Spectrometer based on composite multi-slit and imaging method thereof

Abstract

The present application relates to spectral imaging method, specifically relates to the spectrometer based on composite multi-slit and its imaging method, in order to solve the insufficient of the local weak light area in the existing multi-slit technology cannot obtain sufficient gain effect, the signal-to-noise ratio is difficult to improve.The spectrometer based on composite multi-slit includes detector, spectrometer system and slit assembly arranged in the incident end of spectrometer system, the slit assembly includes N complete dispersion slits and M local dispersion slits arranged in the same plane, and the axis is located in the same plane;N complete dispersion slits are arranged adjacent, and M local dispersion slits are arranged adjacent;The spacing of adjacent complete dispersion slits is n+1 pixel size;The spacing of adjacent local dispersion slits is m+1 pixel size;Meanwhile, the imaging method based on the above spectrometer is provided, N+M spectral data is linearly added, the spectrometer imaging data of the target to be measured is obtained, so as to enhance the signal-to-noise ratio of weak light spectral interval.

Description

TECHNICAL FIELD

[0001] The present application relates to a spectral imaging method, in particular to a spectral imaging method based on a composite multi-slit spectral instrument. BACKGROUND

[0002] The spatial resolution and spectral resolution of the current spectral instrument are constantly improving, and the improvement of resolution and spectral subdivision leads to a decrease of one to two orders of magnitude in single spectral incident energy, making it difficult to improve the signal-to-noise ratio. At the same time, due to the difference in incident radiation energy, there are local weak light areas in some spectral bands. For example, there is a local weak light area in the 2-2.5 μm band of the short-wave infrared spectral instrument. When the solar elevation angle is 60 degrees and the ground reflectivity is 0.2, the entrance pupil radiance curve of the short-wave infrared spectral instrument in the 0.9-2.5 μm band is shown in FIG. 1. As can be seen from FIG. 1, the average entrance pupil radiance in the 0.9-2.5 μm band is 4.98 W / (m2·μm·sr), while the average entrance pupil radiance in the 2-2.5 μm band is only 0.48 W / (m2·μm·sr), which is only 1 / 10 of the average entrance pupil radiance in the 0.9-2.5 μm band. The light intensity in the corresponding area is insufficient, the signal-to-noise ratio is low and difficult to improve, and it is difficult to meet the application requirements of high signal-to-noise ratio. Figure 1 2 ·μm·sr), while the average entrance pupil radiance in the 2-2.5 μm band is only 0.48 W / (m 2 ·μm·sr), which is only 1 / 10 of the average entrance pupil radiance in the 0.9-2.5 μm band. The light intensity in the corresponding area is insufficient, the signal-to-noise ratio is low and difficult to improve, and it is difficult to meet the application requirements of high signal-to-noise ratio.

[0003] In order to achieve high signal-to-noise ratio, the current commonly used methods mainly include increasing the aperture of the optical system, improving the transmittance of the optical system, improving the quantum efficiency of the detector, increasing the integration time, and multiple sampling technology, etc. The increase of the aperture of the optical system and the increase of the volume and weight of the system are nearly cubic relationship. When the aperture increases to a certain extent, the mirror processing, support, assembly and other engineering problems increase exponentially. Increasing the transmittance of the optical system usually reduces the number of optical elements in the optical system and improves the spectral efficiency. This way, the transmittance is difficult to increase significantly, and the signal-to-noise ratio improvement effect is limited. Improving the quantum efficiency of the detector requires breakthroughs in the core technology of high-performance detectors, which is difficult to improve in a short time. Increasing the integration time can be achieved by motion compensation, but it requires high platform stability and cannot obtain spatially continuous images. The multiple sampling technology is a technical solution that achieves multiple exposures of the same target scene through a special way, thereby obtaining greater signal strength and suppressing random noise, and thus obtaining higher signal-to-noise ratio. The cost is relatively small and the signal-to-noise ratio gain is large.

[0004] ​Currently, several mature multi-sampling technical solutions mainly include TDI technology, interferometric spectral imaging technology and multi-slit technology. The TDI technology is only applicable to panchromatic and multispectral and cannot be applied to hyperspectral; the interferometric spectral imaging technology needs to be reconstructed through complex mathematical calculation such as Fourier transform, and the process is relatively complex. The multi-slit technology can be relatively simply realized in engineering, but is limited by the size of the detector array of the spectrometer, and can only realize double or triple slits, and cannot obtain a high signal-to-noise ratio gain, so that sufficient gain effect cannot be obtained for local weak light areas with low energy. SUMMARY

[0005] The purpose of the present application is to solve the problem that the local weak light area cannot obtain sufficient gain effect and the signal-to-noise ratio is difficult to improve in the existing multi-slit technology, and to provide a spectrometer based on a composite multi-slit and an imaging method thereof.

[0006] To achieve the above-mentioned purpose, the technical solution provided by the present application is as follows:

[0007] A spectrometer based on a composite multi-slit, characterized in that it comprises a detector, a light splitting system and a slit assembly arranged at the incident end of the light splitting system, the slit assembly comprising N complete dispersion slits and M local dispersion slits arranged side by side on the same plane; wherein N and M satisfy the following relationship:

[0008] N x (n + 1) + M x (m + 1) - 1 ≤ A,

[0009] wherein A is the number of pixels in the spectral direction of the detector, n is the total number of spectral segments corresponding to the full spectral range of the spectrometer, and m is the number of spectral segments corresponding to the weak light spectral range of the spectrometer;

[0010] The N complete dispersion slits are arranged adjacent to each other, and the M local dispersion slits are arranged adjacent to each other; the spacing between adjacent complete dispersion slits is n+1 pixel size; the spacing between adjacent local dispersion slits is m+1 pixel size.

[0011] Further, it is defined that the full spectral range of the spectrometer is λ1~λ n , and the weak light spectral range is λ s+1 ~ λ s+m ; wherein s+1 is the starting spectral segment number of the weak light spectral range, and s+m is the terminal spectral segment number of the weak light spectral range.

[0012] A local dispersion filter is arranged in the light output direction of the M local dispersion slits, and the bandpass spectral range of the local dispersion filter is λ s+1 ~ λ s+m .

[0013] Further, the weak light spectral range λs+1 ~ λ s+m Close to λ1, N complete dispersion slits and M partial dispersion slits are sequentially arranged along the dispersion direction of the complete dispersion slit, and the interval between adjacent partial dispersion slits and complete dispersion slits is n-s+1 pixel size.

[0014] Further, the weak light spectral interval λ s+1 ~ λ s+m Close to λ n , M partial dispersion slits and N complete dispersion slits are sequentially arranged along the dispersion direction of the complete dispersion slit, and the interval between adjacent partial dispersion slits and complete dispersion slits is s+m+1 pixel size.

[0015] Meanwhile, an imaging method based on the composite multi-slit spectrometer is also provided, and the speciality thereof lies in comprising the following steps:

[0016] S1, aligning the N complete dispersion slits and M partial dispersion slits of the above-mentioned composite multi-slit spectrometer with the target to be measured respectively, performing spectral imaging on the target to be measured, and obtaining N+M spectral data;

[0017] S2, linearly adding the obtained N+M spectral data to obtain the spectral data of the target to be measured after the signal-to-noise ratio is enhanced, and completing the spectral imaging based on the composite multi-slit.

[0018] Further, in step S1, a partial dispersion filter is arranged on the light-emitting direction of the M partial dispersion slits, and the band-pass spectral interval of the partial dispersion filter is λ s+1 ~ λ s+m .

[0019] The beneficial effects of the present application are:

[0020] 1. The composite multi-slit spectrometer based on the composite multi-slit of the present application comprises N complete dispersion slits and M partial dispersion slits, the interval between the N complete dispersion slits is n+1 pixel size, and the interval between the M partial dispersion slits is m+1 pixel size, so as to enhance the signal-to-noise ratio of the weak light spectral interval.

[0021] 2. In the composite multi-slit spectrometer based on the composite multi-slit of the present application, the interval between the N complete dispersion slits is n+1 pixel size, and the interval between the M partial dispersion slits is m+1 pixel size, so that the spectral data of different slits are separated by 1 pixel, the spectral crosstalk between different slits can be reduced, and the spectral data quality can be improved.

[0022] 3. In the composite multi-slit spectrometer based on the composite multi-slit of the present application, a partial dispersion filter is arranged behind the M partial dispersion slits, and the band-pass spectral interval of the partial dispersion filter is λ s+1 ~ λ s+m, adapted to the weak light spectrum range, to achieve maximum signal-to-noise ratio gain and reduce crosstalk between spectral bands.

[0023] 4. In the present invention, a local dispersion filter is set after the local dispersion slit, so the distance between the local dispersion slit and the complete dispersion slit should be as large as possible to reduce the engineering difficulty. The distance between the local dispersion slit and the complete dispersion slit is based on the weak light spectrum range λ s+1 ~λ s+m The position of the weak light spectrum range λ s+1 ~λ s+m When the position satisfies 2s+m≤n, its position is closer to λ1, and the complete dispersion slit and the local dispersion slit are arranged along the dispersion direction, which can ensure that the distance between the local dispersion slit and the complete dispersion slit is the largest, and the distance is n-s+1 pixel size. s+1 ~λ s+m Satisfying 2s+m≥n, its position is closer to λ n At this time, the local dispersion slits and the complete dispersion slits are arranged along the dispersion direction, which can ensure that the distance between the local dispersion slits and the complete dispersion slits is maximized, and the distance between adjacent local dispersion slits and complete dispersion slits is s+m+1 pixel sizes.

[0024] 5. The composite multi-slit spectrometer of the present invention is simple to implement. It only requires replacing the single slit in the original dispersion spectrometer with a composite multi-slit. In the imaging method, it only requires linear addition of the composite multi-slit data to obtain imaging data with a high signal-to-noise ratio, thus solving the problem of difficulty in improving the signal-to-noise ratio in the weak light spectral range of the spectrometer.

[0025] 6. The imaging method of the composite multi-slit spectrometer of the present invention reasonably sets the number of M local dispersion slits and N complete dispersion slits under the premise of meeting the detector array size, so that it satisfies N×(n+1)+M×(m+1)-1≤A. By linearly adding the spectral data of the composite multi-slit dispersion, it is possible to achieve the low-light spectral range λ s+1 ~λ s+m Improved signal-to-noise ratio times, improving the signal-to-noise ratio of the entire spectral range times, achieving an improvement in the signal-to-noise ratio in the full spectrum range and low-light spectrum range. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a schematic diagram of the entrance pupil radiance curve of a short-wave infrared spectrometer in the prior art in the wavelength range of 0.9 to 2.5 μm;

[0027] Figure 2 Schematic diagram of the principle of a composite multi-slit and a local dispersion filter in an embodiment of a composite multi-slit spectrometer according to the present invention;

[0028] Figure 3 The dispersion diagram of the short-wave infrared spectrometer in the embodiment of the composite multi-slit spectrometer of the present invention;

[0029] Figure 4 A comparison chart of the signal-to-noise ratio of imaging using the composite multi-slit spectrometer of the present invention and the signal-to-noise ratio of imaging using a single-slit spectrometer, where the unit of the signal-to-noise ratio is S / N;

[0030] Figure 5 The figure is a comparison chart of the signal-to-noise ratio of imaging using the composite multi-slit spectrometer of the present invention and the signal-to-noise ratio of imaging using a single-slit spectrometer, where the unit of the signal-to-noise ratio is dB. DETAILED DESCRIPTION

[0031] The present invention provides a spectrometer based on a composite multi-slit, comprising a detector, a spectroscopic system, and a slit assembly arranged at the incident end of the spectroscopic system. The spectroscopic system disperses light and images it onto the detector. In order to achieve full-spectrum signal-to-noise ratio enhancement and signal-to-noise ratio enhancement in the weak light range, under the condition that the detector array size is limited, the present invention configures the slit assembly of the spectrometer as a composite multi-slit, which includes N complete dispersion slits and M local dispersion slits arranged in parallel on the same plane. The N complete dispersion slits are arranged adjacent to each other, and the spacing between adjacent complete dispersion slits is n+1 pixel sizes. The M local dispersion slits are arranged adjacent to each other, and the spacing between adjacent local dispersion slits is m+1 pixel sizes. N and M satisfy the following relationship:

[0032] N×(n+1)+M×(m+1)-1≤A,

[0033] Among them, A is the number of pixels in the spectral direction of the detector, n is the total number of spectral segments, and m is the number of spectral segments corresponding to the weak light spectral range.

[0034] Definition: The full spectrum range of the spectrometer is λ1~λ n , the weak light spectrum range is λ s+1 ~λ s+m , where s+1 is the starting spectral segment number of the weak light spectral interval, and s+m is the ending spectral segment number of the weak light spectral interval. To maximize the signal-to-noise ratio gain and reduce crosstalk between spectral segments, a local dispersion filter is placed in the light-emitting direction of the M local dispersion slits. The filter has a bandpass spectral interval of λ s+1 ~λ s+m , and in the spectral range λ1~λ s and λ s+m+1 ~λ n Internal deadline.

[0035] The distance between the local dispersion slit and the complete dispersion slit depends on the weak light spectrum range λ s+1 ~λ s+m The position is determined if the weak light spectrum interval λ s+1~λ s+m Close to λ1, that is, satisfying 2s+m≤n, then N complete dispersion slits and M local dispersion slits are set in sequence along the dispersion direction of the complete dispersion slit, and the spacing between adjacent local dispersion slits and complete dispersion slits is n-s+1 pixel size; if the weak light spectrum interval λ s+1 ~λ s+m Close to λ n , that is, 2s+m>n is satisfied, then M local dispersion slits and N complete dispersion slits are arranged in sequence along the dispersion direction of the complete dispersion slit, and the spacing between adjacent local dispersion slits and complete dispersion slits is s+m+1 pixel size.

[0036] The imaging method based on the composite multi-slit spectrometer comprises the following steps:

[0037] S1. Perform spectral imaging using the composite multi-slit spectrometer, aligning N complete dispersion slits and M partial dispersion slits with the target to be measured, and perform spectral imaging on the target to be measured to obtain N+M spectral data of the target to be measured;

[0038] S2. Linearly add N+M spectral data of the target to be measured to obtain spectral data of the target after the signal-to-noise ratio is enhanced.

[0039] The random noise is suppressed by linearly adding the spectral data of the composite multi-slit dispersion to achieve signal-to-noise ratio enhancement. The signal-to-noise ratio of the spectral data in different spectral intervals after using the above spectrometer is calculated respectively, and the weak light spectral interval λ s+1 ~λ s+m The signal-to-noise ratio SNR(λ p ) is calculated by the following formula:

[0040]

[0041] Where N is the number of complete dispersion slits, M is the number of local dispersion slits, is the wavelength λ corresponding to the spectrometer p The number of signal electrons, σ n It is the comprehensive noise electron count including readout noise, quantization noise, dark current noise, circuit noise and other noises.

[0042] Spectral range λ1~λ s and λ s+m+1 ~λ n The signal-to-noise ratio SNR(λ q ) is calculated by the following formula:

[0043]

[0044] in, is the wavelength λ corresponding to the spectrometer qThe number of signal electrons.

[0045] Full spectrum range λ1~λ n The average signal-to-noise ratio Calculated by the following formula:

[0046]

[0047] Compared with the signal-to-noise ratio of spectral data obtained by using an ordinary spectrometer, under the premise of meeting the detector array size, the number of M local dispersion slits and N complete dispersion slits can be reasonably set to achieve the low-light spectral range λ s+1 ~λ s+m Improved signal-to-noise ratio times, non-weak light spectrum range λ1~λ s and λ s+m+1 ~λ n The signal-to-noise ratio of nm spectral bands is improved times, improving the signal-to-noise ratio of the entire spectral range times, achieving improved signal-to-noise ratio in the full spectrum range and low-light areas.

[0048] In this embodiment, a short-wave infrared spectrometer is used as an example. The short-wave infrared spectrometer can disperse 98 spectral bands from 0.9 to 2.5 μm. The detector array used is 1024×256, and the 256 pixel direction is the spectral direction. According to the above method, a composite multi-slit is set at the slit of the spectrometer, including two complete dispersion slits M1 and M2 and two partial dispersion slits N1 and N2, as shown in FIG. Figure 2 As shown in the figure, the full spectrum range of the spectrometer is 0.9-2.5μm, and the weak light spectrum range of the spectrometer is 2.0-2.35μm. Since 2.35-2.5μm is the atmospheric absorption peak, it is no longer enhanced. In order to maximize the signal-to-noise ratio gain and reduce the crosstalk between spectral bands, a local dispersion filter is placed after the two local dispersion slits. The filter has a bandpass spectrum range of 2.0-2.35μm and is cut off in the spectral range of 0.9-2.0μm and 2.35-2.5μm. The dispersion results are shown in the figure. Figure 3 shown.

[0049] Since the shortwave infrared spectrometer disperses 98 spectral bands from 0.9 to 2.5 μm, 23 spectral bands from 2.0 to 2.35 μm, and 11 spectral bands from 2.35 to 2.5 μm, n = 98, m = 23, and s = 64. Therefore, the spacing n+1 between the two full dispersion slits is 99 pixels, and the spacing m+1 between the two partial dispersion slits is 24 pixels. The two partial dispersion slits and the two full dispersion slits are arranged sequentially along the dispersion direction of the full dispersion slits. The spacing between adjacent partial dispersion slits and full dispersion slits is 88 pixels, and the spectral data of the two adjacent slits is separated by one pixel.

[0050] By linearly adding the spectral data from two complete dispersion slits and two partial dispersion slits to suppress random noise, the signal-to-noise ratio can be enhanced. The signal-to-noise ratio in the weak light spectral range of 2.0 to 2.35 μm is calculated according to the following formula:

[0051]

[0052] λ i ∈[2.0,2.35]

[0053] Among them, SNR(λ i ) is the wavelength λ i The signal-to-noise ratio, is the spectrometer wavelength λ i The number of signal electrons, σ n is the comprehensive noise electron number including readout noise, quantization noise, dark current noise, circuit noise, etc., L(λ i ) is the wavelength λ i The entrance pupil radiance, A d is the pixel area of ​​the detector in the spectrometer, 1 / F is the relative aperture of the system, τ(λ i ) is the wavelength λ i The optical transmittance, η(λ i ) is the wavelength λ i The quantum efficiency, T int is the exposure time.

[0054] The signal-to-noise ratios for other spectral ranges, 0.9-2.0 μm and 2.35-2.5 μm, are:

[0055]

[0056] λ i ∈[0.9,2.0]∪[2.35,2.5]

[0057] Use a single-slit spectrometer to perform spectral imaging, obtain corresponding spectrometer imaging data, and calculate the signal-to-noise ratio after removing the atmospheric absorption peak. At the same time, calculate the signal-to-noise ratio after removing the atmospheric absorption peak in this embodiment. The results are as follows: Figure 4 and Figure 5 As shown in the figure, by comparing the two, it can be seen that the present invention can achieve a 2-fold increase in the signal-to-noise ratio in the weak light spectral range of 2.0 to 2.35 μm, a 1.55-fold increase in the signal-to-noise ratio in the full spectral range, a 5.1 dB increase in the average signal-to-noise ratio, and a 7.2 dB increase in the minimum signal-to-noise ratio, thereby achieving an improvement in the signal-to-noise ratio in the full spectral range and the weak light spectral range. The engineering implementation is simple, and it provides a technical means for achieving high signal-to-noise ratio detection in high-resolution spectral remote sensing.

Claims

1. A composite multi-slit based spectrometer, comprising a detector, a light splitting system, and a slit assembly disposed at an entrance end of the light splitting system, the slit assembly comprising N complete dispersion slits and M partial dispersion slits arranged side by side on the same plane, wherein N and M satisfy the following relationship: N × (n + 1) + M × (m + 1) - 1 ≤ A, wherein A is the number of pixels in the spectral direction of the detector, n is the total number of spectral segments corresponding to the full spectral range of the spectrometer, and m is the number of spectral segments corresponding to the weak light spectral range of the spectrometer; the N complete dispersion slits are arranged adjacently, and the M partial dispersion slits are arranged adjacently; the spacing between adjacent complete dispersion slits is n + 1 pixel sizes; and the spacing between adjacent partial dispersion slits is m + 1 pixel sizes. 2.The composite multi-slit based spectrometer of claim 1, wherein the N complete dispersion slits and the M partial dispersion slits are arranged in a staggered manner. 3.The composite multi-slit based spectrometer of claim 2, wherein the N complete dispersion slits and the M partial dispersion slits are arranged in a staggered manner. 4.The composite multi-slit based spectrometer of claim 2, comprising the following steps: S1, aligning the N complete dispersion slits and the M partial dispersion slits of the composite multi-slit based spectrometer of claim 1 with a target to be measured, respectively, performing spectral imaging on the target to be measured, and obtaining N + M spectral data; S2, linearly adding the obtained N + M spectral data to obtain spectral data of the target to be measured with enhanced signal-to-noise ratio, and completing the spectral imaging based on the composite multi-slit. 6.The imaging method of the composite multi-slit based spectrometer of claim 5, wherein the N complete dispersion slits and the M partial dispersion slits are arranged in a staggered manner. ​ ​ ​ Definition, the full spectral range of the spectrometer is λ1~λ n ; the weak light spectral range is λ s+1 ~λ s+m ; wherein, s+1 is the starting spectral segment number of the weak light spectral range, and s+m is the terminal spectral segment number of the weak light spectral range; M pieces of the local dispersion slits are provided with a local dispersion filter in the light-out direction of the local dispersion slits, and the local dispersion filter has a band-pass spectral interval of λ s+1 ~ λ s+m . ​ The weak light spectral interval λ s+1 ~ λ s+m Close to λ1, N complete dispersion slits and M partial dispersion slits are sequentially arranged along the dispersion direction of the complete dispersion slit, and the spacing between adjacent partial dispersion slits and complete dispersion slits is n-s+1 pixel size. ​ The weak light spectral interval λ s+1 ~ λ s+m Close to λ n When, M partial dispersion slits and N complete dispersion slits are sequentially arranged along the dispersion direction of the complete dispersion slit, and the spacing between adjacent partial dispersion slit and complete dispersion slit is s+m+1 pixel size.

5. An imaging method based on a composite multi-slit spectrometer, characterized in that, ​ ​ ​ ​ In step S1, a partial dispersion filter is arranged in the light-out direction of the M partial dispersion slits, and the band-pass spectral range of the partial dispersion filter is λ s+1 ~ λ s+m .

Images