CN104316183A - Method and device for correcting bent spectral lines of dispersion type spectral imager - Google Patents

Abstract

The invention discloses a method and device for correcting bent spectral lines of a dispersion type spectral imager. The method for correcting the bent spectral lines comprises the steps of calculating the Euclidean distance between central wavelengths and non-deviation central wavelengths lambda j of all spectral channels in the row m, selecting the position points s with the minimum distance and central wavelengths lambda m, s, selecting position points and central wavelengths adjacent to the s from the adjacent spectral channels, calculating the Euclidean distance between the central wavelengths lambda m, s corresponding to s and the non-deviation central wavelengths lambda j, and the Euclidean distance between the central wavelengths corresponding to the adjacent position points and the non-deviation central wavelengths lambda j, according to the distance calculation result, calculating the inverse distance weighting value of the central wavelengths lambda m, s corresponding to the s, and the inverse distance weighting value of the central wavelengths corresponding to the adjacent position points, and according to the calculation result of the inverse distance weighting values, calculating the spectral radiance energy value, corresponding to the non-deviation central wavelengths lambda j, of the row m. The inverse distance weighting method is adopted to carry out correction of the bent spectral lines, the precision of spectral line correction can be improved, the algorithm is simple, and the calculation speed is high.

Description

A kind of color dispersion-type optical spectrum imagers Spectral line bend bearing calibration and device

Technical field

The present invention relates to optical spectrum imagers technical field, particularly relate to a kind of color dispersion-type optical spectrum imagers Spectral line bend bearing calibration and device.

Background technology

Spectral imaging technology is a kind of novel information acquiring technology grown up the eighties in 20th century, can obtain two-dimensional space information and the one dimension spectral information of target simultaneously, be widely applied in civil and military field.Different according to spectroscopic modes, spectral imaging technology can be divided into color dispersion-type, interfere type, Tunable filters type, calculating computed tomography type etc.Wherein color dispersion-type spectral imaging technology comparative maturity, apply also extensive, but due to the existence of optical system alignment error, its Spectral line bend (smile effect) is larger, therefore, need to carry out accurate spectral calibration to each pixel, determine operating central wavelength and the bandwidth of emission spectrum of each pixel, can Spectral line bend correction be carried out according to this centre wavelength matrix, thus realize the consistance of emittance collection.

Demarcation about pixel operating central wavelength has two kinds of methods usually: spectral line lamp method and monochromator method.Because most of image analysis system process data all think that all spaces pixel of a corresponding spectrum channel has identical centre wavelength, Spectral line bend destroys the consistance that color dispersion-type optical spectrum imagers wears the same row/column diverse location pixel spectral radiant energy collection of rail direction, so need to system acquisition to emittance data according to carrying out resampling to correct the deviation of emittance without departing from wavelength data, the consistance of raising spectral radiant energy.

Spectral line bend bearing calibration mainly method of interpolation conventional at present: think that a jth spectrum channel i-th+m pixel is i-th pixel central wavelength lambda _jspectral radiant energy approximate value φ _{i+m, j}rebuild, namely by wavelength X on the curve of spectrum the spectral radiant energy data that can collect from the i-th+m row pixel _jneighbouring some discrete data point interpolations obtain.Suppose that the spectral radiant energy data that the i-th+m row pixel collects are φ _{i+m, 1}(λ _{i+m, 1}), φ _{i+m, 2}(λ _{i+m, 2}) ..., φ _{i+m, j-1}(λ _{i+m, j-1}), φ _{i+m, j}(λ _{i+m, j}), φ _{i+m, j+1}(λ _{i+m, j+1}) ..., φ _{i+m, n-1}(λ _{i+m, n-1}), φ _{i+m, n}(λ _{i+m, n}) wherein n be port number, only need by these discrete datas and have limit for length's bump a to respond central wavelength lambda that convolution can obtain (i+m, j) individual pixel corresponding (i, j) individual pixel being called as interpolation kernel _jspectral radiant energy value φ ' _{i+m, j}(λ _{i, j}).The concrete form of this convolution kernel can be linear, 3 quadratic polynomials, 43 Lagrange, 54 Lagrange, three Hermite and cubic spline interpolation methods etc.

The selection of interpolation method is relevant with spectral resolution, and different interpolation method its precision under different spectral resolutions is different, so versatility is not strong.

Summary of the invention

The object of the embodiment of the present invention is to provide a kind of color dispersion-type optical spectrum imagers Spectral line bend bearing calibration and device, improves the precision correcting spectrum.

The object of the embodiment of the present invention is achieved through the following technical solutions:

The bearing calibration of a kind of color dispersion-type optical spectrum imagers Spectral line bend, comprising:

Calculate m arrange all spectrum channel centre wavelength with without departing from central wavelength lambda _jbetween Euclidean distance, the location point s that chosen distance is minimum and centre wavelength corresponding to described location point are λ _{m, s}, on adjacent spectrum channel, select the location point adjacent with described location point s and centre wavelength corresponding to described adjacent location point, wherein, m value is 1 ~ N, j value be 1 ~ M, M is the number of pixels of wearing rail direction, and N is the number of pixels along rail direction;

Calculate the central wavelength lambda that described location point s is corresponding _{m, s}central wavelength lambda is departed from nothing _jbetween Euclidean distance, and centre wavelength corresponding to described adjacent location point with without departing from central wavelength lambda _jbetween Euclidean distance;

Central wavelength lambda corresponding to described location point s is calculated according to above-mentioned distance result of calculation _{m, s}anti-distance weighting value, and the anti-distance weighting value of centre wavelength corresponding to described adjacent location point;

Calculate the corresponding described nothing of described m row according to above-mentioned anti-distance weighting value result of calculation and depart from central wavelength lambda _jspectral radiant energy value.

A kind of color dispersion-type optical spectrum imagers Spectral line bend means for correcting, comprising:

Position determination unit, for calculate m arrange all spectrum channel centre wavelength with without departing from central wavelength lambda _jbetween Euclidean distance, the location point s that chosen distance is minimum and centre wavelength corresponding to described location point are λ _{m, s}, on adjacent spectrum channel, select the location point adjacent with described location point s and centre wavelength corresponding to described adjacent location point, wherein, m value is 1 ~ N, j value be 1 ~ M, M is the number of pixels of wearing rail direction, and N is the number of pixels along rail direction;

Distance determining unit, for calculating central wavelength lambda corresponding to described location point s _{m, s}central wavelength lambda is departed from nothing _jbetween Euclidean distance, and centre wavelength corresponding to described adjacent location point with without departing from central wavelength lambda _jbetween Euclidean distance;

Weight determining unit, for calculating central wavelength lambda corresponding to described location point s according to the result of calculation of above-mentioned distance determining unit _{m, s}anti-distance weighting value, and the anti-distance weighting value of centre wavelength corresponding to described adjacent location point;

Spectral radiant energy determining unit, departs from central wavelength lambda for calculating the corresponding described nothing of described m row according to the result of calculation of weight determining unit _jspectral radiant energy value.

The technical scheme provided as can be seen from the invention described above embodiment, by calculating the distance departing from the immediate multiple point of centre wavelength with nothing, calculate these depart from centre wavelength weight contribution to nothing, the spectral radiant energy corresponding to this centre wavelength according to weight contribution value recalculates assignment, thus the relation farthest remained between adjacent band, spectrum fidelity is higher, and algorithm is simple, and computing velocity is fast.

Accompanying drawing explanation

In order to be illustrated more clearly in the technical scheme of the embodiment of the present invention, below the accompanying drawing used required in describing embodiment is briefly described, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawings can also be obtained according to these accompanying drawings.

Fig. 1 is the schematic flow sheet of embodiment of the present invention color dispersion-type optical spectrum imagers Spectral line bend bearing calibration.

Fig. 2 is the formation schematic diagram of embodiment of the present invention color dispersion-type optical spectrum imagers Spectral line bend means for correcting.

Fig. 3 is Spectral line bend schematic diagram in the bearing calibration of embodiment of the present invention color dispersion-type optical spectrum imagers Spectral line bend.

Fig. 4 is the application schematic diagram of embodiment of the present invention color dispersion-type optical spectrum imagers Spectral line bend bearing calibration.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on embodiments of the invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to protection scope of the present invention.

As shown in Figure 1, the bearing calibration of a kind of color dispersion-type optical spectrum imagers of embodiment of the present invention Spectral line bend, comprising:

11, calculate m arrange all spectrum channel centre wavelength with without departing from central wavelength lambda _jbetween Euclidean distance, the location point s that chosen distance is minimum, and centre wavelength corresponding to described location point is λ _{m, s}, on adjacent spectrum channel, select the location point adjacent with described location point s and centre wavelength corresponding to described adjacent location point, wherein, m value is 1 ~ N, j value be 1 ~ M, M is the number of pixels of wearing rail direction, and N is the number of pixels along rail direction;

12, central wavelength lambda corresponding to described location point s is calculated _{m, s}central wavelength lambda is departed from nothing _jbetween Euclidean distance, and centre wavelength corresponding to described adjacent location point with without departing from central wavelength lambda _jbetween Euclidean distance;

13, central wavelength lambda corresponding to described location point s is calculated according to above-mentioned distance result of calculation _{m, s}anti-distance weighting value, and the anti-distance weighting value of centre wavelength corresponding to described adjacent location point;

14, calculate the corresponding described nothing of described m row according to above-mentioned anti-distance weighting value result of calculation and depart from central wavelength lambda _jspectral radiant energy value.

The technical scheme provided as can be seen from the invention described above embodiment, by calculating the distance departing from the immediate some points of centre wavelength with nothing, calculate these depart from centre wavelength weight contribution to nothing, the spectral radiant energy corresponding to this centre wavelength according to weight contribution value recalculates assignment, thus the relation farthest remained between adjacent band, spectrum fidelity is higher, and algorithm is simple, and computing velocity is fast.

According to prior art, those skilled in the art are appreciated that calculating m arranges all spectrum channel centre wavelength and departs from central wavelength lambda with nothing _jbetween the implementation of Euclidean distance, therefore not to repeat here.

It will be understood by those skilled in the art that calculating m arranges all spectrum channel centre wavelength and departs from central wavelength lambda with nothing _jbetween Euclidean distance, the location point that chosen distance is minimum, can be assumed to be s.

It will be understood by those skilled in the art that the Euclidean distance related to all refers to Euclidean distance herein, it will be appreciated by those skilled in the art that Euclidean distance, therefore not to repeat here.

The bearing calibration of embodiment of the present invention color dispersion-type optical spectrum imagers Spectral line bend, can select and be two points, three points, four points, five points etc. without departing from the immediate point of centre wavelength, does not do strict restriction.

Optionally, determine that the location point adjacent with described location point s can be:

Determine front position point s-1 adjacent with described location point s, or, determine rear location point s+1 adjacent with described location point s.

Visible, select 1 point near location point s, namely departing from the immediate point of centre wavelength with nothing is 2.

Optionally, determine that the location point adjacent with described location point s is:

Determine with described location point s before and after adjacent 3 location points;

Or, determine with described location point s before and after adjacent 4 location points.

Visible, select 3 points near location point s, namely departing from the immediate point of centre wavelength with nothing is 4.

Select 4 points near location point s, namely departing from the immediate point of centre wavelength with nothing is 5.

Considering computational accuracy and computing velocity, is good with three to four points, for three points:

Determine front position point s-1 adjacent with described location point s and rear location point s+1;

Now, the central wavelength lambda that described front position point s-1 is corresponding is obtained _{m, s-1}and the central wavelength lambda that described rear location point s+1 is corresponding _{m, s+1};

Computing center's wavelength X _{m, s-1}central wavelength lambda is departed from nothing _jbetween Euclidean distance Δ _s-1, central wavelength lambda _{m, s}central wavelength lambda is departed from nothing _jbetween Euclidean distance Δ _s, and central wavelength lambda _{m, s+1}central wavelength lambda is departed from nothing _jbetween Euclidean distance Δ _s+1, wherein, Δ _s-1=| λ _{m, s-1}-λ _j|; Δ _s=| λ _{m, s}-λ _j|; Δ _s+1=| λ _{m, s+1}-λ _j|;

Computing center's wavelength X _{m, s-1}anti-distance weighting value Q _s-1, central wavelength lambda _{m, s}anti-distance weighting value Q _s, and central wavelength lambda _{m, s+1}anti-distance weighting value Q _s+1, wherein, $Q_{s-1}=\frac{1/{\mathrm{\Δ}}_{s-1}}{1/{\mathrm{\Δ}}_{s-1}+1/{\mathrm{\Δ}}_s+1/{\mathrm{\Δ}}_{s+1}};Q_s=\frac{1/{\mathrm{\Δ}}_s}{1/{\mathrm{\Δ}}_{s-1}+1/{\mathrm{\Δ}}_s+1/{\mathrm{\Δ}}_{s+1}};$ $Q_{s+1}=\frac{1/{\mathrm{\Δ}}_{s+1}}{1/{\mathrm{\Δ}}_{s-1}+1/{\mathrm{\Δ}}_s+1/{\mathrm{\Δ}}_{s+1}};$

Calculate the corresponding described nothing of described m row and depart from central wavelength lambda _jspectral radiant energy value φ ' _{m, j}(λ _j)=φ _{m, s-1}(λ _{m, s-1}) Q _s-1+ φ _{m, s}(λ _{m, s}) Q _s+ φ _{m, s+1}(λ _{m, s+1}) Q _s+1wherein, for the response of the respective pixel that the measurement of color dispersion-type optical spectrum imagers obtains.

Visible, depart from central wavelength lambda with nothing _jmore close spectrum channel, its centre wavelength and λ _jdistance is shorter, and anti-distance weighting value corresponding is with it then larger, then this wavelength is to calculating φ ' _{m, j}(λ _j) contribution larger, this is just in time true to life.

The spectral radiant energy corresponding to this centre wavelength according to weight contribution value recalculates assignment, thus farthest remains the relation between adjacent band, and spectrum fidelity is higher, and algorithm is simple, and computing velocity is fast.

As shown in Figure 2, the color dispersion-type optical spectrum imagers Spectral line bend bearing calibration of corresponding above-described embodiment, the embodiment of the present invention provides a kind of color dispersion-type optical spectrum imagers Spectral line bend means for correcting, comprising:

Position determination unit 21, for determining that m arranges all spectrum channel centre wavelength and departs from central wavelength lambda with nothing _jbetween nearest location point s, determine the location point adjacent with described location point s, and obtain central wavelength lambda corresponding to described location point s _{m, s}and the centre wavelength that described adjacent location point is corresponding, wherein, m value is 1 ~ N, j value be 1 ~ M, M is the number of pixels of wearing rail direction, and N is the number of pixels along rail direction;

Distance determining unit 22, for calculating central wavelength lambda corresponding to described location point s _{m, s}central wavelength lambda is departed from nothing _jbetween Euclidean distance, and centre wavelength corresponding to described adjacent location point with without departing from central wavelength lambda _jbetween Euclidean distance;

Weight determining unit 23, for calculating central wavelength lambda corresponding to described location point s according to the result of calculation of above-mentioned distance determining unit 22 _{m, s}anti-distance weighting value, and the anti-distance weighting value of centre wavelength corresponding to described adjacent location point;

Spectral radiant energy determining unit 24, departs from central wavelength lambda for calculating the corresponding described nothing of described m row according to the result of calculation of weight determining unit 23 _jspectral radiant energy value.

The technical scheme provided as can be seen from the invention described above embodiment, by calculating the distance departing from the immediate some points of centre wavelength with nothing, calculate these depart from centre wavelength weight contribution to nothing, the spectral radiant energy corresponding to this centre wavelength according to weight contribution value recalculates assignment, thus the relation farthest remained between adjacent band, spectrum fidelity is higher, and algorithm is simple, and computing velocity is fast.

Embodiment of the present invention color dispersion-type optical spectrum imagers Spectral line bend means for correcting, can select and be two points, three points, four points, five points etc. without departing from the immediate point of centre wavelength, does not do strict restriction.

Optionally, position determination unit, specifically may be used for: determine front position point s-1 adjacent with described location point s, or, determine rear location point s+1 adjacent with described location point s.

Visible, select 1 point near location point s, namely departing from the immediate point of centre wavelength with nothing is 2.

Optionally, position determination unit, specifically may be used for: determine with described location point s before and after adjacent 3 location points;

Or, determine with described location point s before and after adjacent 4 location points.

Visible, select 3 points near location point s, namely departing from the immediate point of centre wavelength with nothing is 4.

Select 4 points near location point s, namely departing from the immediate point of centre wavelength with nothing is 5.

Considering computational accuracy and computing velocity, is good with three to four points, for three points:

Position determination unit 21, specifically may be used for: determine front position point s-1 adjacent with described location point s and rear location point s+1; Now, the central wavelength lambda that described front position point s-1 is corresponding is obtained _{m, s-1}and the central wavelength lambda that described rear location point s+1 is corresponding _{m, s+1};

Distance determining unit 22, specifically may be used for computing center's wavelength X _{m, s-1}central wavelength lambda is departed from nothing _jbetween Euclidean distance Δ _s-1, central wavelength lambda _{m, s}central wavelength lambda is departed from nothing _jbetween Euclidean distance Δ _s, and central wavelength lambda _{m, s+1}central wavelength lambda is departed from nothing _jbetween Euclidean distance Δ _s+1, wherein, Δ _s-1=| λ _{m, s-1}-λ _j|; Δ _s=| λ _{m, s}-λ _j|; Δ _s+1=| λ _{m, s+1}-λ _j|;

Weight determining unit 23, specifically may be used for computing center's wavelength X _{m, s-1}anti-distance weighting value Q _s-1, central wavelength lambda _{m, s}anti-distance weighting value Q _s, and central wavelength lambda _{m, s+1}anti-distance weighting value Q _s+1, wherein, $Q_{s-1}=\frac{1/{\mathrm{\Δ}}_{s-1}}{1/{\mathrm{\Δ}}_{s-1}+1/{\mathrm{\Δ}}_s+1/{\mathrm{\Δ}}_{s+1}};$ $Q_s=\frac{1/{\mathrm{\Δ}}_s}{1/{\mathrm{\Δ}}_{s-1}+1/{\mathrm{\Δ}}_s+1/{\mathrm{\Δ}}_{s+1}};Q_{s+1}=\frac{1/{\mathrm{\Δ}}_{s+1}}{1/{\mathrm{\Δ}}_{s-1}+{1/\mathrm{\Δ}}_s+1/{\mathrm{\Δ}}_{s+1}};$

Spectral radiant energy determining unit 24, specifically may be used for calculating the corresponding described nothing of described m row and departs from central wavelength lambda _jspectral radiant energy value wherein, φ ' _{m, j}(λ _j)=φ _{m, s-1}(λ _{m, s-1}) Q _s-1+ φ _{m, s}(λ _{m, s}) Q _s+ φ _{m, s+1}(λ _{m, s+1}) Q _s+1, for the response of the respective pixel that the measurement of color dispersion-type optical spectrum imagers obtains.

Visible, the spectral radiant energy corresponding to this centre wavelength according to weight contribution value recalculates assignment, thus farthest remains the relation between adjacent band, and spectrum fidelity is higher, and algorithm is simple, and computing velocity is fast.

Be illustrated in figure 3 color dispersion-type optical spectrum imagers Spectral line bend schematic diagram, detector size is M × N, wears rail direction 32 and has M pixel, have N number of pixel, be spectrum direction along rail direction 31 along rail direction.

Ideally wear M the pixel corresponding Same Wavelength position of rail direction with a line, be assumed to be λ _jbut in fact due to the existence of systematic error and alignment error, make to wear rail direction and occur deviation with the centre wavelength of a line diverse location pixel, thus destroy color dispersion-type optical spectrum imagers and wear the consistance that rail direction spectral radiant energy gathers, so need to be corrected to same spectral position.

As shown in Figure 4, the bearing calibration of embodiment of the present invention color dispersion-type optical spectrum imagers Spectral line bend, comprising:

40, initial value is set, m=1, j=1.

41, calculate m arrange all channel center wavelength with without departing from central wavelength lambda _jnearest location point s.

42, the centre wavelength value λ that s point is corresponding is found _{m, s}, and front and back value λ _{m, s-1}, λ _{m, s+1}.

43, three centre wavelength values and λ is calculated _jeuclidean distance Δ _s-1, Δ _s, Δ _s+1.

If 44 Δs _s=0, the spectral radiant energy so corresponding to this wavelength location is without departing from energy value corresponding to central wavelength, that is: φ ' _{m, j}(λ _j)=φ _{m, s}(λ _{m, s}), forward step 47 to, otherwise forward 45-46 to.

45, the weighted value Q corresponding to three centre wavelength is calculated by inverse distance weighting _s-1, Q _s, Q _s+1.

46, m row corresponding wavelength λ is calculated _jspectral radiant energy value φ ' _{m, j}(λ _j).

47, judge whether that all path computations are complete, namely whether j equals total spectrum channel number N, if it is forwards 48 to, otherwise enters 49 couples of j+1, then continues step 41-46.

48, judge whether that all column counts are complete, namely whether m equals to wear rail pixel number M, if it is calculates complete, otherwise enters 410 couples of m+1, then repeat step 41-47.

Above; be only the present invention's preferably embodiment, but protection scope of the present invention is not limited thereto, is anyly familiar with those skilled in the art in the technical scope that the present invention discloses; the change that can expect easily or replacement, all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection domain of claims.

Claims (8)

1. a color dispersion-type optical spectrum imagers Spectral line bend bearing calibration, is characterized in that, comprising:

Calculate m arrange all spectrum channel centre wavelength with without departing from central wavelength lambda _jbetween Euclidean distance, the location point s that chosen distance is minimum and centre wavelength corresponding to described location point are λ _{m, s}, on adjacent spectrum channel, select the location point adjacent with described location point s and centre wavelength corresponding to described adjacent location point, wherein, m value is 1 ~ N, j value be 1 ~ M, M is the number of pixels of wearing rail direction, and N is the number of pixels along rail direction;

Calculate the central wavelength lambda that described location point s is corresponding _{m, s}central wavelength lambda is departed from nothing _jbetween Euclidean distance, and centre wavelength corresponding to described adjacent location point with without departing from central wavelength lambda _jbetween Euclidean distance;

Central wavelength lambda corresponding to described location point s is calculated according to above-mentioned distance result of calculation _{m, s}anti-distance weighting value, and the anti-distance weighting value of centre wavelength corresponding to described adjacent location point;

Calculate the corresponding described nothing of described m row according to above-mentioned anti-distance weighting value result of calculation and depart from central wavelength lambda _jspectral radiant energy value.

2. color dispersion-type optical spectrum imagers Spectral line bend according to claim 1 bearing calibration, is characterized in that, determines that the location point adjacent with described location point s is: determine front position point s-1 adjacent with described location point s and rear location point s+1;

Now, the central wavelength lambda that described front position point s-1 is corresponding is obtained _{m, s-1}and the central wavelength lambda that described rear location point s+1 is corresponding _{m, s+1};

Computing center's wavelength X _{m, s-1}central wavelength lambda is departed from nothing _jbetween Euclidean distance Δ _s-1, central wavelength lambda _{m, s}central wavelength lambda is departed from nothing _jbetween Euclidean distance Δ _s, and central wavelength lambda _{m, s+1}central wavelength lambda is departed from nothing _jbetween Euclidean distance Δ _s+1, wherein, Δ _s-1=| λ _{m, s-1}-λ _f|; Δ _s=| λ _{m, s}-λ _f|; Δ _s+1=| λ _{m, s+1-}λ _f|;

Computing center's wavelength X _{m, s-1}anti-distance weighting value Q _s-1, central wavelength lambda _{m, s}anti-distance weighting value Q _s, and central wavelength lambda _{m, s+1}anti-distance weighting value Q _s+1, wherein, $Q_{s-1}=\frac{1/{\mathrm{\Δ}}_{s-1}}{1/{\mathrm{\Δ}}_{s-1}+1/{\mathrm{\Δ}}_s+1/{\mathrm{\Δ}}_{s+1}};$ $Q_s=\frac{1/{\mathrm{\Δ}}_s}{1/{\mathrm{\Δ}}_{s-1}+1/{\mathrm{\Δ}}_s+1/{\mathrm{\Δ}}_{s+1}};$ $Q_{s+1}=\frac{1/{\mathrm{\Δ}}_{s+1}}{1/{\mathrm{\Δ}}_{s-1}+1/{\mathrm{\Δ}}_s+1/{\mathrm{\Δ}}_{s+1}};$

Calculate the corresponding described nothing of described m row and depart from central wavelength lambda _jspectral radiant energy value φ ' _{m, f}(λ _f)=φ _{m, s-1}(λ _{m, s-1}) Q _s-1+ φ _{m, s}(λ _{m, s}) Q _s+ φ _{m, s+1}(λ _{m, s+1}) Q _s+1.

3. color dispersion-type optical spectrum imagers Spectral line bend according to claim 1 bearing calibration, is characterized in that, determines that the location point adjacent with described location point s is:

Determine front position point s-1 adjacent with described location point s, or, determine rear location point s+1 adjacent with described location point s.

4. color dispersion-type optical spectrum imagers Spectral line bend according to claim 1 bearing calibration, is characterized in that, determines that the location point adjacent with described location point s is:

Determine with described location point s before and after adjacent 3 location points;

Or, determine with described location point s before and after adjacent 4 location points.

5. a color dispersion-type optical spectrum imagers Spectral line bend means for correcting, is characterized in that, comprising:

Position determination unit, arranges all spectrum channel centre wavelength with calculating m and departs from central wavelength lambda with nothing _jbetween Euclidean distance, the location point s that chosen distance is minimum and centre wavelength corresponding to described location point are λ _{m, s}, on adjacent spectrum channel, select the location point adjacent with described location point s and centre wavelength corresponding to described adjacent location point, wherein, m value is 1 ~ N, j value be 1 ~ M, M is the number of pixels of wearing rail direction, and N is the number of pixels along rail direction;

Distance determining unit, for calculating central wavelength lambda corresponding to described location point s _{m, s}central wavelength lambda is departed from nothing _jbetween Euclidean distance, and centre wavelength corresponding to described adjacent location point with without departing from central wavelength lambda _jbetween Euclidean distance;

Weight determining unit, for calculating central wavelength lambda corresponding to described location point s according to the result of calculation of above-mentioned distance determining unit _{m, s}anti-distance weighting value, and the anti-distance weighting value of centre wavelength corresponding to described adjacent location point;

Spectral radiant energy determining unit, departs from central wavelength lambda for calculating the corresponding described nothing of described m row according to the result of calculation of weight determining unit _jspectral radiant energy value.

6. color dispersion-type optical spectrum imagers Spectral line bend means for correcting according to claim 5, is characterized in that, described position determination unit, specifically for: determine front position point s-1 adjacent with described location point s and rear location point s+1;

Now, described distance determining unit, for obtaining central wavelength lambda corresponding to described front position point s-1 _{m, s-1}and the central wavelength lambda that described rear location point s+1 is corresponding _{m, s+1};

Described weight determining unit, for computing center's wavelength X _{m, s-1}central wavelength lambda is departed from nothing _jbetween Euclidean distance Δ _s-1, central wavelength lambda _{m, s}central wavelength lambda is departed from nothing _jbetween Euclidean distance Δ _s, and central wavelength lambda _{m, s+1}central wavelength lambda is departed from nothing _jbetween Euclidean distance Δ _s+1, wherein, Δ _s-1=| λ _{m, s-1}-λ _f|; Δ _s=| λ _{m, s}-λ _f|; Δ _s+1=| λ _{m, s+1}-λ _f|;

Described spectral radiant energy determining unit, for computing center's wavelength X _{m, s-1}anti-distance weighting value Q _s-1, central wavelength lambda _{m, s}anti-distance weighting value Q _s, and central wavelength lambda _{m, s+1}anti-distance weighting value Q _s+1, wherein, $Q_{s-1}=\frac{1/{\mathrm{\Δ}}_{s-1}}{1/{\mathrm{\Δ}}_{s-1}+1/{\mathrm{\Δ}}_s+1/{\mathrm{\Δ}}_{s+1}};$ $Q_s=\frac{1/{\mathrm{\Δ}}_s}{1/{\mathrm{\Δ}}_{s-1}+1/{\mathrm{\Δ}}_s+1/{\mathrm{\Δ}}_{s+1}};$ $Q_{s+1}=\frac{1/{\mathrm{\Δ}}_{s+1}}{1/{\mathrm{\Δ}}_{s-1}+1/{\mathrm{\Δ}}_s+1/{\mathrm{\Δ}}_{s+1}};$

Calculate the corresponding described nothing of described m row and depart from central wavelength lambda _jspectral radiant energy value φ ' _{m, f}(λ _f)=φ _{m, s-1}(λ _{m, s-1}) Q _s-1+ φ _{m, s}(λ _{m, s}) Q _s+ φ _{m, s+1}(λ _{m, s+1}) Q _s+1.

7. color dispersion-type optical spectrum imagers Spectral line bend means for correcting according to claim 5, is characterized in that, described position determination unit, specifically for:

Determine front position point s-1 adjacent with described location point s, or, determine rear location point s+1 adjacent with described location point s.

8. color dispersion-type optical spectrum imagers Spectral line bend means for correcting according to claim 5, is characterized in that, described position determination unit, specifically for:

Determine with described location point s before and after adjacent 3 location points;

Or, determine with described location point s before and after adjacent 4 location points.