JPH07198602A - Component content estimating method by near infrared spectral analysis - Google Patents

Abstract

PURPOSE:To reduce a measurement error due to temperature in quantitative analysis by a near infrared spectral analysis method. CONSTITUTION:At the time of obtaining absorbance of a near infrared ray and an analytical curve of a component content, if the temperature dependency of a term composed of dominant wavelengthes is positive or negative, a term composed of absorbances of negative or positive inverse characteristic wavelengthes is used to cancel the temperature fluctuation of absorbance in each wavelength. Thus, a temperature error due to temperature in the component content measurement of an unknown sample by near infrared spectral analysis can be reduced so as to improve the measurement accuracy. In the case of measurement by grinding a sample, though heretofore it is necessary to perform measurement after the ground sample is cooled to the atmosphere temperature, the component content can be measured without cooling.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for estimating the amount of components such as proteins contained in foods by near infrared spectroscopy.

[0002]

2. Description of the Related Art Near infrared spectrophotometers are widely used for nondestructively measuring (estimating) the amounts of various components such as proteins contained in grains and the like.

When estimating the content of a component of a sample by a near-infrared spectrophotometer, a sample having a known content of a predetermined component (eg, protein, starch, water, etc.) has a predetermined region in which the component is absorbed. The near-infrared ray having a wavelength is irradiated to obtain the absorbance. Next, this absorbance is secondarily differentiated to calculate the second derivative absorbance, and the second derivative absorbance is used as an explanatory variable and regression analysis is performed by using the sample component content as an objective variable to obtain a calibration curve and the calibration curve. Estimate the component content of the unknown sample by the line.

[0004]

In such an ingredient content estimation method, an error may occur in the estimated value of the ingredient due to a change in the sample temperature due to a change in the outside air temperature, and a solution to this has been desired. .

Therefore, an object of the present invention is to eliminate the influence of temperature, which has been a problem in the prior art, in the method of estimating the content of components by near-infrared spectroscopy, thereby improving the estimation accuracy.

[0006]

According to the present invention, a sample having a known content of a component is irradiated with near-infrared light having a wavelength in a region where a predetermined component is absorbed, the absorbance is measured, and the absorbance is calculated to be an explanatory variable. And the calibration curve is obtained by performing regression analysis with the content of the predetermined component of the sample as the target variable.
In the component content estimation method for estimating the content of a predetermined component of a sample whose component content is unknown by the calibration curve, when the temperature dependence of the term constituted by the dominant wavelength in obtaining the calibration curve is positive or negative, It is characterized by using a term composed of an absorbance of a wavelength having a negative or positive inverse characteristic.

[0007]

[Function] When measuring by near-infrared spectroscopic analysis, the predetermined wavelength of near-infrared ray has positive and negative temperature dependence. Therefore, the use of both wavelengths cancels the temperature fluctuation of the absorbance at each wavelength. To be done. Therefore, it is possible to reduce the measurement error due to the temperature when measuring the components contained in the unknown sample by the near infrared spectroscopic analysis, and to improve the measurement accuracy. Further, in the case where the sample is crushed at the time of measurement, it is necessary to measure the sample after crushing after cooling it to the ambient temperature, but the present invention has an advantage that it can be measured without cooling.

[0008]

EXAMPLES Examples of the present invention will be described below.

The present invention is an improvement of the method for obtaining a calibration curve required when measuring the content of components contained in an unknown sample by a near infrared spectrophotometer, and the calibration curve is obtained by the following procedure.

Taking a protein as an example, first, a sample having a known protein content is attached to a known near-infrared spectrophotometer, and this is irradiated with near-infrared light to detect the absorbance for each wavelength of near-infrared light. . These treatments are performed on N samples of known protein content.

Next, each absorbance is secondarily differentiated by a computer incorporated in the near-infrared spectrophotometer to calculate a second derivative absorbance, and the second derivative absorbance is 2180 nm as the first wavelength and the second wavelength. Is used as an explanatory variable and the protein content of the sample is used as an objective variable, and multiple regression analysis is performed to obtain a calibration curve. The third
The wavelength is 1610nm, 1492nm or 195nm
A wavelength of 2 nm can be used for the reasons described below.

The reason why the second-order differential absorbances of the above three wavelengths are used is as follows.

As shown in FIG. 1, the absorbance at a wavelength of 2180 nm, which has a strong correlation with the protein content, has a correlation with temperature change that is almost uncorrelated, and even if there is a temperature change,
It has almost no effect on the second derivative absorbance. However, the absorbance at a wavelength of 1224 nm has an inverse correlation (negative correlation) with the temperature change, and the second derivative absorbance decreases as the temperature rises, and the second derivative absorbance increases as the temperature falls.
On the other hand, the wavelength is 1610 nm, 1492 nm, or 19
The secondary differential absorbance at 52 nm (see FIG. 1) has an inverse correlation (positive correlation) with the temperature change. When the temperature rises, the secondary differential absorbance increases, and when the temperature decreases, the secondary differential absorbance decreases. Become.

Therefore, in obtaining the calibration curve as described above, 2 at the wavelength of 2180 nm, which has no temperature dependence, is obtained.
The second derivative is the first term, and the temperature dependence is negative 1224
The second-order differential absorbance at the wavelength of nm is used as the second term, and the temperature dependence is positive at 1610 nm, 1492 nm, or 1952 n.
The second-order differential absorbance at the wavelength of m was defined as the third term.

Here, a sample having a known protein content was estimated from the calibration curve obtained as described above, and the relationship between the actual measured value of protein and the estimated value is shown in the scatter diagram of FIG. On the other hand, for comparison with this, there is no temperature dependence 21
The second derivative absorbance at a wavelength of 80 nm is set to the first term, and the second derivative absorbance at a wavelength of 1224 nm, which has a negative temperature dependence, is set to the second term, and a calibration curve is obtained only from these, and the same measurement as above is performed. Then, the scatter diagram as shown in FIG. 3 is obtained.

Comparing the two scatter plots thus obtained, the effect of adding the second-order differential absorbance at a wavelength of 1610 nm with positive temperature dependence as a correction term is clear. This is because the temperature variation of the secondary differential absorbance at the wavelength 1224 nm having a negative temperature dependency is canceled out by the temperature variation of the secondary differential absorbance at the wavelength 1610 nm having a positive temperature dependency.

As is clear from the above description, by using the calibration curve obtained as described above, it is possible to reduce the measurement error due to the temperature when measuring an unknown sample by the near-infrared spectrophotometer, and to improve the measurement accuracy. It is possible to improve. Further, in the case where the sample is crushed, it is necessary to measure the crushed sample after cooling it to the ambient temperature, but it can be measured without cooling.

[0018]

As described above, according to the present invention, when the calibration curve is prepared, the predetermined wavelengths of the near infrared rays are those having a positive temperature dependence and those having a negative temperature dependence. The temperature fluctuation of the absorbance at each wavelength is canceled. Therefore, it is possible to reduce the measurement error due to the temperature when measuring the components of the unknown sample by the near infrared spectroscopic analysis, and improve the measurement accuracy. Further, in the case where the sample is crushed, it is necessary to measure the crushed sample after cooling it to the ambient temperature, but it is convenient to measure without cooling.

[Brief description of drawings]

FIG. 1 is a diagram in which the correlation between absorbance of near infrared rays and temperature is plotted for each wavelength.

FIG. 2 is a scatter diagram showing a relationship between an actually measured value of protein and its estimated value, in which a sample of known protein is estimated by the calibration curve obtained as described above.

FIG. 3 is a scatter diagram for comparison with FIG.

Claims (1)

[Claims] 1. A sample having a known content of a component is irradiated with near-infrared light having a wavelength in a region where the predetermined component is absorbed, the absorbance is measured, the absorbance is calculated and used as an explanatory variable, and the predetermined component of the sample is calculated. In the component content estimation method of estimating the content of the predetermined component of the sample whose component content is unknown by using the calibration curve to obtain a calibration curve by performing regression analysis using the content of When the temperature dependence of the term consisting of wavelengths is positive or negative, the term consisting of the absorbance of the wavelength of the negative or positive inverse characteristic is used to estimate the component content by near-infrared spectroscopy. Law.