Spectrometer stray light correction method based on optical filter

Abstract

The invention relates to a spectrum technology, and particularly provides an optical filter-based spectrograph stray light correction method, which comprises the following steps of: measuring an original spectrum DN and a spectrum DNfi after light of a light source passes through an ith optical filter fi under the same condition; the spectral range [lambda min, lambda max] of the spectrometer is divided into (N + 1) wavebands by the cut-off wavelength lambda fi of each optical filter; the spectral response difference value of two adjacent optical filters fi and fi + 1 in the jth wave band is delta DNjfi, fi + 1 = DNfi (lambdafj-1, lambdafj)-DNfi + 1 (lambdafj-1, lambdafj), i > = j, and j = 1, 2,..., N-1; the light of the wave band [lambda fi, lambda fi + 1 + gfi + 1] is shielded by an optical filter fi + 1, and the reduced spectral response area of the shielded DNfi is Afi; the proportion of stray light contributed to the jth wave band by Afi is kjfi, fi + 1; calculating the area A 'fi of a theoretical bright spectrum DN' fi after the original spectrum DN in the wave band [lambda fi, lambda fi + 1 + gfi + 1) passes through the optical filter fi; calculating a proportion alpha i = Afi / A 'fi; and applying the correction coefficient to the actually measured spectrum DN to obtain the spectrum DNC after stray light correction. The method has the advantages of good correction effect and the like.

Description

Technical Field

[0001] This invention relates to spectral technology, and in particular to a method for correcting stray light in a spectrometer based on filters. Background Technology

[0002] Stray light is a ubiquitous and difficult-to-eliminate interference factor in optical instruments such as spectrometers. Its causes are complex, and even with precise optomechanical design and optimization, a significant proportion of stray light may still remain, greatly affecting the measurement accuracy of the instrument. In addition to hardware-level suppression measures, post-processing image techniques can be used to correct this residual stray light.

[0003] The matrix method is one of the methods for correcting stray light in post-image processing. Its core principle is to measure the stray light distribution characteristics of the instrument at each pixel using a monochromatic light source (such as a tunable laser or monochromator) and establish a response matrix that characterizes the energy transfer law of stray light. Specifically, by systematically measuring the monochromatic light input signal covering the full spectrum of the instrument and the resulting stray light distribution, the cross-interference weights of stray light between different wavelengths can be quantified, and then a stray light correction matrix can be constructed to correct the instrument's stray light.

[0004] Among the monochromatic light sources used to measure stray light correction matrices, tunable lasers are extremely expensive to manufacture and maintain, making them unaffordable for most spectrometer manufacturers or laboratories. While monochromators are more readily available than tunable lasers, measuring stray light matrices requires high-quality stray light from the monochromator, and monochromators with low stray light are also quite expensive.

[0005] Chinese patent CN101813519B proposes a filter correction method for stray light. This method establishes a formula based on the original spectrum, the spectrum after passing through the filter, and the characteristics of the filter to calculate the proportion of stray light within the filter's cutoff region, and then corrects the original spectrum using this proportion. However, this method is only effective for spectral signals obtained from normal spectral measurements. If stray light testing is performed, such as using a filtering method to test stray light in the ultraviolet band, the signal in the ultraviolet band is filtered out by the solution, and the signal value is equivalent to the stray light in the ultraviolet band. In this case, the method will not have a significant correction effect. Summary of the Invention

[0006] To address the shortcomings of the existing technical solutions, this invention provides a method for correcting stray light in a spectrometer based on a filter.

[0007] The objective of this invention is achieved through the following technical solution: A method for correcting stray light in a spectrometer based on filters, wherein the correction method is as follows: Under the same conditions, the original spectrum DN is measured, and the light from the light source passes through the i-th filter f. iThe subsequent spectral DN fi , i = 1, 2, ..., N; Spectral range of the spectrometer [λ] min ,λ max The cutoff wavelength λ of each filter fi Divided into (N+1) bands, including [λ min ,λ f1 ),[λ f1 ,λ f2 )···[λ fN-1 ,λ fN ),[λ fN ,λ max Let λ min =λ f0 ,λ max =λ fN+1 ; Two adjacent filters f in the j-th band i and f i+1 The difference in spectral response is; ΔDN j fi,fi+1 =DN fi (λ fj-1 ,λ fj )-DN fi+1 (λ fj-1 ,λ fj ), i≥j, j=1,2, ···,N-1; DN fi (λ fj-1 ,λ fj ) is the light from the light source passing through the i-th filter f. i Then in the j-th band [λ fj-1 ,λ fj Spectral response within ) Band [λ] fi ,λ fi+1 +g fi+1 The light is filtered by filter f i+1 Obstruction, DN after being obstructed fi The area of ​​the reduced spectral response is g fi+1 It is the width of the transition region of the (i+1)th filter; By A fi The proportion of stray light contributed to the j-th band, k j fi,fi+1 =ΔDN j fi,fi+1 / A fi ; Calculate band [λ] fi ,λ fi+1 +g fi+1The original spectral DN within the filter f i Theoretical Brightness Spectrum DN ˊ fi area ; Calculate the proportion α i =A fi / A ˊ fi DN ˊ fi The light from the light source passes through the filter f i The theoretical brightness spectrum afterwards.

[0008] Compared with the prior art, the beneficial effects of the present invention are as follows: This application estimates the stray light ratio based on the intensity of the uncorrected interval, thus it can still achieve effective correction during stray light testing and has wider applicability. Attached Figure Description

[0009] The disclosure of this invention will become more readily understood with reference to the accompanying drawings. It will be readily understood by those skilled in the art that these drawings are merely illustrative of the technical solutions of this invention and are not intended to limit the scope of protection of this invention. In the drawings: Figure 1 This is a schematic diagram of the spectrum of light from the light source according to the present invention after passing through each filter; Figure 2 This is a schematic diagram of the theoretical bright spectrum and corresponding area within a certain wavelength range according to the present invention; Figure 3 This is a schematic diagram for calculating the cutoff wavelength and transition region width based on the measured transmittance curve of a certain filter according to the present invention. Figure 4 This is the corrected spectrum. Detailed Implementation

[0010] Figures 1-4 The following description illustrates optional embodiments of the invention to teach those skilled in the art how to implement and reproduce the invention. Some conventional aspects have been simplified or omitted to explain the technical solutions of the invention. Those skilled in the art should understand that variations or substitutions derived from these embodiments will be within the scope of the invention. Those skilled in the art should understand that the following features can be combined in various ways to form multiple variations of the invention. Therefore, the invention is not limited to the following optional embodiments, but is defined only by the claims and their equivalents.

[0011] Example 1

[0012] The filter-based spectrometer stray light correction method of this invention is as follows: Under the same conditions, the original spectrum DN is measured, and the light from the light source passes through the i-th filter f. i The subsequent spectral DN fi , i=1,2,···,N, such as Figure 1 As shown.

[0013] Spectral range of the spectrometer [λ] min ,λ max The cutoff wavelength λ of each filter fi Divided into (N+1) bands, including [λ min ,λ f1 ),[λ f1 ,λ f2 )···[λ fN-1 ,λ fN ),[λ fN ,λ max Let λ min =λ f0 ,λ max =λ fN+1 ; Two adjacent filters f in the j-th band i and f i+1 The difference in spectral response is; ΔDN j fi,fi+1 =DN fi (λ fj-1 ,λ fj )-DN fi+1 (λ fj-1 ,λ fj ), i≥j, j=1,2, ···,N-1; DN fi (λ fj-1 ,λ fj ) is the light from the light source passing through the i-th filter f. i Then in the j-th band [λ fj-1 ,λ fj Spectral response within ) Band [λ] fi ,λ fi+1 +g fi+1 The light is filtered by filter f i+1 Obstruction, DN after being obstructed fi The area of ​​the reduced spectral response is ,like Figure 2 As shown.

[0014] g fi+1 It is the width of the transition zone of the (i+1)th filter.

[0015] By A fi The proportion of stray light contributed to the j-th band, kj fi,fi+1 =ΔDN j fi,fi+1 / A fi。

[0016] Calculate band [λ] fi ,λ fi+1 +g fi+1 The original spectral DN within the filter f i Post-theoretical brightness spectrum DN ˊ fi area ,like Figure 2 As shown.

[0017] Calculate the proportion α i =A fi / A ˊ fi DN ˊ fi The light from the light source passes through the filter f i The theoretical brightness spectrum afterwards.

[0018] For the spectral DN actually measured by the spectrometer, the band [λ] fi ,λ fi+1 +g fi+1 The spectral area within ) is .

[0019] The total amount of stray light in the j-th band is .

[0020] ΔDN is assembled according to the band order. j The missing bands are filled with preset values ​​to generate a stray light vector ΔDN with the same dimension as the original signal DN.

[0021] Corrected spectral DN C =DN-ΔDN.

[0022] Theoretical Brightness Spectrum DN ˊ fi for: .

[0023] mean(τ p fi ) represents the transmittance τ p fi The average value of CubicSpline([mean(τ)) p fi ),τ p fi ]) represents the use of cubic spline interpolation in the transition region to approximate the transmittance of the filter in this region, τ pfi It is the passband transmittance of the i-th filter.

[0024] The transmittance curve τ(λ) of a certain filter in the filter group is measured using a spectrometer, and its first derivative is obtained as τˊ(λ)=dτ(λ) / dλ.

[0025] A skewed Gaussian fit is performed on τˊ(λ), the peak point λ0 is located from the fitted curve, and the left half-width HW is calculated. l and right half width HW r。

[0026] Cutoff wavelength λ c =λ0-3HW l The transition zone width g = 3 (HW) r +HW l ),like Figure 3 As shown.

[0027] Example 2

[0028] An application example of the filter-based spectrometer stray light correction method according to Embodiment 1 of the present invention.

[0029] Under the same conditions, the original spectral DN was measured. The spectral DN of the light source after passing through nine long-pass filters f1, f2, ..., f9 was measured. fi , i=1,2,···,9, such as Figure 1 As shown.

[0030] Calculate k respectively j fi,fi+1 and α i For the spectral DN obtained from actual measurements, a correction algorithm is applied to obtain the corrected spectral DN. C ,like Figure 4 As shown.

[0031] For the long-pass filter used, the algorithm has a better correction effect in the short-wavelength range.

Claims

1. A method for correcting stray light in a spectrometer based on filters, characterized in that, The correction method is as follows: Under the same conditions, the original spectrum DN is measured, and the light from the light source passes through the i-th filter f. i The subsequent spectral DN fi , i = 1, 2, ..., N; Spectral range of the spectrometer [λ] min ,λ max The cutoff wavelength λ of each filter fi Divided into (N+1) bands, including [λ min ,λ f1 ),[λ f1 ,λ f2 )···[λ fN-1 ,λ fN ),[λ fN ,λ max Let λ min =λ f0 , λ max =λ fN+1 ; Two adjacent filters f in the j-th band i and f i+1 The difference in spectral response is; ΔDN j fi,fi+1 =DN fi (l fj-1 ,l fj )-DN fi+1 (l fj-1 ,l fj ),i≥j,j=1,2, ···,N-1; DN fi (λ fj-1 ,λ fj ) is the light from the light source passing through the i-th filter f. i Then in the j-th band [λ fj-1 ,λ fj Spectral response within ) Band [λ] fi ,λ fi+1 +g fi+1 The light is filtered by filter f i+1 Obstruction, DN after being obstructed fi The area of ​​the reduced spectral response is g fi+1 It is the width of the transition region of the (i+1)th filter; By A fi The proportion of stray light contributed to the j-th band is k. j fi,fi+1 =ΔDN j fi,fi+1 / A fi ; Calculate band [λ] fi ,λ fi+1 +g fi+1 The original spectral DN within the filter f i Theoretical Brightness Spectrum DN ˊ fi area ; Calculate the proportion α i =A fi / A ˊ fi .

2. The correction method according to claim 1, characterized in that, Theoretical Brightness Spectrum DN ˊ fi for: ； mean(τ p fi ) represents the transmittance τ p fi The average value of CubicSpline([mean(τ)) p fi ),τ p fi ]) indicates that cubic spline interpolation is used to approximate the transmittance of the filter in the transition region; τ p fi It is the i-th filter f i The passband transmittance.

3. The correction method according to claim 1, characterized in that, For the spectral DN actually measured by the spectrometer, the band [λ] fi ,λ fi+1 +g fi+1 The spectral area within ) is ; The total amount of stray light in the j-th band is ; ΔDN is assembled according to the band order. j The missing bands are filled with preset values ​​to generate stray light vector ΔDN with the same dimension as the original signal DN. Corrected spectral DN C =DN-ΔDN.

4. The correction method according to claim 2, characterized in that, The transmittance curve τ(λ) of the filter was measured using a spectrometer, and τˊ(λ) = dτ(λ) / dλ was obtained. A skewed Gaussian fit is performed on τˊ(λ), the peak point λ0 is located from the fitted curve, and the left half-width HW is calculated. l and right half width HW r ; Cutoff wavelength λ c =λ0-3HW l The transition zone width g = 3 (HW) r +HW l ).

5. The correction method according to claim 1, characterized in that, The filter is a long-pass filter, a short-pass filter, or a bandpass filter.

Images