Multispectral camera based on single CMOS array sensor and imaging method

Abstract

The invention discloses a multispectral camera based on a single CMOS array sensor and an imaging method, and relates to the technical field of image processing. The multispectral camera comprises a single CMOS array sensor and a PCB control board. The single CMOS array sensor is provided with a plurality of pixels, adjacent pixels are taken as a group, coating and filtering are carried out according to wave bands, and detection signals of the wave bands are obtained; a control program is embedded in the PCB control board, and the PCB control board firstly preprocesses a detection signal to generate a meta-image; splitting the element image according to the pixel group to generate a multi-band image; interpolating the single-band image by adopting a cubic convolution method to obtain a full-resolution single-band image; and finally, a Gram-Schmidt algorithm is adopted to carry out fusion sharpening on the wave band with the low resolution based on the wave band with the high resolution, a multi-band image with the improved resolution is obtained, and therefore high-resolution multispectral output is achieved.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, and in particular to a multispectral camera and imaging method based on a single CMOS array sensor. Background Technology

[0002] With the increasing demand for multispectral imaging technology in fields such as remote sensing monitoring, precision agriculture, and environmental monitoring, traditional multispectral cameras that use multiple sensors or beam-splitting prism systems to achieve spectral separation in different bands face the following technical bottlenecks:

[0003] 1) High Hardware Complexity: Multi-sensor beam splitting systems employ multiple independent sensors (such as CCD / CMOS) equipped with filters for different wavelengths. An optical beam splitter separates the incident light to each sensor, enabling parallel wavelength acquisition. For example, a quadrature prism beam splitting system divides visible and near-infrared light into 4-6 wavelengths, allowing for simultaneous imaging by four sensors. Multi-sensor systems require precise optical registration, making optical assembly and adjustment difficult. After long-term use, sensor response drift necessitates frequent calibration. Furthermore, the equipment's size and weight increase significantly, resulting in high costs and making it difficult to meet lightweight design requirements.

[0004] 2) Limited real-time performance and high cost: Beam-splitter / filter wheel systems, based on beam-splitter prisms (such as the Offner structure) or rotating filter wheels, time-division switching of wavelength light to a single sensor, achieving multispectral reconstruction through time-series exposure. The filter wheel system's mechanical structure has a lifespan of only about 500,000 rotations. Time-division imaging in dynamic scenes leads to motion artifacts, affecting the accuracy of dynamic scene analysis. Beam-splitter prisms also suffer from band crosstalk, such as cross-contamination between visible and near-infrared light. Furthermore, beam-splitter / filter wheel systems rely on complex structures, resulting in high maintenance costs.

[0005] 3) Low resolution: Single-sensor multispectral solutions use pixel-level filtering (such as Bayer arrays) or microlenses to split the light, allowing a single sensor to capture different bands in different regions. For example, in an RGB-NIR filter array, pixels are distributed in a 2×2 period, covering 4 bands. The number of bands in a single-sensor Bayer array is ≤4, which cannot meet the needs of agricultural monitoring (requiring 6-10 bands); pixel utilization is low (only 25% is used for effective bands), resulting in significant spatial resolution loss. Summary of the Invention

[0006] To address the aforementioned problems, the present invention aims to provide a multispectral camera and imaging method based on a single CMOS array sensor. Through single-CMOS pixel-level filter design, it achieves simultaneous imaging across multiple bands (6-10 bands), overcoming the band limitations of traditional Bayer arrays (≤ 4 bands). The multispectral imaging module is integrated into a conventional CMOS sensor, achieving device miniaturization, reducing size by more than 50%, and lowering costs by 30%-50%. Simultaneous exposure of all band pixels eliminates temporal errors in time-division imaging, making it suitable for dynamic scenes (such as moving target detection). Combining cubic convolution interpolation with the Gram-Schmidt fusion algorithm, it reconstructs high-resolution multispectral images under single-sensor hardware conditions, solving the problems of insufficient bands and resolution loss in traditional solutions, and achieving high-resolution multispectral output.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0008] On the one hand, a multispectral camera based on a single CMOS array sensor is provided, the multispectral camera including a single CMOS array sensor and a PCB control board;

[0009] The single CMOS array sensor has multiple pixels, wherein the pixels are arranged in adjacent order. Each pixel is grouped together, and the film is applied and filtered according to the wavelength band to obtain... Detection signals in each band;

[0010] The PCB control board embeds a control program that first preprocesses the detection signal to generate a meta-image; then, it decomposes the meta-image into pixel groups to generate a multi-band image; next, it uses cubic convolution to interpolate the single-band image to obtain a full-resolution single-band image; finally, it uses the Gram-Schmidt algorithm to fuse and sharpen the lower-resolution bands based on the higher-resolution bands to obtain a multi-band image with improved resolution.

[0011] Optionally, the multispectral camera further includes an optical lens that captures light from the subject and focuses it onto the single CMOS array sensor.

[0012] Optionally, the optical lens, the single CMOS array sensor, and the PCB control board are assembled in a housing structure to form the multispectral camera.

[0013] Optionally, the detection signal of the single CMOS array sensor is a current signal, and a meta-image is generated by amplifying and denoising the current signal.

[0014] Optionally, by employing the Gram-Schmidt algorithm, the lower-resolution B1-B6 bands are fused and sharpened based on the higher-resolution PB band, resulting in a multi-band image with improved resolution.

[0015] On the other hand, an imaging method for a multispectral camera based on a single CMOS array sensor is provided, the method comprising the following steps:

[0016] S1. Acquire detection signals using a single CMOS array sensor, with the nearest neighboring [element / substrate] being the most [specifically, the sensor]. Each pixel is grouped together, and the film is applied and filtered according to the wavelength band to obtain... Detection signals in each band;

[0017] S2. The PCB control board embeds a control program to preprocess the detection signal and generate a meta-image.

[0018] S3. Decompose the meta-image into pixel groups to generate a multi-band image;

[0019] S4. Interpolate the single-band image using the cubic convolution method to obtain a full-resolution single-band image;

[0020] S5. Using the Gram-Schmidt algorithm, the lower-resolution bands are fused and sharpened based on the higher-resolution bands to obtain a multi-band image with improved resolution.

[0021] Optionally, the detection signal of the single CMOS array sensor is a current signal, and in step S2, a meta-image is generated by amplifying and denoising the current signal.

[0022] Optionally, in step S3, the pixel value image sequence of the multi-band image is generated as follows:

[0023]

[0024]

[0025]

[0026]

[0027]

[0028]

[0029] Where h and v represent the number of rows and columns of a single CMOS array sensor pixel, respectively, n and m represent the number of rows and columns of the neighboring pixels, respectively, and B1, B2, B3, B4, B5, B6, and PB represent bands.

[0030] Optionally, in step S4, a cubic convolution method is used to interpolate the single-band image, and the calculation formula is as follows:

[0031]

[0032]

[0033] Where W(d) represents the cubic convolution kernel function, used to calculate the weight of the neighboring pixel to the target pixel during interpolation, d represents the distance between the target pixel and the neighboring pixel in the x or y direction, and a represents the shape parameter of the cubic convolution kernel.

[0034] f(x, y) represents the pixel value at the target location (x, y) after three convolutional interpolations. i y i ) indicates the area around the target pixel The original value of the (i, j)th pixel in the neighborhood, where i and j range from 0 to 3, with a total of Each neighboring cell participates in the interpolation;

[0035] W(xx i ) represents the target location x and its neighboring pixels x i The weights, W(yy), are calculated using a cubic convolution kernel function to represent the distance in the x-direction. i ) represents the target location y and its neighboring pixels y i The weights are calculated using a cubic convolution kernel function to represent the distance in the y-direction.

[0036] Optionally, in step S5, the Gram-Schmidt algorithm is used to perform fusion and sharpening on the lower resolution B1-B6 bands based on the higher resolution PB band. The calculation process is as follows:

[0037]

[0038]

[0039]

[0040]

[0041] GS T (i, j) represents the value of the T-th band at pixel (i, j) after Gram-Schmidt orthogonalization, B T(i, j) represents the value of the Tth original low-resolution band at pixel (i, j); μ T This represents the mean of the T-th band. GS represents the correlation coefficient between the T-th band and the l-th orthogonalized band. l (i, j) represents the value of the l-th orthogonalized band at pixel (i, j);

[0042] C represents the number of columns in the image, and R represents the number of rows in the image; σ(B T GS l ) represents the covariance between the T-th band and the l-th orthogonalized band, σ(GS) l GS l ) 2 Let σ represent the variance of the l-th orthogonalized band. T This represents the standard deviation of the T-th band.

[0043] The beneficial effects of the technical solution provided by this invention include at least the following:

[0044] This invention achieves a breakthrough in multispectral imaging through the innovative fusion of single-CMOS pixel-level filtering and computational imaging technologies. Its core advantages are: First, the hardware structure employs a pixel-level filter array design, enabling simultaneous acquisition of 6-10 bands of data in a single exposure, significantly increasing spectral information compared to traditional Bayer arrays (≤4 bands). Simultaneously, it eliminates the need for a multi-sensor spectral splitting system, reducing device size by over 50% and cost by 30%-50%. Second, through cubic convolution interpolation and Gram-Schmidt algorithm optimization, spatial resolution is improved by 20%-30% while simplifying hardware, and pixel utilization is increased from 25% to over 80%. Furthermore, the full-band simultaneous exposure characteristic effectively eliminates motion artifacts in dynamic scenes. This invention provides a lightweight and cost-effective solution for precision agriculture (such as crop health monitoring) and environmental inspection (drone pollution detection), resolving the contradictions of high cost, low real-time performance, and band limitations inherent in traditional technologies. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a schematic diagram of a single CMOS array sensor provided in an embodiment of the present invention;

[0047] Figure 2 This is a schematic diagram of wavelength band coating and filtering provided in an embodiment of the present invention;

[0048] Figure 3 This is a schematic diagram of the generated pixel image provided in an embodiment of the present invention;

[0049] Figure 4 This is a schematic diagram of the multispectral camera structure provided in an embodiment of the present invention;

[0050] Figure 5 This is a flowchart of the imaging method provided in an embodiment of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0052] In embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Rather, the term "exemplary" is used to present concepts in a specific manner.

[0053] This invention provides a multispectral camera based on a single CMOS array sensor, the multispectral camera comprising a single CMOS array sensor and a PCB control board. (Reference) Figure 1 and Figure 2 As shown, the single CMOS array sensor has multiple pixels, wherein the pixels are arranged in adjacent order. Each pixel is grouped together, and the film is applied and filtered according to the wavelength band to obtain... Detection signals in each band.

[0054] The PCB control board embeds a control program that processes the detection signal as follows:

[0055] First, the detected signal is preprocessed to generate a meta-image. The detected signal from a single CMOS array sensor is a current signal. This current signal is amplified and denoised to generate a meta-image. (Refer to...) Figure 3 As shown.

[0056] The meta-image is then decomposed into pixel groups to generate a multi-band image. Next, a cubic convolution method is used to interpolate the single-band image to obtain a full-resolution single-band image. Finally, the Gram-Schmidt algorithm is used to fuse and sharpen the lower-resolution bands based on the higher-resolution bands to obtain a multi-band image with improved resolution.

[0057] Specifically, by employing the Gram-Schmidt algorithm, based on the higher-resolution PB band, the lower-resolution B1-B6 bands are fused and sharpened to obtain a multi-band image with improved resolution.

[0058] In this embodiment of the invention, a single CMOS pixel-level filter array design is adopted to achieve simultaneous imaging in 6-10 bands, breaking through the band limitations of traditional Bayer arrays. In terms of hardware, the multi-sensor beam splitting system is eliminated, reducing the volume by more than 50% and the cost by 30%-50%. In terms of algorithms, spatial resolution is improved by 20%-30% through cubic convolution interpolation and Gram-Schmidt algorithm, and the pixel utilization rate reaches more than 80%. It can meet the needs of precision agriculture, environmental inspection and other fields for lightweight, cost-effective multispectral imaging.

[0059] Furthermore, the multispectral camera also includes an optical lens that captures light from the subject and focuses it onto the single CMOS array sensor. (Reference) Figure 4 As shown, the optical lens, the single CMOS array sensor, and the PCB control board are assembled in the housing structure to form the multispectral camera.

[0060] Accordingly, embodiments of the present invention also provide an imaging method for a multispectral camera based on a single CMOS array sensor, such as... Figure 5 As shown, the method includes the following steps:

[0061] S1. Acquire detection signals using a single CMOS array sensor, with the nearest neighboring [element / substrate] being the most [specifically, the sensor]. Each pixel is grouped together, and the film is applied and filtered according to the wavelength band to obtain... Detection signals in each band.

[0062] After the optical lens collects the light from the object being photographed, it focuses it onto a single CMOS array sensor, generating a current signal as a detection signal.

[0063] S2. The PCB control board embeds a control program to preprocess the detection signal and generate a meta-image.

[0064] The control program is used to amplify and denoise the current signal to generate a meta-image.

[0065] S3. Decompose the meta-image into pixel groups to generate multi-band images. The pixel value image sequence of the multi-band images is as follows:

[0066]

[0067]

[0068]

[0069]

[0070]

[0071]

[0072] Where h and v represent the number of rows and columns of a single CMOS array sensor pixel, respectively, n and m represent the number of rows and columns of the neighboring pixels, respectively, and B1, B2, B3, B4, B5, B6, and PB represent bands.

[0073] S4. Interpolate the single-band image using the cubic convolution method to obtain a full-resolution single-band image. The calculation formula is as follows:

[0074]

[0075]

[0076] Where W(d) represents the cubic convolution kernel function, used to calculate the weight of the neighboring pixel to the target pixel during interpolation, d represents the distance between the target pixel and the neighboring pixel in a certain direction (such as the x or y direction), and a represents the shape parameter of the cubic convolution kernel. Different values ​​of a will affect the shape of the convolution kernel, and thus affect the smoothness and detail preservation of the interpolation result. Usually, a value such as a = -0.5 can better balance smoothness and detail.

[0077] Different intervals of the piecewise function: based on the distance between the target pixel and its neighboring pixels. Different function expressions are used to calculate weights for pixels in different ranges to ensure that the weight calculation is more precise when the distance is close, and the weight is 0 when the distance exceeds the effective range, that is, pixels outside a certain range have no interpolation contribution to the target pixel.

[0078] f(x, y) represents the pixel value at the target location (x, y) after three convolutional interpolations. i y i ) indicates the area around the target pixel The original value of the (i, j)th pixel in the neighborhood, where i and j range from 0 to 3, with a total of Each neighboring cell participates in the interpolation.

[0079] W(xx i ) represents the target location x and its neighboring pixels x i The weights, W(yy), are calculated using a cubic convolution kernel function to represent the distance in the x-direction. i ) represents the target location y and its neighboring pixels y i The weights are calculated using a cubic convolution kernel function to represent the distance in the y-direction.

[0080] Double summation For all pixels in a 4×4 neighborhood around the target pixel, calculate their weights in the x and y directions respectively, and then sum the products of the neighborhood pixel values ​​and their corresponding weights to obtain the interpolated pixel values ​​at the target location, thereby achieving resolution enhancement (interpolation) of a single-band image.

[0081] S5. Using the Gram-Schmidt algorithm, the lower-resolution bands are fused and sharpened based on the higher-resolution bands to obtain a multi-band image with improved resolution.

[0082] Specifically, the Gram-Schmidt algorithm is used to perform fusion and sharpening on the lower-resolution B1-B6 bands, based on the higher-resolution PB band. The calculation process is as follows:

[0083]

[0084]

[0085]

[0086]

[0087] GS T (i, j) represents the value of the T-th band at pixel (i, j) after Gram-Schmidt orthogonalization, B T (i, j) represents the value of the Tth original low-resolution band at pixel (i, j); μ T This represents the mean of the T-th band. GS represents the correlation coefficient between the T-th band and the l-th orthogonalized band. l (i, j) represents the value of the l-th orthogonalized band at pixel (i, j);

[0088] C represents the number of columns in the image, and R represents the number of rows in the image; σ(B T GS l ) represents the covariance between the T-th band and the l-th orthogonalized band, σ(GS) lGS l ) 2 Let σ represent the variance of the l-th orthogonalized band. T This represents the standard deviation of the T-th band.

[0089] This invention achieves a breakthrough in multispectral imaging through the innovative fusion of single-CMOS pixel-level filtering and computational imaging technologies. Its core advantages are: First, the hardware structure employs a pixel-level filter array design, enabling simultaneous acquisition of 6-10 bands of data in a single exposure, significantly increasing spectral information compared to traditional Bayer arrays (≤4 bands). Simultaneously, it eliminates the need for a multi-sensor spectral splitting system, reducing device size by over 50% and cost by 30%-50%. Second, through cubic convolution interpolation and Gram-Schmidt algorithm optimization, spatial resolution is improved by 20%-30% while simplifying hardware, and pixel utilization is increased from 25% to over 80%. Furthermore, the full-band simultaneous exposure characteristic effectively eliminates motion artifacts in dynamic scenes. This invention provides a lightweight and cost-effective solution for precision agriculture (such as crop health monitoring) and environmental inspection (drone pollution detection), resolving the contradictions of high cost, low real-time performance, and band limitations inherent in traditional technologies.

[0090] In an exemplary embodiment, the present invention also provides an electronic device, the electronic device comprising:

[0091] processor;

[0092] A memory storing computer-readable instructions, which, when loaded and executed by the processor, implement the steps of the imaging method for a multispectral camera based on a single CMOS array sensor as described above.

[0093] In an exemplary embodiment, the present invention also provides a computer-readable storage medium storing at least one instruction, which is loaded and executed by a processor to implement the steps of the imaging method for a multispectral camera based on a single CMOS array sensor as described above. For example, the computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, or optical data storage device, etc.

[0094] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0095] The use of terms such as "an embodiment," "an embodiment," "an exemplary embodiment," and "some embodiments" in the specification indicates that the described embodiment may include a specific feature, structure, or characteristic, but not every embodiment necessarily includes that specific feature, structure, or characteristic. Furthermore, when a specific feature, structure, or characteristic is described in connection with an embodiment, implementing such a feature, structure, or characteristic in conjunction with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the art.

[0096] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0097] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0098] In the embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0099] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0100] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0101] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0102] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0103] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multispectral camera based on a single CMOS array sensor, characterized in that, The multispectral camera includes a single CMOS array sensor and a PCB control board; The single CMOS array sensor has multiple pixels, wherein the pixels are arranged in adjacent order. Each pixel is grouped together, and the film is applied and filtered according to the wavelength band to obtain... Detection signals in each band; The PCB control board embeds a control program that first preprocesses the detection signal to generate a meta-image; then, it decomposes the meta-image into pixel groups to generate a multi-band image; next, it uses cubic convolution to interpolate the single-band image to obtain a full-resolution single-band image; finally, it uses the Gram-Schmidt algorithm to fuse and sharpen the lower-resolution bands based on the higher-resolution bands to obtain a multi-band image with improved resolution.

2. The multispectral camera based on a single CMOS array sensor according to claim 1, characterized in that, The multispectral camera also includes an optical lens that captures light from the subject and focuses it onto the single CMOS array sensor.

3. The multispectral camera based on a single CMOS array sensor according to claim 2, characterized in that, The optical lens, the single CMOS array sensor, and the PCB control board are assembled in the housing structure to form the multispectral camera.

4. The multispectral camera based on a single CMOS array sensor according to claim 1, characterized in that, The detection signal of the single CMOS array sensor is a current signal. By amplifying and denoising the current signal, a meta-image is generated.

5. The multispectral camera based on a single CMOS array sensor according to claim 1, characterized in that, By employing the Gram-Schmidt algorithm, based on the higher-resolution PB band, the lower-resolution B1-B6 bands are fused and sharpened to obtain a multi-band image with improved resolution.

6. An imaging method for a multispectral camera based on a single CMOS array sensor, characterized in that, The multispectral camera is the multispectral camera according to any one of claims 1 to 5, and the method includes the following steps: S1. Acquire detection signals using a single CMOS array sensor, with the nearest neighboring [element / substrate] being the most [specifically, the sensor]. Each pixel is grouped together, and the film is applied and filtered according to the wavelength band to obtain... Detection signals in each band; S2. The PCB control board embeds a control program to preprocess the detection signal and generate a meta-image. S3. Decompose the meta-image into pixel groups to generate a multi-band image; S4. Interpolate the single-band image using the cubic convolution method to obtain a full-resolution single-band image; S5. Using the Gram-Schmidt algorithm, the lower-resolution bands are fused and sharpened based on the higher-resolution bands to obtain a multi-band image with improved resolution.

7. The imaging method for a multispectral camera based on a single CMOS array sensor according to claim 6, characterized in that, The detection signal of the single CMOS array sensor is a current signal. In step S2, the current signal is amplified and denoised to generate a meta-image.

8. The imaging method for a multispectral camera based on a single CMOS array sensor according to claim 6, characterized in that, In step S3, the pixel value image sequence of the multi-band image is generated as follows: ； ； ； ； ； ； ； Where h and v represent the number of rows and columns of a single CMOS array sensor pixel, respectively, n and m represent the number of rows and columns of the neighboring pixels, respectively, and B1, B2, B3, B4, B5, B6, and PB represent bands.

9. The imaging method for a multispectral camera based on a single CMOS array sensor according to claim 6, characterized in that, In step S4, the single-band image is interpolated using a cubic convolution method, and the calculation formula is as follows: ； ； Where W(d) represents the cubic convolution kernel function, used to calculate the weight of the neighboring pixel to the target pixel during interpolation, d represents the distance between the target pixel and the neighboring pixel in the x or y direction, and a represents the shape parameter of the cubic convolution kernel. f(x, y) represents the pixel value at the target location (x, y) after three convolutional interpolations. i y i ) indicates the area around the target pixel The original value of the (i, j)th pixel in the neighborhood, where i and j range from 0 to 3, with a total of Each neighboring cell participates in the interpolation; W(xx i ) represents the target location x and its neighboring pixels x i The weights, W(yy), are calculated using a cubic convolution kernel function to represent the distance in the x-direction. i ) represents the target location y and its neighboring pixels y i The weights are calculated using a cubic convolution kernel function to represent the distance in the y-direction.

10. The imaging method for a multispectral camera based on a single CMOS array sensor according to claim 6, characterized in that, In step S5, the Gram-Schmidt algorithm is used to perform fusion and sharpening on the lower resolution B1-B6 bands based on the higher resolution PB band. The calculation process is as follows: ； ； ； ； GS T (i, j) represents the value of the T-th band at pixel (i, j) after Gram-Schmidt orthogonalization, B T (i, j) represents the value of the Tth original low-resolution band at pixel (i, j); μ T This represents the mean of the T-th band. GS represents the correlation coefficient between the T-th band and the l-th orthogonalized band. l (i, j) represents the value of the l-th orthogonalized band at pixel (i, j); C represents the number of columns in the image, and R represents the number of rows in the image; σ(B T GS l ) represents the covariance between the T-th band and the l-th orthogonalized band, σ(GS) l GS l ) 2 Let σ represent the variance of the l-th orthogonalized band. T This represents the standard deviation of the T-th band.

Images