relay the modulation pattern to the panel of the grayscale pattern sensor, but also control the dispersion amplitude again, thereby further improving the imaging quality.

[0016] In one possible implementation, the first relay lens, the second relay lens, the third relay lens, and the fourth relay lens are all achromatic lenses; thereby, chromatic aberration can be eliminated on the basis of relay imaging, thereby further improving the imaging quality. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of a single-exposure hyperspectral imaging device based on orthogonal mask disclosed in this embodiment;

[0018] Figure 2 This is a schematic diagram of the orthogonal mask structure disclosed in this embodiment;

[0019] Figure 3 This is a schematic diagram of a mathematical model of a single-exposure hyperspectral imaging device structure based on orthogonal mask disclosed in this embodiment;

[0020] Figure 4 This is a schematic diagram of the orthogonal mask design scheme disclosed in this embodiment.

[0021] Explanation of reference numerals in the attached figures:

[0022] 1. Object-side dispersive prism, 2. Orthogonal mask, 3. Image-side dispersive prism, 4. Grayscale pattern sensor, 5. First relay lens, 6. Second relay lens, 7. Third relay lens, 8. Fourth relay lens. Detailed Implementation

[0023] First, those skilled in the art should understand that these embodiments are merely used to explain the technical principles of the embodiments of this application and are not intended to limit the scope of protection of the embodiments of this application. Those skilled in the art can make adjustments as needed to adapt to specific application scenarios.

[0024] Secondly, in the embodiments of this application, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0025] The present application will be further described in detail below with reference to an embodiment, the accompanying drawings, and specific embodiments.

[0026] See Figures 1-4This application discloses a single-exposure hyperspectral imaging device based on orthogonal masks, see [link to relevant documentation]. Figure 1 The single-exposure hyperspectral imaging device includes a first relay lens 5, an object-side dispersive prism 1, a second relay lens 6, an orthogonal mask 2, a third relay lens 7, an image-side dispersive prism 3, a fourth relay lens 8, and a grayscale pattern sensor 4, arranged sequentially along the direction of propagation of hyperspectral imaging information formed by the object. The object-side dispersive prism 1 disperses the hyperspectral imaging information formed by the object to obtain an object-side dispersive pattern; the orthogonal mask 2 modulates the object-side dispersive pattern in binary space to obtain a modulation pattern; the image-side dispersive prism 3, with its prism orientation opposite to that of the object-side dispersive prism 1, disperses the modulation pattern to obtain an image-side dispersive pattern; and the grayscale pattern sensor 4 performs exposure measurements on the image-side dispersive pattern to obtain measurement results.

[0027] After the hyperspectral imaging information of the object enters the single-exposure hyperspectral imaging device, it is first relayed to the object-side dispersive prism 1 by the first relay lens 5. Then, it is dispersed by the object-side dispersive prism 1 to cause different degrees of lateral displacement of the patterns in different spectral bands. After obtaining the object-side dispersive pattern, it is relayed to the orthogonal mask 2 by the second relay lens 6. The orthogonal mask 2 is carefully modulated in binary space to obtain the modulation pattern. Subsequently, the modulation pattern is relayed to the image-side dispersive prism 3 by the third relay lens 7. It is dispersed by the image-side dispersive prism 3 to obtain the image-side dispersive pattern. The image-side dispersive pattern is then relayed to the panel of the grayscale pattern sensor 4 by the fourth relay lens 8. Exposure measurement is performed by superposition to form the final measurement result.

[0028] See Figure 2 and Figure 4 In this embodiment, the orthogonal mask 2 is a matrix composed of multiple light-transmitting units and light-blocking units. Figure 2 In the matrix, white squares represent light-transmitting units, and black squares represent light-blocking units. In each row of this matrix, the same number of light-transmitting units are placed between any two adjacent light-transmitting units. For each column, the light-transmitting and light-blocking units are randomly arranged. A schematic diagram of the orthogonal mask 2 construction scheme is shown below. Figure 4 , Figure 2 and Figure 4 The first row is shown in the blue box.

[0029] Meanwhile, in this embodiment, the first relay lens 5, the second relay lens 6, the third relay lens 7, and the fourth relay lens 8 are all achromatic lenses. The object is positioned on the object plane of the first relay lens 5, and the object-side dispersive prism 1 is positioned on the focal plane of the first relay lens 5. Figure 1 In this context, f1 represents the focal length of the first relay lens 5. The object-side dispersive prism 1 is positioned on the object plane of the second relay lens 6, while the orthogonal mask 2 is positioned on the focal plane of the second relay lens 6. Figure 1In this context, f2 represents the focal length of the second relay lens 6. The orthogonal mask 2 is positioned on the object plane of the third relay lens 7, and the image-square dispersive prism 3 is positioned on the focal plane of the third relay lens 7. Figure 1 In this context, f3 represents the focal length of the third relay lens 7. The image-square dispersive prism 3 is positioned on the object plane of the fourth relay lens 8, and the grayscale pattern sensor 4 is positioned on the focal plane of the fourth relay lens 8. Figure 1 In this context, f4 represents the focal length of the third relay lens 7.

[0030] In this embodiment, the focal lengths of the first relay lens 5, the second relay lens 6, the third relay lens 7, and the fourth relay lens 8 are all 50mm, the number of units in the orthogonal mask 2 is 512*543, the number of exposure pixels in the grayscale pattern sensor 4 is 512*512, and the number of measurable spectral bands is 32. The grayscale pattern sensor 4 can be a CCD, CMOS, or SCMOS sensor.

[0031] To verify the beneficial effects of this single-exposure hyperspectral imaging device, the mathematical model of the single-exposure hyperspectral imaging device is further disclosed below.

[0032] See Figure 3 ,make A spectral data cube representing hyperspectral imaging information of an object contains N λ Spectral channels, each channel consisting of N x ×N y It consists of spatial pixels. The single-exposure hyperspectral imaging device disclosed in this embodiment mainly involves four steps in measuring X.

[0033] X is first cut by the object-space dispersive prism 1 into... For spectral channel X k ∈R Nx×Ny and have

[0034] Then, the sheared data cube X' is spatially modulated through orthogonal mask 2, which is a matrix that can be represented as follows: A modulation data cube of the same size as X' is obtained through orthogonal mask 2. That is, the modulation pattern, which has each spectral channel It can be represented as

[0035] Z k =X′ k ☉M * (2)

[0036] Here, ☉ represents matrix dot product. Following this, a second shear operation is applied to Z by the image-side dispersive prism 3, in the opposite direction to the first shear, resulting in a twisted modulation data cube. That is, a square dispersion pattern, in which each spectral channel Z′ k ∈R Nx×Ny It can be represented as being M * Spectral channels X modulated by different translation versions k :

[0037] Z′ k =X k ☉M k (3)

[0038] Where M k (i,j)=M * (i,j+k-1). For convenience, let Indicated by M * If a 3D mask is composed of different translation versions, then Z′ can be written as

[0039] Z′=X☉M (4)

[0040] Finally, summing Z′ along the spectral dimension yields the final 2D encoded measurement result.

[0041]

[0042] To ensure the validity of the encoding, M (or M * Certain conditions must be met. In traditional single-exposure spectral compression imaging systems, M... * The matrix elements are usually randomly assigned to ensure The incoherence between them. In this embodiment, M... * Special design is employed to ensure that the orthogonal mask 2 possesses orthogonality, thereby achieving... This forms an orthogonal set, namely M. k1 ☉M k2 =0, and Where J is an all-one matrix.

[0043] The orthogonality of orthogonal mask 2 ensures independent (crosstalk-free) measurements of different spectral channels; that is, each pixel in Y measures only one spectral channel. More precisely, each spectral channel X... k Different subsets of the 4-pixel array of the grayscale pattern sensor are sampled separately. Due to the existence of orthogonality, Y (Equation 5) is compared with the mask M of a certain channel. k Multiplication allows the extraction of the corresponding masked spectral image X from Y. k ☉M k Therefore, the subsequent reconstruction task involves extracting these masked spectral images. The original spectral image set was recovered from the image. This is essentially an image inpainting problem, rather than the compressed sensing reconstruction problem in traditional single-exposure spectral compressed imaging. Certain algorithms can handle this type of reconstruction problem better, thus providing better reconstruction quality.

[0044] In summary, the single-exposure hyperspectral imaging device based on orthogonal masks disclosed in this embodiment has the advantage of simple structure compared with the prior art. The orthogonal mask 2 used can significantly improve the final pattern quality of the imaging system, including peak signal-to-noise ratio, structural similarity, and spectral accuracy. Compared with the single-exposure spectral imaging technology using color masks that has been disclosed in recent years, the orthogonal mask 2 used in this embodiment modulates a binary (i.e., black and white) spatial pattern. The resulting modulation pattern is a binary pattern, which has a simple and mature manufacturing process, low cost, and can achieve spectral modulation with equal or even higher efficiency, thus not being limited by cost and operational complexity.

[0045] In the description of the embodiments of this application, it should be noted that the terms "inner", "outer", "front", "rear", etc., indicating the direction or positional relationship are based on the direction or positional relationship shown in the drawings. This is only for the convenience of description and does not indicate or imply that the device or component must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this application.

[0046] In the description of this application, the references to terms such as "an embodiment," "some embodiments," "in this embodiment," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0047] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A single-exposure hyperspectral imaging device based on orthogonal masks, characterized in that, The single-exposure hyperspectral imaging device comprises, arranged in sequence along a propagation direction of hyperspectral imaging information formed by an object: an object-side dispersion prism (1) for dispersing hyperspectral imaging information formed by an object to obtain an object-side dispersion pattern; an orthogonal mask (2) for binary spatial modulation of the object-side dispersion pattern to obtain a modulation pattern; an image-side dispersion prism (3) with a prism orientation opposite to that of the object-side dispersion prism (1) for dispersing the modulation pattern to obtain an image-side dispersion pattern; a gray pattern sensor (4) for exposure measurement of the image-side dispersion pattern to obtain a measurement result.

2. The orthogonal mask based single-shot hyperspectral imaging device of claim 1, wherein, The orthogonal mask (2) is a matrix composed of a plurality of light transmission units and light shielding units, and in each row of the matrix, the same number of light shielding units are arranged between any two adjacent light transmission units.

3. The orthogonal mask based single-shot hyperspectral imaging apparatus according to claim 1 or 2, wherein, Along the propagation direction of the light signal, a first relay lens (5) is arranged between the object and the object-side dispersion prism (1), and a second relay lens (6) is arranged between the object-side dispersion prism (1) and the orthogonal mask (2).

4. The orthogonal mask based single-shot hyperspectral imaging apparatus of claim 3, wherein, Along the propagation direction of the light signal, a third relay lens (7) is arranged between the orthogonal mask (2) and the image-side dispersion prism (3), and a fourth relay lens (8) is arranged between the image-side dispersion prism (3) and the gray pattern sensor (4).

5. The orthogonal mask based single-shot hyperspectral imaging device of claim 4, wherein, The first relay lens (5), the second relay lens (6), the third relay lens (7), and the fourth relay lens (8) are all achromatic lenses.

Images