CN104915956A - Sulfur fumigated traditional Chinese medicine discriminating method and system thereof based on imaging spectrometer - Google Patents

Abstract

The invention discloses a sulfur fumigated traditional Chinese medicine discriminating method and a system thereof based on an imaging spectrometer, and relates to the technical field of remote sensing. The method comprises the steps that S1: a hyperspectral image of a traditional Chinese medicine to be discriminated within the preset wave band is acquired, and the spectral index characteristics of the traditional Chinese medicine to be discriminated are calculated according to the hyperspectral image of the traditional Chinese medicine to be discriminated within the preset wave band; S2: and the traditional Chinese medicine to be discriminated is classified according to the hyperspectral image of the traditional Chinese medicine to be discriminated within the preset wave band and the spectral index characteristics of the traditional Chinese medicine to be discriminated so that whether the traditional Chinese medicine to be discriminated is fumigated by sulfur can be confirmed. The traditional Chinese medicine to be discriminated is classified through the hyperspectral image of the traditional Chinese medicine to be discriminated within the preset wave band and the spectral index characteristics of the traditional Chinese medicine to be discriminated so that discriminating precision and efficiency are enhanced.

Description

Based on Chinese medicine sulfur fumigation discrimination method and the system of imaging spectrometer

Technical field

The present invention relates to remote sensing technology field, particularly a kind of Chinese medicine sulfur fumigation discrimination method based on imaging spectrometer and system.

Background technology

Sulphur smokes the traditional method that Chinese crude drug was once China's processing of crude drugs, is mainly used in insect protected and protection against the tide, is convenient to the storage of medicinal material, reduces medicinal material loss, also can increase color and luster.But owing to can produce sulphuric dioxide after sulfur burning, as easy as rolling off a logly remain in the middle of medicinal material, long-term taking can injure alimentary canal and respiratory tract, and the internal organs such as liver, kidney.Therefore, state food Drug Administration management board just specially hereafter, prohibitted the use sulphur to smoke Chinese crude drug in 2004.

At present still made in this way by many pharmacists are illegal, traditional discrimination method be make to smell with one's nose, naked eyes distinguish and taste.But due to some Chinese crude drugs inherently scent of or is poisonous itself, so this discrimination method also exists significant limitation, the precision of discriminating and efficiency are very low.

Summary of the invention

In order to improve precision and the efficiency of TCD identificafion, the invention provides a kind of Chinese medicine sulfur fumigation discrimination method based on imaging spectrometer, described method comprises:

S1: obtain the high spectrum image of Chinese medicine to be identified at default wave band, and according to the high spectrum image calculating spectrum index feature to described to be identified Chinese medicine of described Chinese medicine to be identified at default wave band;

S2: classify to described Chinese medicine to be identified at the high spectrum image of default wave band and the spectrum index feature of described Chinese medicine to be identified according to described Chinese medicine to be identified, to determine that whether described Chinese medicine to be identified is through sulfur fumigation.

Alternatively, before step S1, described method also comprises:

S0: obtain the high spectrum image of sample Chinese medicine at default wave band, and according to the high spectrum image calculating spectrum index feature to described sample Chinese medicine of described sample Chinese medicine at default wave band;

Correspondingly, step S2 specifically comprises:

At the high spectrum image of default wave band and the spectrum index feature of described sample Chinese medicine, described Chinese medicine to be identified is classified at the spectrum index feature of the high spectrum image of default wave band, described Chinese medicine to be identified, described sample Chinese medicine according to described Chinese medicine to be identified, to determine that whether described Chinese medicine to be identified is through sulfur fumigation.

Alternatively, in step S2, by following formula, described Chinese medicine to be identified is classified,

$f\left(x_j\right)=\mathrm{sgn}(\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{y}_i{\mathrm{\α}}_{i}K(x_i,x_j)+b),$

Wherein, K (x _i, x _j)=μ K ₁(x _i, x _j)+(1-μ) K ₂(x ' _i, x ' _j); γ be greater than 0 constant; J is the pixel corresponding with Chinese medicine to be identified; x _jfor the Chinese medicine to be identified pixel value that pixel j is corresponding on the high spectrum image of default wave band; X ' _jfor the spectrum index feature that pixel j is corresponding; α _iand b is the parameter of being tried to achieve at the high spectrum image of default wave band and the spectrum index Feature-solving optimization problem of sample Chinese medicine by sample Chinese medicine; μ is the constant in [0,1] interval; I is the pixel that sample Chinese medicine is corresponding; N is the pixel sum that sample Chinese medicine is corresponding; x _ifor the sample Chinese medicine pixel value that pixel i is corresponding on the high spectrum image of default wave band; X ' _ifor the spectrum index feature that pixel i is corresponding; y _iat the Chinese medicine that pixel i is through sulfur fumigation, its value is-1; When pixel i is the Chinese medicine without sulfur fumigation, its value is 1; As f (x _j) <0 time, then differentiate that pixel j is the Chinese medicine through sulfur fumigation; As f (x _j) >0 time, then differentiate that pixel j is the Chinese medicine without sulfur fumigation.

Alternatively, before step S0, described method also comprises:

Obtain the high spectrum image of described sample Chinese medicine at all wave bands, calculate the spatial autocorrelation indicators that the high spectrum image of each wave band is corresponding, using described spatial autocorrelation indicators higher than the wave band of predetermined threshold value as described default wave band.

Alternatively, described spectrum index is characterized as the spectrum index feature to Chinese traditional medicine Physicochemical sense parameters.

The invention also discloses a kind of Chinese medicine sulfur fumigation identification system based on imaging spectrometer, described system comprises:

Data capture unit to be identified, for obtaining the high spectrum image of Chinese medicine to be identified at default wave band, and according to the high spectrum image calculating spectrum index feature to described to be identified Chinese medicine of described Chinese medicine to be identified at default wave band;

Classification of TCD unit to be identified, for classifying to described Chinese medicine to be identified at the high spectrum image of default wave band and the spectrum index feature of described Chinese medicine to be identified according to described Chinese medicine to be identified, to determine that whether described Chinese medicine to be identified is through sulfur fumigation.

Alternatively, described system also comprises:

Sample data acquiring unit, for obtaining the high spectrum image of sample Chinese medicine at default wave band, and according to the high spectrum image calculating spectrum index feature to described sample Chinese medicine of described sample Chinese medicine at default wave band;

Correspondingly, described classification of TCD unit to be identified, also for classifying to described Chinese medicine to be identified at the high spectrum image of default wave band and the spectrum index feature of described sample Chinese medicine at the spectrum index feature of the high spectrum image of default wave band, described Chinese medicine to be identified, described sample Chinese medicine according to described Chinese medicine to be identified, to determine that whether described Chinese medicine to be identified is through sulfur fumigation.

Alternatively, described classification of TCD unit to be identified, also for being classified to described Chinese medicine to be identified by following formula,

$f\left(x_j\right)=\mathrm{sgn}(\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{y}_i{\mathrm{\&alpha;}}_{i}K(x_i,x_j)+b),$

Wherein, K (x _i, x _j)=μ K ₁(x _i, x _j)+(1-μ) K ₂(x ' _i, x ' _j); γ be greater than 0 constant; J is the pixel corresponding with Chinese medicine to be identified; x _jfor the Chinese medicine to be identified pixel value that pixel j is corresponding on the high spectrum image of default wave band; X ' _jfor the spectrum index feature that pixel j is corresponding; α _iand b is the parameter of being tried to achieve at the high spectrum image of default wave band and the spectrum index Feature-solving optimization problem of sample Chinese medicine by sample Chinese medicine; μ is the constant in [0,1] interval; I is the pixel that sample Chinese medicine is corresponding; N is the pixel sum that sample Chinese medicine is corresponding; x _ifor the sample Chinese medicine pixel value that pixel i is corresponding on the high spectrum image of default wave band; X ' _ifor the spectrum index feature that pixel i is corresponding; y _iat the Chinese medicine that pixel i is through sulfur fumigation, its value is-1; When pixel i is the Chinese medicine without sulfur fumigation, its value is 1; As f (x _j) <0 time, then differentiate that pixel j is the Chinese medicine through sulfur fumigation; As f (x _j) >0 time, then differentiate that pixel j is the Chinese medicine without sulfur fumigation.

Alternatively, described system also comprises:

Band selection unit, for obtaining the high spectrum image of described sample Chinese medicine at all wave bands, calculates the spatial autocorrelation indicators that the high spectrum image of each wave band is corresponding, using described spatial autocorrelation indicators higher than the wave band of predetermined threshold value as described default wave band.

Alternatively, described spectrum index is characterized as the spectrum index feature to Chinese traditional medicine Physicochemical sense parameters.

The present invention is classified to described Chinese medicine to be identified at the high spectrum image of default wave band and the spectrum index feature of Chinese medicine to be identified by Chinese medicine to be identified, thus improves precision and the efficiency of discriminating.

Accompanying drawing explanation

Fig. 1 is the process flow diagram of the Chinese medicine sulfur fumigation discrimination method based on imaging spectrometer of one embodiment of the present invention;

Fig. 2 is the structured flowchart of the Chinese medicine sulfur fumigation identification system based on imaging spectrometer of one embodiment of the present invention.

Embodiment

Below in conjunction with drawings and Examples, the specific embodiment of the present invention is described in further detail.Following examples for illustration of the present invention, but are not used for limiting the scope of the invention.

Fig. 1 is the process flow diagram of the Chinese medicine sulfur fumigation discrimination method based on imaging spectrometer of one embodiment of the present invention; With reference to Fig. 1, described method comprises:

S1: obtain the high spectrum image of Chinese medicine to be identified at default wave band, and according to the high spectrum image calculating spectrum index feature to described to be identified Chinese medicine of described Chinese medicine to be identified at default wave band;

S2: classify to described Chinese medicine to be identified at the high spectrum image of default wave band and the spectrum index feature of described Chinese medicine to be identified according to described Chinese medicine to be identified, to determine that whether described Chinese medicine to be identified is through sulfur fumigation.

Present embodiment is classified to described Chinese medicine to be identified at the high spectrum image of default wave band and the spectrum index feature of Chinese medicine to be identified by Chinese medicine to be identified, thus improves precision and the efficiency of discriminating.

In specific implementation, classify to described Chinese medicine to be identified for the ease of realizing, alternatively, before step S1, described method also comprises:

S0: obtain the high spectrum image of sample Chinese medicine at default wave band, and according to the high spectrum image calculating spectrum index feature to described sample Chinese medicine of described sample Chinese medicine at default wave band;

When obtaining high spectrum image, Halogen lamp LED can be used to do light source, to Chinese medicine Uniform Illumination to be identified, the imaging spectral instrument system centering medicinal material built is adopted to carry out light spectrum image-forming, known sulfur fumigation and the known Chinese medicine of the same race without fumigating are placed in same visual field simultaneously, certainly, also by other obtain manner, present embodiment is not limited this.

Correspondingly, step S2 specifically comprises:

At the high spectrum image of default wave band and the spectrum index feature of described sample Chinese medicine, described Chinese medicine to be identified is classified at the spectrum index feature of the high spectrum image of default wave band, described Chinese medicine to be identified, described sample Chinese medicine according to described Chinese medicine to be identified, to determine that whether described Chinese medicine to be identified is through sulfur fumigation.

For improving the efficiency of classifying to described Chinese medicine to be identified further, alternatively, in step S2, by following formula, described Chinese medicine to be identified is classified,

$f\left(x_j\right)=\mathrm{sgn}(\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{y}_i{\mathrm{\&alpha;}}_{i}K(x_i,x_j)+b),$

Wherein, K (x _i, x _j)=μ K ₁(x _i, x _j)+(1-μ) K ₂(x ' _i, x ' _j); γ be greater than 0 constant; J is the pixel corresponding with Chinese medicine to be identified; x _jfor the Chinese medicine to be identified pixel value that pixel j is corresponding on the high spectrum image of default wave band; X ' _jfor the spectrum index feature that pixel j is corresponding; α _iand b is the parameter of being tried to achieve at the high spectrum image of default wave band and the spectrum index Feature-solving optimization problem of sample Chinese medicine by sample Chinese medicine; μ is the constant in [0,1] interval; I is the pixel that sample Chinese medicine is corresponding; N is the pixel sum that sample Chinese medicine is corresponding; x _ifor the sample Chinese medicine pixel value that pixel i is corresponding on the high spectrum image of default wave band; X ' _ifor the spectrum index feature that pixel i is corresponding; y _iat the Chinese medicine that pixel i is through sulfur fumigation, its value is-1; When pixel i is the Chinese medicine without sulfur fumigation, its value is 1; As f (x _j) <0 time, then differentiate that pixel j is the Chinese medicine through sulfur fumigation; As f (x _j) >0 time, then differentiate that pixel j is the Chinese medicine without sulfur fumigation.

Because the wave band number of high spectrum image is many, wavelength cover is wide, but the picture quality of subband is poor, serious mistake can be brought to divide when classifying, for avoiding this problem, need first to carry out band selection, namely Geary ' s c value is calculated to the high-spectral data obtained, remove the wave band that Geary ' s c value is larger, the wave band that retention is less, and then remove the wave band that between poor image quality or wave band, correlativity is strong, reduce the redundancy of high-spectral data, play the effect of Data Dimensionality Reduction, alternatively, before step S0, described method also comprises:

Obtain the high spectrum image of described sample Chinese medicine at all wave bands, calculate the spatial autocorrelation indicators that the high spectrum image of each wave band is corresponding, using described spatial autocorrelation indicators higher than the wave band of predetermined threshold value as described default wave band.

In present embodiment, described spatial autocorrelation indicators adopts Geary ' s c value to weigh, and computing formula is as follows:

$c=\frac{\underset{m}{\mathrm{\&Sigma;}}\underset{n}{\mathrm{\&Sigma;}}{w}_{\mathrm{mn}}{c}_{\mathrm{mn}}}{2\underset{m}{\mathrm{\&Sigma;}}\underset{n}{\mathrm{\&Sigma;}}{w}_{\mathrm{mn}}{\mathrm{\&sigma;}}^2}$

Wherein, c is Geary ' s c value, m and n is any two pixels on same high spectrum image, c _mn=(z _m-z _n) ², z _mfor the gray-scale value of pixel m, z _nfor the gray-scale value of pixel n, w _mnfor space weight variable, when the spacing of pixel m and pixel n is preset value, value is 1, otherwise value is 0, σ ²for the variance yields of gray scale on this high spectrum image.

Wherein, σ ²computing formula be:

${\mathrm{\&sigma;}}^2=\frac{1}{n-1}\underset{m}{\mathrm{\&Sigma;}}{(z_m-\stackrel{\&OverBar;}{z})}^2$

Wherein, n is the sum of pixel on this high spectrum image, for the mean value of gray scale on this high spectrum image.

It should be noted that, Geary ' s c value can reflect picture quality (Geary ' s c value is less, picture quality is higher; Otherwise picture quality is lower), and to wave band decorrelation.

Differentiate ground accuracy to improve, alternatively, described spectrum index is characterized as the spectrum index feature to Chinese traditional medicine Physicochemical sense parameters, and through repeatedly verifying, described spectrum index feature can choose 50 kinds, specifically can refer to following table:

Fig. 2 is the structured flowchart of the Chinese medicine sulfur fumigation identification system based on imaging spectrometer of one embodiment of the present invention; With reference to Fig. 2, described system comprises:

Data capture unit to be identified, for obtaining the high spectrum image of Chinese medicine to be identified at default wave band, and according to the high spectrum image calculating spectrum index feature to described to be identified Chinese medicine of described Chinese medicine to be identified at default wave band;

Classification of TCD unit to be identified, for classifying to described Chinese medicine to be identified at the high spectrum image of default wave band and the spectrum index feature of described Chinese medicine to be identified according to described Chinese medicine to be identified, to determine that whether described Chinese medicine to be identified is through sulfur fumigation.

Alternatively, described system also comprises:

Sample data acquiring unit, for obtaining the high spectrum image of sample Chinese medicine at default wave band, and according to the high spectrum image calculating spectrum index feature to described sample Chinese medicine of described sample Chinese medicine at default wave band;

Correspondingly, described classification of TCD unit to be identified, also for classifying to described Chinese medicine to be identified at the high spectrum image of default wave band and the spectrum index feature of described sample Chinese medicine at the spectrum index feature of the high spectrum image of default wave band, described Chinese medicine to be identified, described sample Chinese medicine according to described Chinese medicine to be identified, to determine that whether described Chinese medicine to be identified is through sulfur fumigation.

Alternatively, described classification of TCD unit to be identified, also for being classified to described Chinese medicine to be identified by following formula,

$f\left(x_j\right)=\mathrm{sgn}(\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{y}_i{\mathrm{\&alpha;}}_{i}K(x_i,x_j)+b),$

Wherein, K (x _i, x _j)=μ K ₁(x _i, x _j)+(1-μ) K ₂(x ' _i, x ' _j); γ be greater than 0 constant; J is the pixel corresponding with Chinese medicine to be identified; x _jfor the Chinese medicine to be identified pixel value that pixel j is corresponding on the high spectrum image of default wave band; X ' _jfor the spectrum index feature that pixel j is corresponding; α _iand b is the parameter of being tried to achieve at the high spectrum image of default wave band and the spectrum index Feature-solving optimization problem of sample Chinese medicine by sample Chinese medicine; μ is the constant in [0,1] interval; I is the pixel that sample Chinese medicine is corresponding; N is the pixel sum that sample Chinese medicine is corresponding; x _ifor the sample Chinese medicine pixel value that pixel i is corresponding on the high spectrum image of default wave band; X ' _ifor the spectrum index feature that pixel i is corresponding; y _iat the Chinese medicine that pixel i is through sulfur fumigation, its value is-1; When pixel i is the Chinese medicine without sulfur fumigation, its value is 1; As f (x _j) <0 time, then differentiate that pixel j is the Chinese medicine through sulfur fumigation; As f (x _j) >0 time, then differentiate that pixel j is the Chinese medicine without sulfur fumigation.

Alternatively, described system also comprises:

Band selection unit, for obtaining the high spectrum image of described sample Chinese medicine at all wave bands, calculates the spatial autocorrelation indicators that the high spectrum image of each wave band is corresponding, using described spatial autocorrelation indicators higher than the wave band of predetermined threshold value as described default wave band.

Alternatively, described spectrum index is characterized as the spectrum index feature to Chinese traditional medicine Physicochemical sense parameters.

Above embodiment is only for illustration of the present invention; and be not limitation of the present invention; the those of ordinary skill of relevant technical field; without departing from the spirit and scope of the present invention; can also make a variety of changes and modification; therefore all equivalent technical schemes also belong to category of the present invention, and scope of patent protection of the present invention should be defined by the claims.

Claims (10)

1., based on a Chinese medicine sulfur fumigation discrimination method for imaging spectrometer, it is characterized in that, described method comprises:

S1: obtain the high spectrum image of Chinese medicine to be identified at default wave band, and according to the high spectrum image calculating spectrum index feature to described to be identified Chinese medicine of described Chinese medicine to be identified at default wave band;

S2: classify to described Chinese medicine to be identified at the high spectrum image of default wave band and the spectrum index feature of described Chinese medicine to be identified according to described Chinese medicine to be identified, to determine that whether described Chinese medicine to be identified is through sulfur fumigation.

2. the method for claim 1, is characterized in that, before step S1, described method also comprises:

S0: obtain the high spectrum image of sample Chinese medicine at default wave band, and according to the high spectrum image calculating spectrum index feature to described sample Chinese medicine of described sample Chinese medicine at default wave band;

Correspondingly, step S2 specifically comprises:

At the high spectrum image of default wave band and the spectrum index feature of described sample Chinese medicine, described Chinese medicine to be identified is classified at the spectrum index feature of the high spectrum image of default wave band, described Chinese medicine to be identified, described sample Chinese medicine according to described Chinese medicine to be identified, to determine that whether described Chinese medicine to be identified is through sulfur fumigation.

3. method as claimed in claim 2, be is characterized in that, in step S2, classified by following formula to described Chinese medicine to be identified,

$f\left(x_j\right)=\mathrm{sgn}(\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{y}_i{\mathrm{\&alpha;}}_{i}K(x_i,x_j)+b),$

Wherein, K (x _i, x _j)=μ K ₁(x _i, x _j)+(1-μ) K ₂(x ' _i, x ' _j); γ be greater than 0 constant; J is the pixel corresponding with Chinese medicine to be identified; x _jfor the Chinese medicine to be identified pixel value that pixel j is corresponding on the high spectrum image of default wave band; X ' _jfor the spectrum index feature that pixel j is corresponding; α _iand b is the parameter of being tried to achieve at the high spectrum image of default wave band and the spectrum index Feature-solving optimization problem of sample Chinese medicine by sample Chinese medicine; μ is the constant in [0,1] interval; I is the pixel that sample Chinese medicine is corresponding; N is the pixel sum that sample Chinese medicine is corresponding; x _ifor the sample Chinese medicine pixel value that pixel i is corresponding on the high spectrum image of default wave band; X ' _ifor the spectrum index feature that pixel i is corresponding; y _iat the Chinese medicine that pixel i is through sulfur fumigation, its value is-1; When pixel i is the Chinese medicine without sulfur fumigation, its value is 1; As f (x _j) <0 time, then differentiate that pixel j is the Chinese medicine through sulfur fumigation; As f (x _j) >0 time, then differentiate that pixel j is the Chinese medicine without sulfur fumigation.

4. method as claimed in claim 2, it is characterized in that, before step S0, described method also comprises:

Obtain the high spectrum image of described sample Chinese medicine at all wave bands, calculate the spatial autocorrelation indicators that the high spectrum image of each wave band is corresponding, using described spatial autocorrelation indicators higher than the wave band of predetermined threshold value as described default wave band.

5. the method according to any one of Claims 1 to 4, is characterized in that, described spectrum index is characterized as the spectrum index feature to Chinese traditional medicine Physicochemical sense parameters.

6., based on a Chinese medicine sulfur fumigation identification system for imaging spectrometer, it is characterized in that, described system comprises:

Data capture unit to be identified, for obtaining the high spectrum image of Chinese medicine to be identified at default wave band, and according to the high spectrum image calculating spectrum index feature to described to be identified Chinese medicine of described Chinese medicine to be identified at default wave band;

Classification of TCD unit to be identified, for classifying to described Chinese medicine to be identified at the high spectrum image of default wave band and the spectrum index feature of described Chinese medicine to be identified according to described Chinese medicine to be identified, to determine that whether described Chinese medicine to be identified is through sulfur fumigation.

7. system as claimed in claim 6, it is characterized in that, described system also comprises:

Sample data acquiring unit, for obtaining the high spectrum image of sample Chinese medicine at default wave band, and according to the high spectrum image calculating spectrum index feature to described sample Chinese medicine of described sample Chinese medicine at default wave band;

Correspondingly, described classification of TCD unit to be identified, also for classifying to described Chinese medicine to be identified at the high spectrum image of default wave band and the spectrum index feature of described sample Chinese medicine at the spectrum index feature of the high spectrum image of default wave band, described Chinese medicine to be identified, described sample Chinese medicine according to described Chinese medicine to be identified, to determine that whether described Chinese medicine to be identified is through sulfur fumigation.

8. system as claimed in claim 7, is characterized in that, described classification of TCD unit to be identified, also for being classified to described Chinese medicine to be identified by following formula,

$f\left(x_j\right)=\mathrm{sgn}(\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{y}_i{\mathrm{\&alpha;}}_{i}K(x_i,x_j)+b),$

Wherein, K (x _i, x _j)=μ K ₁(x _i, x _j)+(1-μ) K ₂(x ' _i, x ' _j); γ be greater than 0 constant; J is the pixel corresponding with Chinese medicine to be identified; x _jfor the Chinese medicine to be identified pixel value that pixel j is corresponding on the high spectrum image of default wave band; X ' _jfor the spectrum index feature that pixel j is corresponding; α _iand b is the parameter of being tried to achieve at the high spectrum image of default wave band and the spectrum index Feature-solving optimization problem of sample Chinese medicine by sample Chinese medicine; μ is the constant in [0,1] interval; I is the pixel that sample Chinese medicine is corresponding; N is the pixel sum that sample Chinese medicine is corresponding; x _ifor the sample Chinese medicine pixel value that pixel i is corresponding on the high spectrum image of default wave band; X ' _ifor the spectrum index feature that pixel i is corresponding; y _iat the Chinese medicine that pixel i is through sulfur fumigation, its value is-1; When pixel i is the Chinese medicine without sulfur fumigation, its value is 1; As f (x _j) <0 time, then differentiate that pixel j is the Chinese medicine through sulfur fumigation; As f (x _j) >0 time, then differentiate that pixel j is the Chinese medicine without sulfur fumigation.

9. system as claimed in claim 7, it is characterized in that, described system also comprises:

Band selection unit, for obtaining the high spectrum image of described sample Chinese medicine at all wave bands, calculates the spatial autocorrelation indicators that the high spectrum image of each wave band is corresponding, using described spatial autocorrelation indicators higher than the wave band of predetermined threshold value as described default wave band.

10. the system according to any one of claim 6 ~ 9, is characterized in that, described spectrum index is characterized as the spectrum index feature to Chinese traditional medicine Physicochemical sense parameters.