Multispectral image sensor and imaging method thereof

Abstract

The present invention relates to a multispectral image sensor and an imaging method thereof, wherein the multispectral image sensor comprises: an image sensor, comprising a substrate and a pixel region located on the surface of the substrate, wherein a plurality of array-arranged pixel units are formed in the pixel region; a filter, wherein a spectral channel region corresponding to the pixel region is formed on the surface of the filter, comprising a plurality of spectral channels corresponding to respective wavelengths arranged in a mosaic form, wherein the side length of each spectral channel is greater than the side length of the pixel unit; the filter is fixed to the surface of the image sensor, wherein the rotational offset angle between the pixel region and the spectral channel region is less than a threshold value, so that the projection of each spectral channel on the pixel region covers at least one complete pixel unit. The multispectral image sensor has a low cost.

Description

Technical Field

[0001] The present invention relates to the field of sensor technology, and in particular to a multispectral image sensor and an imaging method thereof. Background Art

[0002] There are two main manufacturing methods for existing multispectral image sensors. One is to use discrete optical devices such as beam splitters to integrate the optical devices and image sensors. The sensors manufactured by this method are large in size, heavy in weight, and high in cost, and their application scenarios are limited. Another method is to use semiconductor manufacturing methods to create a wedge-shaped or mosaic-shaped filter layer on the surface of the image sensor so that different pixels on the image sensor can only receive light of specific wavelengths. This method also has disadvantages such as high cost and low yield due to the need for multiple photolithography and multi-layer material deposition, and its application is therefore limited.

[0003] How to form a small-volume, low-cost multispectral image sensor is a problem that needs to be solved at present. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a multi-spectral image sensor and an imaging method thereof, wherein the multi-spectral image sensor is small in size and low in cost.

[0005] In order to solve the above problems, the present invention provides a multispectral image sensor, comprising: an image sensor, comprising a substrate and a pixel area located on the surface of the substrate, wherein a plurality of array-arranged pixel units are formed in the pixel area; a filter, wherein a spectral channel area corresponding to the pixel area is formed on the surface of the filter, comprising a plurality of spectral channels corresponding to respective wavelengths arranged in a mosaic form, wherein the side length of each spectral channel is greater than the side length of the pixel unit; the filter is fixed to the surface of the image sensor, and a rotational offset angle between the pixel area and the spectral channel area is less than a threshold value, so that the projection of each spectral channel on the pixel area covers at least one complete pixel unit.

[0006] Optionally, the side length of the spectral channel is N times the side length of the pixel unit, where N is an integer greater than or equal to 2.

[0007] Optionally, the threshold is arctan (pixel unit side length / pixel area short side length).

[0008] Optionally, the filter includes: a transparent substrate and a filter layer formed on the surface of the transparent substrate, and the filter layers of spectral channels of different wavelengths can selectively transmit light waves of corresponding wavelengths.

[0009] Optionally, alignment patterns are formed on the filter and the image sensor respectively.

[0010] Optionally, there is a cavity or a transparent adhesive layer between the light channel area and the pixel area.

[0011] Optionally, it also includes: a calibration module for storing a calibration file, wherein the content of the calibration file includes the correspondence between each pixel unit and the spectral channel of each wavelength, and the storage form of the correspondence includes at least one of direct expression, compressed expression and abstract expression.

[0012] Optionally, the corresponding relationship includes light components obtained by each pixel unit from spectral channels of different wavelengths.

[0013] Optionally, it also includes: an imaging module, which forms images corresponding to spectral channels of various wavelengths according to the calibration file and the detection signal obtained from each pixel unit.

[0014] Optionally, the method of forming images corresponding to spectral channels of each wavelength based on the calibration file and the detection signals obtained by each pixel unit further includes: selecting a pixel unit that only obtains the light signal of a single spectral channel as an imaging unit; and obtaining images corresponding to each wavelength based on the wavelength of the spectral channel corresponding to the imaging unit and the output detection signal.

[0015] Optionally, the method of forming images corresponding to spectral channels of each wavelength based on the calibration file and the detection signals obtained by each pixel unit includes: obtaining light components obtained by each pixel unit from the spectral channels of each wavelength based on the correspondence between each pixel unit and the spectral channels of each wavelength; obtaining detection signal components of each wavelength in the detection signal output by each pixel unit based on the light components, and obtaining images corresponding to each wavelength based on the detection signal components of each wavelength.

[0016] The technical solution of the present invention also provides an imaging method for a multispectral image sensor, comprising: obtaining a detection signal output by each pixel unit; obtaining a calibration file, wherein the content of the calibration file includes a correspondence between each pixel unit and a spectral channel of each wavelength; and forming images corresponding to the spectral channels of each wavelength respectively according to the calibration file and the detection signal.

[0017] Optionally, the corresponding relationship includes: the corresponding relationship includes light components obtained by each pixel unit from spectral channels of different wavelengths.

[0018] Optionally, the method for forming images corresponding to each wavelength spectral channel according to the calibration file and the detection signal further includes: selecting a pixel unit that only acquires the light signal of a single spectral channel as an imaging unit; and acquiring images corresponding to each wavelength according to the wavelength of the spectral channel corresponding to the imaging unit and the output detection signal.

[0019] Optionally, the method of forming images corresponding to spectral channels of each wavelength based on the calibration file and the detection signals obtained by each pixel unit further includes: obtaining light components obtained by each pixel unit from the spectral channels of each wavelength based on the correspondence between each pixel unit and the spectral channels of each wavelength; obtaining detection signal components of each wavelength in the detection signal output by each pixel unit based on the light components, and obtaining images corresponding to each wavelength based on the detection signal components of each wavelength.

[0020] Optionally, the method for acquiring the calibration file includes: calculating the correspondence between each pixel unit and the spectral channel of each wavelength through the displacement offset between the spectral channel area of ​​the filter and the pixel area of ​​the image sensor.

[0021] Optionally, the method for obtaining the calibration file includes: actually capturing images of one or more wavelength light sources, and obtaining the correspondence between each pixel unit and the spectral channel of each wavelength according to the intensity of the detection signal output by each pixel unit under each wavelength light source.

[0022] Optionally, the multispectral image sensor is as described in any one of the above items.

[0023] The multi-spectral image sensor of the present invention includes a filter having multiple spectral channels that is fixed to the surface of the image sensor and is separate from the image sensor. The filter and the image sensor are manufactured separately, and both can be manufactured using mature and low-cost manufacturing methods. In addition, the size of the spectral channel is larger than the size of the pixel unit, so that the projection of each spectral channel on the pixel area covers at least one complete pixel unit, thereby reducing the alignment requirements between the image sensor and the filter and reducing the process difficulty. In addition, the multi-spectral image sensor also includes a calibration module for calibrating the offset between the image sensor and the filter, so that the entire system can tolerate a larger relative displacement between the image sensor and the filter, and high-quality multi-spectral images can be obtained under different offsets.

[0024] The imaging method of the present invention can calibrate the offset between the image sensor and the filter when there is a position offset between the image sensor and the filter, so that the entire system can tolerate a larger relative displacement between the image sensor and the filter, and high-quality multispectral images can be obtained under different offsets. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A schematic diagram of the structure of a multispectral image sensor according to a specific embodiment of the present invention;

[0026] Figure 2 It is a partial top view schematic diagram of a spectral channel region and a pixel region according to a specific embodiment of the present invention;

[0027] Figure 3 A schematic diagram of a partial structure of a multispectral image sensor according to another specific embodiment of the present invention;

[0028] Figure 4 The figure is a flow chart of an imaging method of a multispectral image sensor according to a specific embodiment of the present invention. DETAILED DESCRIPTION

[0029] The specific implementation manner of the multispectral image sensor and the imaging method thereof provided by the present invention will be described in detail below with reference to the accompanying drawings.

[0030] Please refer to Figure 1 , is a schematic structural diagram of a multispectral image sensor according to a specific embodiment of the present invention.

[0031] The multispectral image sensor includes: an image sensor 110 and a filter 120 . The filter 120 is fixed on the surface of the image sensor 110 .

[0032] The image sensor 110 includes a substrate 111 and a pixel area 112 located on the surface of the substrate 111. A plurality of array-arranged pixel units 1, 2, ... are formed in the pixel area 112. Specifically, the substrate 111 includes a semiconductor substrate, and the pixel unit is formed in the substrate 111, which may be a CMOS sensor unit or a CCD sensor unit. Structures such as electrical interconnects may be formed in the substrate 111 for outputting the detection signal of the pixel unit to an external circuit for processing. Other circuit modules may also be integrated in the substrate 111 for processing the detection output by the pixel unit.

[0033] In this specific embodiment, a single pixel unit is a square. In other specific embodiments, a single pixel unit can also be other shapes, such as a rectangle. The shape and size of each pixel unit are the same. The pixel area 112 is used to receive external light, and each pixel unit is used to output a detection signal corresponding to the received light intensity.

[0034] The surface of the filter 120 is formed with a spectral channel area 123 corresponding to the pixel area 112, including a plurality of spectral channels ch1, ch2, ch3, ..., corresponding to each wavelength, arranged in a mosaic form. Each spectral channel is used to pass light of different wavelengths, so that the corresponding pixel unit in the pixel area 112 can only receive light waves of a specific wavelength that can pass through the corresponding spectral channel. In this specific embodiment, the filter 120 includes a transparent substrate 121 and a filter layer 122 formed on the first surface of the transparent substrate 121. Spectral channels of different wavelengths use different filter layer materials and structures, so that spectral channels of different wavelengths can selectively pass light waves of different wavelengths. In this specific embodiment, the spectral channels ch1, ch2, ch3, ..., are used to pass different wavelengths. In a specific embodiment of the present invention, the spectral channel area 123 of the filter 120 may include 4, 8, 16 or other numbers of spectral channels, so that the multi-spectral image sensor can output images of 4, 8, and 16 corresponding wavelengths, respectively. Those skilled in the art can set corresponding spectral channels to correspond to 2 or more wavelengths as needed.

[0035] The optical filter 120 is fixed to the surface of the image sensor 110, and the spectral channel region 123 is arranged toward the pixel region 112, and the filter layer 122 is toward the pixel region 112. The optical filter 120 and the image sensor 110 can be bonded together by a bonding process. In this specific embodiment, the optical filter 120 and the surface of the image sensor 110 are bonded together by a transparent adhesive layer 130, and the adhesive layer 130 is also formed between the optical filter 120 and the image sensor 110. The adhesive layer 130 has high light transmittance, so that the light passing through the spectral channel region 123 can be received by the pixel region 112. In other specific embodiments, the edge of the image sensor 110 located at the periphery of the pixel region 112 is bonded to the edge area of ​​the spectral channel region 123 of the optical filter 120, so that there is a cavity between the optical channel region 123 and the pixel region 112.

[0036] In other specific implementations, the filter layer 122 may also be formed on the second surface of the transparent substrate 121 , and the incident light passes through the filter layer 122 and the transparent substrate 121 before reaching the pixel region 112 .

[0037] The side length of each spectral channel is the same and greater than the side length of the pixel unit, so that when there is a certain displacement deviation between the spectral channel and the pixel unit, the entire pixel unit can still be completely covered. Specifically, the rotation offset angle between the pixel area 112 and the spectral channel area 123 is less than a threshold value, so that the projection of each spectral channel on the pixel area 112 covers at least one complete pixel unit.

[0038] Please refer to Figure 2 , is a partial top view schematic diagram of the spectral channel region 123 and the pixel region 112 according to a specific embodiment of the present invention.

[0039] In this specific implementation, the spectral channel region 123 includes eight spectral channels corresponding to eight different wavelengths, namely ch1 to ch8. Figure 2 Only a partial area of ​​the partial pixel area 112 and a partial area of ​​the spectral channel area 123 are shown.

[0040] In this specific implementation, the side length of the spectral channel is twice the side length of the pixel unit, that is, the area of ​​one spectral channel is equivalent to the area of ​​four pixel units. Taking the spectral channel ch1 as an example, when the spectral channel region 123 is completely aligned with the pixel region 112 and there is no relative offset between the two, the spectral channel region 123 covers four complete pixel units, for example, four pixel units with coordinates (0,0), (0.1), (1,0) and (1,1). Since the lengthening of the spectral channel is twice the lengthening of the pixel unit, when there is only a translation position deviation between the spectral channel region 123 and the pixel region 113, no matter how large the translation displacement deviation is, there will always be a complete pixel unit located below a certain pixel channel. Therefore, it is only necessary to control the rotation offset angle between the spectral channel region 123 and the pixel region 113 so that the rotation offset angle is less than a certain threshold. The threshold is arctan (pixel unit side length / pixel area short side length).

[0041] In other specific embodiments, the side length of the spectral channel is N times the side length of the pixel unit, where N is an integer greater than or equal to 2. For example, the length of a single spectral channel is 3, 4, or 5 times the side length of a single pixel unit. In this case, when the rotation offset angle is less than the threshold, there are always (N-1) 2 A complete pixel unit is located under a spectral channel. The detection signal output by the complete pixel unit located under a spectral channel can be directly used to output an image of the corresponding wavelength.

[0042] by Figure 2For example, the specific implementation of , pixel unit (0,0) corresponds to spectral channel ch1, pixel unit (0,2) corresponds to spectral channel ch2, pixel unit (0,4) corresponds to spectral channel ch3, pixel unit (0,6) corresponds to spectral channel ch4, pixel unit (2,0) corresponds to spectral channel ch5, pixel unit (2,2) corresponds to spectral channel ch6, pixel unit (0,4) corresponds to spectral channel ch7, pixel unit (0,6) corresponds to spectral channel ch8, and so on. According to the detection signal output by the complete pixel unit corresponding to each spectral channel, the image compressed by the light wave of the corresponding wavelength is output.

[0043] Please refer to Figure 3 , is a schematic diagram of the local structure of a multispectral image sensor according to another specific embodiment of the present invention.

[0044] The multi-spectral image sensor may further include a calibration module 301 for storing a calibration file, wherein the calibration file content includes a correspondence between each pixel unit and a spectral channel of each wavelength. The multi-spectral image sensor further includes an imaging module 302, connected to each pixel unit in the pixel area 112, and connected to the calibration module 301, for acquiring a detection signal output by each pixel unit, and forming an image corresponding to each spectral channel of each wavelength according to the calibration file and the detection signal acquired from each pixel unit.

[0045] The correspondence between each pixel unit and each wavelength spectral channel includes the light components obtained by each pixel unit from the spectral channels of different wavelengths. For a multi-spectral sensor, the light components obtained by each pixel unit from different optical channels are in a fixed ratio, which can be obtained through calibration.

[0046] In a specific embodiment, the method in which the calibration module forms images corresponding to spectral channels of each wavelength based on the calibration file and the detection signals obtained by each pixel unit further includes: selecting a pixel unit that only obtains the light signal of a single spectral channel as an imaging unit; and obtaining images corresponding to each wavelength based on the wavelength of the spectral channel corresponding to the imaging unit and the output detection signal.

[0047] In another specific embodiment, the method in which the calibration module forms images corresponding to spectral channels of each wavelength based on the calibration file and the detection signals obtained by each pixel unit includes: obtaining light components obtained by each pixel unit from the spectral channels of each wavelength based on the correspondence between each pixel unit and the spectral channels of each wavelength; obtaining detection signal components of each wavelength in the detection signal output by each pixel unit based on the light components, and obtaining images corresponding to each wavelength based on the detection signal components of each wavelength.

[0048] In a specific embodiment of the present invention, alignment patterns are respectively formed on the filter 120 and the image sensor 110 for calculating the relative displacement between the spectral channel region 123 and the pixel region 112, so that the area of ​​each spectral channel projected on each pixel unit below can be known based on the specific size relationship, thereby calculating the correspondence between each pixel unit in the pixel region 112 of the image sensor and each spectral channel in the spectral channel region 123, and storing the correspondence as a calibration file.

[0049] In another specific embodiment of the present invention, by actually shooting one or more images of light sources with wavelengths corresponding to each spectral channel, and by outputting detection signals of each pixel unit under each wavelength light source, the light components of each spectral channel acquired by each pixel unit can be obtained, and the correspondence between each pixel unit in the pixel area 112 and each spectral channel in the spectral channel area 123 can be obtained, and the correspondence can be stored as a calibration file. The calibration file can store the correspondence in multiple forms, such as at least one of direct expression, compressed expression, or abstract expression of the correspondence.

[0050] The above-mentioned multi-spectral image sensor includes a filter having multiple spectral channels fixed to the surface of the image sensor and separated from the image sensor. The filter and the image sensor are manufactured separately, and both can be manufactured using mature and low-cost manufacturing methods. In addition, the size of the spectral channel is larger than the size of the pixel unit, so that the projection of each spectral channel on the pixel area covers at least one complete pixel unit, thereby reducing the alignment requirements between the image sensor and the filter and reducing the process difficulty. In addition, the multi-spectral image sensor also includes a calibration module for calibrating the offset between the image sensor and the filter, so that the entire system can tolerate a larger relative displacement between the image sensor and the filter, and high-quality multi-spectral images can be obtained under different offsets.

[0051] A specific embodiment of the present invention also provides an imaging method of a multi-spectral image sensor.

[0052] Please refer to Figure 4 , is a schematic flow chart of an imaging method of a multispectral image sensor according to a specific embodiment of the present invention.

[0053] The multispectral image sensor is as described in the above specific embodiment, and the imaging method comprises the following steps:

[0054] S101: Acquire a detection signal output by each pixel unit.

[0055] The pixel area of ​​the image sensor includes a plurality of pixel units for receiving external light, and each pixel unit is used to output a detection signal corresponding to the received light intensity. Since some pixel units may correspond to spectral channels corresponding to multiple wavelengths, the detection signal output by the pixel unit may include a detection signal generated by light waves of one or more wavelengths.

[0056] S102: Acquire a calibration file, wherein the calibration file includes a correspondence between each pixel unit and a spectral channel of each wavelength.

[0057] The correspondence includes the light components obtained by each pixel unit from spectral channels of different wavelengths, wherein some pixel units can only receive light components obtained from a single spectral channel; some pixel units can receive light components obtained from two or more spectral channels. The calibration file records the proportion of light components obtained by each spectral channel from each spectral channel.

[0058] In some specific embodiments, the method for obtaining the calibration file includes: calculating the correspondence between each pixel unit and the spectral channel of each wavelength through the displacement offset between the spectral channel area of ​​the filter and the pixel area of ​​the image sensor. According to the specific size relationship, the area of ​​each spectral channel projected on each pixel unit below can be known, so as to calculate the correspondence between each pixel unit in the pixel area 112 of the image sensor and each spectral channel in the spectral channel area 123, and store the correspondence as a calibration file.

[0059] In other specific embodiments, the method for obtaining the calibration file includes: actually capturing images of one or more wavelength light sources, obtaining the correspondence between each pixel unit and the spectral channel of each wavelength according to the intensity of the detection signal output by each pixel unit under each wavelength light source, and storing the correspondence as a calibration file.

[0060] The calibration file may store the correspondence between each pixel unit and each wavelength spectral channel in various forms, such as at least one of direct expression, compressed expression or abstract expression of the correspondence. The correspondence may be obtained by using a corresponding decoding or reading method according to the form in which the data is stored in the calibration file.

[0061] S103: forming images corresponding to spectral channels of respective wavelengths according to the calibration file and the detection signal.

[0062] In some specific embodiments, a pixel unit that only acquires a light signal of a single spectral channel can be selected as an imaging unit; and images corresponding to each wavelength are acquired according to the wavelength of the spectral channel corresponding to the imaging unit and the output detection signal.

[0063] In other specific embodiments, the light component obtained by each pixel unit from the spectral channel of each wavelength can be obtained based on the correspondence between each pixel unit and the spectral channel of each wavelength; based on the light component, the detection signal component of each wavelength in the detection signal output by each pixel unit is obtained; and based on the detection signal component of each wavelength, the image corresponding to each wavelength is obtained.

[0064] For example, the pixel unit (0,0) only receives the light component obtained from the spectral channel ch1, while the pixel unit (0,1) simultaneously obtains the light components obtained from the spectral channels ch1 and ch2, and the ratio of the two is 1:1. Then, when imaging the wavelength corresponding to the spectral channel ch1, only the detection signal obtained by the pixel unit (0,0) can be used for imaging to form an image of the wavelength corresponding to the spectral channel ch1; or the detection signal of the pixel unit (0,0) and 50% of the detection signal of the pixel unit (0,1) can be used at the same time to form an image of the wavelength corresponding to the spectral channel ch1.

[0065] The above imaging method can calibrate the offset between the image sensor and the filter, so that the entire system can tolerate a larger relative displacement between the image sensor and the filter, and high-quality multispectral images can be obtained under different offsets.

[0066] The above is only a preferred embodiment of the present invention. It should be pointed out that ordinary technicians in this technical field can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A multispectral image sensor, characterized in that: include: An image sensor comprises a substrate and a pixel region located on a surface of the substrate, wherein a plurality of pixel units arranged in an array are formed in the pixel region; A filter, wherein a spectral channel region corresponding to the pixel region is formed on the surface of the filter, including a plurality of spectral channels corresponding to respective wavelengths arranged in a mosaic form, and the side length of each spectral channel is greater than the side length of the pixel unit; The filter is fixed on the surface of the image sensor, there is a rotational offset between the pixel area and the spectral channel area, and the rotational offset angle is less than a threshold value, so that the projection of each spectral channel on the pixel area covers at least one complete pixel unit; A calibration module, used to store a calibration file, wherein the calibration file content includes a correspondence between each pixel unit and a spectral channel of each wavelength, wherein the calibration module is used to calibrate the offset between the image sensor and the filter, wherein the correspondence includes light components obtained by each pixel unit from spectral channels of different wavelengths; An imaging module, forming images corresponding to spectral channels of each wavelength respectively according to the calibration file and the detection signal obtained from each pixel unit; The method of forming images corresponding to spectral channels of respective wavelengths according to the calibration file and the detection signals acquired by each pixel unit further comprises: selecting a pixel unit that acquires only the light signal of a single spectral channel as an imaging unit; According to the wavelength of the spectral channel corresponding to the imaging unit and the output detection signal, the image corresponding to each wavelength is acquired.

2. The multispectral image sensor according to claim 1, characterized in that: The side length of the spectral channel is N times the side length of the pixel unit, where N is an integer greater than or equal to 2.

3. The multispectral image sensor according to claim 1, characterized in that: The threshold is arctan (pixel unit side length / pixel area short side length).

4. The multispectral image sensor according to claim 1, characterized in that: The optical filter comprises: a transparent substrate and a filter layer formed on the surface of the transparent substrate, and the filter layers of spectral channels of different wavelengths can selectively transmit light waves of corresponding wavelengths respectively.

5. The multispectral image sensor according to claim 1, characterized in that: Alignment patterns are formed on the filter and the image sensor respectively.

6. The multispectral image sensor according to claim 1, characterized in that: There is a cavity or a transparent adhesive layer between the spectral channel region and the pixel region.

7. The multispectral image sensor according to claim 1, characterized in that: The storage form of the corresponding relationship includes at least one of direct expression, compressed expression and abstract expression.

8. The multispectral image sensor according to claim 1, characterized in that: The method for forming images corresponding to spectral channels of each wavelength based on the calibration file and the detection signals obtained by each pixel unit includes: obtaining the light components obtained by each pixel unit from the spectral channels of each wavelength based on the correspondence between each pixel unit and the spectral channels of each wavelength; obtaining the detection signal components of each wavelength in the detection signals output by each pixel unit based on the light components, and obtaining the images corresponding to each wavelength based on the detection signal components of each wavelength.

9. An imaging method using the multispectral image sensor according to any one of claims 1 to 8, characterized in that: include: Acquire a detection signal output by each pixel unit; Acquire a calibration file, wherein the calibration file includes a correspondence between each pixel unit and a spectral channel of each wavelength, wherein the correspondence includes light components obtained by each pixel unit from spectral channels of different wavelengths; The offset between the image sensor and the filter is calibrated according to the calibration file and the detection signal to form images corresponding to the spectral channels of each wavelength.

10. The imaging method according to claim 9, characterized in that: The method of forming images corresponding to each wavelength spectral channel according to the calibration file and the detection signal further includes: selecting a pixel unit that only obtains the light signal of a single spectral channel as an imaging unit; and obtaining images corresponding to each wavelength according to the wavelength of the spectral channel corresponding to the imaging unit and the output detection signal.

11. The imaging method according to claim 9, characterized in that: The method of forming images corresponding to spectral channels of each wavelength based on the calibration file and the detection signals obtained by each pixel unit further includes: obtaining light components obtained by each pixel unit from the spectral channels of each wavelength based on the correspondence between each pixel unit and the spectral channels of each wavelength; obtaining detection signal components of each wavelength in the detection signal output by each pixel unit based on the light components, and obtaining images corresponding to each wavelength based on the detection signal components of each wavelength.

12. The imaging method according to claim 9, characterized in that: The method for obtaining the calibration file includes: calculating the correspondence between each pixel unit and the spectral channel of each wavelength through the displacement offset between the spectral channel area of ​​the filter and the pixel area of ​​the image sensor.

13. The imaging method according to claim 9, characterized in that: The method for obtaining the calibration file includes: actually capturing images of one or more wavelength light sources, and obtaining the correspondence between each pixel unit and the spectral channel of each wavelength according to the intensity of the detection signal output by each pixel unit under each wavelength light source.

Images