KR102141154B1 - Non-destructive Quality Evaluation Method of Sweet Potatoes Using Near Infrared and Chemometrics - Google Patents

Abstract

The present invention relates to a quality evaluation method of sweet potatoes, and more specifically, to a non-destructive quality evaluation method of sweet potatoes using near-infrared rays and chemometrics. The quality evaluation method of sweet potatoes, according to the present invention, is a non-destructive analysis method in which a raw sweet potato can be used as a sample by means of near-infrared rays. Measurement of moisture, sugar, and white spot contents, which are indicators of sweet potatoes immediately after being harvested or before being processed, and states of sweet potatoes according to storage conditions can be simultaneously analyzed through a single measurement by means of near-infrared rays. When the method of the present invention is used, it is possible to classify sweet potatoes according to a processing use before the process of the sweet potatoes, and to simultaneously analyze various factors affecting quality of the sweet potatoes, such as moisture, sugar, white spot contents of sweet potatoes and states of sweet potatoes according to storage conditions, at a speed of processing 20 sweet potatoes per minute.

Description

Non-destructive Quality Evaluation Method of Sweet Potatoes Using Near Infrared and Chemometrics}

The present invention relates to a method for quality evaluation of sweet potatoes, and more particularly, to a method for non-destructive quality evaluation of sweet potatoes using near infrared rays and chemometrics.

The country of origin of sweet potatoes is known as Central America and South America, and the total worldwide production of sweet potatoes is 107,639 thousand tons as of 2010, mainly producing about 82% in Asia. The main native regions of Korea are Jeonnam, Jeonbuk, and Gyeongbuk, and are typically known as the sweet potato production area of Haenam, Jeollanam-do. In accordance with the recent well-being era, it is recognized as a non-polluting health food for consumers, and it is a food that is increasing in consumption as a meal replacement and diet food. The main ingredient of sweet potato is sugar and contains sweet ingredients such as sucrose, glucose, and fructose. It is a high-calorie food with twice as much calories. In addition, since it contains a large amount of dietary fiber and yalaffin and seraphin, it is effective in preventing constipation, and it is reported that it contains a large amount of inorganic components such as vitamin C and potassium. In addition, vitamins and anthocyanins, which are antioxidants contained in a large amount of purple sweet potato, have an excellent effect on skin beauty and anti-aging, and contain a large amount of potassium as an alkaline food good for preventing adult diseases. Potassium acts to excrete sodium into the body along with urine, preventing adult diseases such as hypertension and preventing stroke.

Recently, as the yield of sweet potato has increased, utilization as a processed food has increased, but the standard for the internal quality level of sweet potato has not been established, and thus the quality has not been managed. When preparing sweets or dried fish, if sweet potato content is high after cooking, it must be discarded, which has the disadvantage of high cooking and disposal costs. In addition, the internal quality differs depending on the sweet potato storage conditions, but it is difficult to distinguish in appearance.

Accordingly, the present inventors tried to develop a method for analyzing sweet potato quality using a non-destructive technique that can be analyzed at a processing site in order to solve this disadvantage, and as a result, near-infrared (near-infrared; abbreviated as'NIR' for short). When measuring the transmission spectrum of sweet potato by irradiating, and substituting the measured spectral values into the calibration curves of moisture, sugar, and heavy protein content of the sweet potato having a specific formula, moisture, sugar, and heavy protein content of the target sweet potato to be measured It was confirmed that can be effectively derived. In addition, the present invention was completed by confirming that it was possible to classify sweet potatoes stored under unsuitable conditions and sweet potatoes stored under suitable conditions with the measured transmission spectrum of sweet potatoes.

Korean Patent Publication No. 10-2012-0063679 Korean Registered Patent No. 10-1027434

Accordingly, an object of the present invention is to provide a method for evaluating the quality of non-destructive sweet potatoes.

In order to achieve the object of the present invention as described above,

The present invention comprises the steps of: a) measuring a spectrum value by irradiating near-infrared light on a sweet potato sample; And b) calculating moisture, sugar, and heavy white contents of the sweet potato using the measured spectral values.

In one embodiment of the present invention, the irradiation wavelength of near infrared rays for calculating the moisture content of the sweet potato may be 500 to 600 nm.

In an embodiment of the present invention, the irradiation wavelength of near infrared rays for calculating the sugar content of the sweet potato may be 800 to 1,000 nm.

In one embodiment of the present invention, the irradiation wavelength of the near-infrared ray for calculating the white matter content of the sweet potato may be 870 to 1,000 nm.

In one embodiment of the present invention, in step b), the measured spectral values can be substituted into the calibration curves of the moisture, sugar, and heavy protein contents of sweet potato to calculate moisture, sugar, and heavy protein contents of the sweet potato sample, respectively.

In one embodiment of the present invention, the moisture content calibration curve of the sweet potato is represented by Equation 1 below, and when x is the spectral value of the sweet potato sample in Equation 1, the moisture content measurement value y(%) of the sweet potato sample is calculated. Can be.

[Equation 1]

y = 0.5944 x + 22.36

In one embodiment of the present invention, the sugar content calibration curve of the sweet potato is represented by Equation 2 below, and when x′ in Equation 2 is the spectral value of the sweet potato sample, the measured value of the sugar content of the sweet potato sample y′ (%) Can be calculated.

[Equation 2]

y′= 0.5792 x′+ 0.9183

In one embodiment of the present invention, the calibration curve of the sweet potato content of the sweet potato is represented by the following Equation 3, and when x″ in Equation 3 is the spectral value of the sweet potato sample, the measured value of the heavy protein content of the sweet potato sample y″ (%) Can be calculated.

[Equation 3]

y″= 0.7418 x″+ 0.05369

In one embodiment of the present invention, the method for evaluating the quality of sweet potatoes may further include a step of determining whether the storage conditions are suitable.

In one embodiment of the present invention, the irradiation wavelength of the near-infrared for determining the suitability of the storage conditions may be 0 to 2000 nm.

In one embodiment of the present invention, the suitability of the storage condition may be determined as a suitability class when the value calculated by multiplying the near-infrared spectral value and the eigenvalue is greater than -0.15 and a non-conformance class when less than -0.15.

At this time, the near infrared spectral value is a spectral value measured by irradiating a near-infrared light on a sweet potato sample, and the eigenvalue is the eigenvalue having the greatest dispersion degree obtained by putting Equation 4 below into a matlab program.

[Equation 4]

In the above equation, W is a transformation matrix, S _B is a variance matrix between classes, and S _w is a variance matrix within a class.

The method for evaluating sweet potato quality of the present invention is a non-destructive analysis method that can use raw sweet potato itself as a sample using near infrared rays, and analyzes moisture, sugar, and heavy protein contents as a quality indicator of sweet potato immediately before harvesting or before processing using near infrared rays. Since it can be analyzed simultaneously with this single measurement, when using the method of the present invention, it is possible to classify according to the processing purpose before processing sweet potatoes, and at 20 processing speeds per minute, the moisture, sugar content, heavy protein content and storage conditions It has the advantage of simultaneously analyzing various factors affecting sweet potato quality, such as sweet potato status.

1 shows an NIR spectrum for a sweet potato sample. Figure 1A is a NIR spectrum of 0-2000 nm, Figure 1B is a NIR spectrum of 200-1000 nm, Figure 1C is a 200-1000 nm NIR spectrum primary differential spectrum, Figure 1D is a 200-1000 nm NIR spectrum secondary differential Spectrum.
Figure 2 shows the method used when measuring the content of heavy white. FIG. 2A is an image of a cross-section taken with a camera after baking at 180°C for 90 minutes, cutting a sample of sweet potato at the equator. FIG. 2B is an image obtained by converting an original image into black and white, and FIG. 2C is histogram data of FIG. 2B.
Figure 3 is a diagram showing the results of Partial Least Squares (PLS) Regression analysis optimized for moisture (%) of sweet potato samples.
4 is a diagram showing the results of a Partial Least Squares (PLS) Regression analysis optimized for the sweetness (g/100 g) of a sweet potato sample.
FIG. 5 is a diagram showing the results of a Partial Least Squares (PLS) Regression analysis optimized for the white protein of a sweet potato sample.
FIG. 6 is a diagram showing the results of Linear Discriminant Analysis (LDA) analysis of sweet potatoes (Abnormal) stored under unsuitable conditions and sweet potatoes (Normal) stored under suitable conditions. LDA 1 in FIG. 6A is a value obtained by multiplying the first eigenvector by each NIR spectrum, and LDA 2 is a value obtained by multiplying the second eigenvector by each NIR spectrum. 6B shows the LDA 1 value as a histogram, and it can be confirmed that the sweet potato (Abnormal) stored under unsuitable conditions and the sweet potato (Normal) stored under suitable conditions are classified as LDA 1 values.
FIG. 7A is a photograph showing a cross-section after roasting of sweet potato without a white phenomenon, and FIG. 7B is a photograph showing a cross-section after roasting of sweet potatoes having a heavy white phenomenon.

The present invention comprises the steps of: a) measuring a spectrum value by irradiating near-infrared light on a sweet potato sample; And b) calculating moisture, sugar and heavy white contents of sweet potato using the measured spectral values.

The “moisture content of sweet potato” referred to in the present invention is a characteristic of sweet potato that is defined as a percentage of the weight occupied by moisture relative to the total weight of the sweet potato sample. The moisture content of the sweet potato is affected by the storage condition, and has a great influence on the sweet potato quality.

“Sugar content of sweet potato” referred to in the present invention is a characteristic of sweet potato defined as a sugar content for 100 g of sweet potato samples. The sugar content can give a lot of nutrition to the taste after sweet potato processing.

"Medium content of sweet potato" referred to in the present invention means the area of the portion where the saccharification is not achieved when the sweet potato is baked at 180°C for 90 minutes. When the content of heavy white rice is high, it is not good in appearance and taste is also poor. Therefore, the content of heavy white rice is closely related to the quality of sweet potatoes, and it also affects the preferences of sellers and consumers.

Step a) of the present invention is a step of measuring spectral values by irradiating near-infrared rays to sweet potatoes, which are samples to be analyzed.

In the present invention, the near-infrared spectrum can be obtained by scanning in all manners conventionally used in the art in the state of sweet potatoes, and preferably by a transmission method.

In one embodiment of the present invention, the near-infrared wavelength irradiated in step a) is a wavelength in the range of 500 to 600 nm when the object to be calculated is'moisture content of sweet potato', and the object to be calculated is'sugar content of sweet potato' In the case of', it is a wavelength in the range of 800 to 1,000 nm, and when the object to be calculated is'the white matter content of sweet potato', it may be a wavelength in the range of 870 to 1,000 nm.

Step b) of the present invention is a step of calculating the moisture, sugar, and heavy white content of the sweet potato sample by substituting the spectral values measured through the step a) into the calibration curves of the sweet water, sugar, and heavy white content, respectively.

In step b) of the present invention, the calibration curve of the moisture content of the sweet potato is optimized by comparing the value derived by applying the second derivative to the NIR spectrum of the sweet potato sample, and comparing the sweet potato sample with the moisture content measured by the dry weight loss method of the food industry. Can be derived. In the following example of the present invention, the value derived from the NIR spectrum through the multivariate regression analysis on the same sample based on the moisture content measured by the moisture analysis method for the sweet potato sample is the value measured by the moisture measurement method. The moisture content calibration curve optimized for sweet potatoes was derived by optimizing in the direction of minimizing the error to be close. The moisture analysis method for measuring the moisture content of the standard sweet potato sample as the reference value is not limited to the above-listed methods, and a moisture analysis method known in the industry can be used without limitation.

In one embodiment of the present invention, the moisture content calibration curve of the sweet potato is represented by Equation 1 below, and in Equation 1, when x is a spectral value of the sweet potato sample, the moisture content measurement value y(%) of the sweet potato sample is calculated. Can.

[Equation 1]

y = 0.5944 x + 22.36

In step b) of the present invention, the calibration curve of sugar content of the sweet potato is optimized by comparing the value derived by applying the first derivative to the NIR spectrum of the sweet potato sample, and comparing the sweet potato sample with the sugar content measured by the phenol-sulfuric acid method. Can be derived. In the following example of the present invention, a value derived from the NIR spectrum through multivariate regression analysis on the same sample based on the sugar content measured by the sugar analysis method for a sweet potato sample is a value measured by the sugar content measurement method The sugar content calibration curve optimized for sweet potato was derived by optimizing in the direction of minimizing the error so as to approximate to. As the reference value, the sugar analysis method for measuring the sugar content of the standard sweet potato sample is not limited to the above-listed methods, and the sugar content analysis method known in the industry can be used without limitation.

In one embodiment of the present invention, the sugar content calibration curve of the sweet potato is represented by Equation 2 below, and when x′ in Equation 2 is the spectral value of the sweet potato sample, the y′(%) of the sugar content measurement of the sweet potato sample is Can be calculated.

[Equation 2]

y′= 0.5792 x′+ 0.9183

In step b) of the present invention, the calibration curve of the sweet potato content of the sweet potato can be derived by optimizing the value derived by applying the second derivative to the NIR spectrum of the sweet potato sample and comparing the sweet potato sample with the heavy protein content measured by the image analysis method. Can. In the following examples of the present invention, values derived from NIR spectra for multi-variate regression analysis for the same sample based on the protein content measured by the protein analysis method for sweet potato samples are measured by the protein content measurement method Optimized in the direction of minimizing the error so as to approximate, the weight curve for the heavy whitetail content optimized for sweet potatoes was derived. As a reference value, the method for measuring the content of the medium to measure the content of the medium to the standard sweet potato sample is not limited to the above-listed methods, and the method of analyzing the content of the medium known in the industry can be used without limitation.

In one embodiment of the present invention, the calibration curve of the heavy protein content of the sweet potato is represented by the following Equation 3, and when x″ in Equation 3 is the spectral value of the sweet potato sample, the measurement value of the heavy protein content of the sweet potato sample y″ (%) is Can be calculated.

[Equation 3]

y″= 0.7418 x″+ 0.05369

The method for evaluating the quality of the sweet potato of the present invention may further include a step of determining whether the storage conditions are suitable.

In one embodiment of the present invention, the near-infrared irradiation wavelength for determining the suitability of the storage conditions may be a wavelength in the range of 0 to 2000 nm.

In another embodiment of the present invention, the suitability of the storage condition may be determined as a suitability class when a value calculated by multiplying a near-infrared spectrum value and an eigenvalue is greater than -0.15 and a non-conformance class when less than -0.15.

At this time, the near infrared spectral value is a spectral value measured by irradiating a near-infrared light on a sweet potato sample, and the eigenvalue is the eigenvalue having the greatest dispersion degree obtained by putting Equation 4 below into a matlab program.

[Equation 4]

In the above equation, W is a transformation matrix, S _B is a variance matrix between classes, and S _w is a variance matrix within a class.

Hereinafter, the present invention will be described in more detail through examples. These examples are intended to illustrate the present invention more specifically, but the scope of the present invention is not limited to these examples.

< Example >

material

As a test material, sweet potato of Beniharuka species stored at Gongdeok Agricultural Cooperative in Gimje, North Jeolla Province was used. In addition to Beniharuka species, NIR spectra similar to Beniharuka were analyzed in sweet potato species grown in Korea.

< Example 1>

Destructive quality analysis method (conventional quality analysis method)

For comparison with the results of non-destructive quality analysis using near infrared rays according to the present invention, moisture and sugar content were measured by a conventional destructive quality analysis method. Moisture was analyzed according to the dry weight loss method before food. The average value was used by measuring the moisture content of all three parts of the upper, middle, and lower parts of the raw sweet potato sample. Sugar content was performed by the phenol-sulfuric acid method, and baked at 180° C. for 90 minutes, and then analyzed with a group of sweet potato samples.

<1-1> moisture content analysis

The weighing bottle was left for 2 hours in a dryer set at 105° C., and then left in a desiccator for 30 minutes, cooled to room temperature, and then weighed (a). 3 g of sweet potato sample was put into a weighing bottle, and the weight (b) was measured. After standing for 3 hours in a dryer set at 105°C, it was left for 30 minutes in a desiccator, cooled to room temperature, and then weighed (c).

<1-2> Sugar content analysis

After baking at 180° C. for 90 minutes, 0.1 g was collected from the central portion of the group of sweet potatoes and 10 mL of distilled water was mixed and centrifuged at 12,000 rpm. 0.5 mL of supernatant and 0.5 mL of 5% phenol solution were mixed. To this solution, 2 mL of sulfuric acid was added, followed by vortexing and cooling at room temperature for 10 minutes to measure the absorbance at 490 nm. Glucose at a concentration of 0.025-0.5 mg/mL was used as a standard substance.

<1-3> Jungbaek Content analysis method

After baking at 180° C. for 90 minutes, a sample of the sweet potato was cut at the equator and the cross section was taken with a camera. After saving as an image file, the original image was converted to a black and white image to obtain histogram data. The group of sweet potatoes in which the whitening phenomenon occurred had a histogram value (a) of 120 or more, and the section of group of sweet potatoes in which the whitening did not occur had a histogram value (b) of 120 or less. Therefore, the content of heavy protein was calculated by the following equation.

< Example 2>

Near infrared Using sweet potato quality analysis method

In this experiment, raw sweet potato samples were placed on a near-infrared light-receiving sensor, and then measured using a “non-destructive sugar sorter” manufactured and sold by Hansung Engineering equipped with a near-infrared spectroscopy and sensor. The spectrum was obtained by irradiating near infrared rays to receive transmitted light. In order to prepare a calibration curve, the spectrum was pre-treated with 1st and 2nd derivatives based on the spectrum for all samples, and the optimal value was developed by developing an optimal model using PLS (Partial Least Squares) analysis among the chemometrics methods. Was derived.

The moisture content was 500-600 nm, the sugar content was 800-1,000 nm, and the white matter content was optimal in the wavelength range of 870-1,000 nm.

The moisture content was analyzed by PLS after the second differentiation, the PLS analysis after the first differentiation by the sugar content, and the PLS analysis after the second differentiation.

The calibration formula analyzed by the above method is as follows.

Moisture content: y = 0.5944 x + 22.36

Sugar content: y′= 0.5792 x′+ 0.9183

Heavy white content: y″= 0.7418 x″+ 0.05369

In each of the above formulas, x to x″ are the spectral values for moisture, sugar, and heavy white, respectively. After spectral measurement of individual samples, the spectrum in the wavelength range is quantified using the least squares method and applied to the formula, which is the predicted value (NIR measurement content). To y″ can be quantified and obtained by content. Optimized calibration curves for each component are shown in Figures 3-5.

Meanwhile, the spectrum was normalized in the classification analysis of the storage conditions of sweet potatoes. Linear Discriminant Analysis (LDA) is a method of classifying data by maximizing the ratio of between class scatter and within class scatter by reducing the data dimension. The NIR spectrum (x ₁ , x ₂ , ..., x ₂₀₀₀ ) was defined as a vector, and vectors were specified for each class (Abnormal stored under unsuitable conditions, and Sweet potato stored under suitable conditions). . The fisher's criterion was transformed to find the matrix with the largest variance between classes and the ratio of variances within the class, and the transformation matrix (W) was obtained as follows.

fisher's criterion:

의 최대값을 구하는 함수, W^T는 변환행렬(W)의 전치행렬, S_B는 클래스 간의 분산 행렬, S_w는 클래스 내의 분산 행렬을 의미한다. argmax is

A function for obtaining the maximum value of, W ^T is the transpose matrix of the transformation matrix (W), S _B is the variance matrix between classes, and S _w is the variance matrix within the class.

Conversion matrix:

만족하기 때문에, 변환행렬 (W)의 고유벡터 (V)와 고유값 (λ) 값을 구할 수 있다. 최종적으로 NIR 스펙트럼에 고유벡터 (V)를 곱해주면 LDA 분석으로 축소된 데이터를 구할수 있다. 고유벡터는 NIR 스펙트럼의 x의 수와 똑같은 수로 나오며, 2000개의 고유벡터 중 클래스 간 분류가 잘 되는 2개의 고유벡터를 사용하였다. LDA 1은 1차 고유값을 각각의 NIR 스펙트럼에 곱한 값이며, LDA 2는 2차 고유값을 각각의 NIR 스펙트럼에 곱한 값으로서, 개별 시료에 대한 NIR 스펙트럼 측정 후 2개의 분류 조건에 대하여 최적의 견지에서 데이터를 축소한 수치이다. 참고로 상기 ‘1차 고유값’이란 NIR 스펙트럼 값에 대한 고유값 중 분류하고자하는 두 그룹 클래스 1 및 클래스 2에서 가장 분산이 큰 고유값(클래스 간 분류가 가장 잘 되는 고유값)을 의미한다. The eigenvector (V) and eigenvalue (λ) values are

Since it is satisfactory, the eigenvector (V) and eigenvalue (λ) of the transform matrix (W) can be obtained. Finally, if the NIR spectrum is multiplied by the eigenvector (V), the reduced data can be obtained by LDA analysis. The eigenvectors appear as the same number as the number of x in the NIR spectrum, and among the 2000 eigenvectors, two eigenvectors with good classification between classes were used. LDA 1 is the first eigenvalue multiplied by each NIR spectrum, and LDA 2 is the second eigenvalue multiplied by each NIR spectrum, and it is optimal for two classification conditions after measuring the NIR spectrum for individual samples. From a standpoint, this is a reduction in data. For reference, the'primary eigenvalue' means the eigenvalue with the greatest variance (the eigenvalue with the best classification between classes) in the two group classes 1 and 2, which are to be classified among the eigenvalues for the NIR spectrum value.

The LDA 1 value and histogram distribution for each classification condition are shown in FIG. 6.

Classification range for sweet potato storage conditions:

(Not suitable conditions) LDA 1 <-0.15

(Suitable conditions) LDA 1> -0.15

)을 매트랩(matlab) 프로그램에 넣어 도출할 수 있다.The above has been described to explain the general contents of the Linear Discriminant Analysis (LDA), but in order to actually obtain the LDA 1 value of the present invention, the transmission spectrum value and the conversion obtained by examining NIR in sweet potatoes Matrix formula (

) In a MATLAB program.

< Experimental Example 1>

Of the present invention Near infrared Moisture, sugar, and sweet potato samples measured by the quality analysis method and the conventional quality analysis (general analysis) method Jungbaek Content analysis

In this experiment, compared with the values measured by the actual conventional quality analysis (general analysis) method, it was evaluated how much the values measured by the sweet potato quality analysis method using the near infrared ray of the present invention match.

To this end, the spectrum of the sweet potato sample measured by NIR spectroscopy is substituted into the optimized calibration curve derived from <Example 2>, and the moisture, sugar and heavy white contents obtained and general analysis values for the same sample are shown in Table 1 below. Shown. A total of 50 sweet potato samples were used, and the average values for each of these values were shown. On the other hand, the average value is rounded off to the third decimal place.

Moisture, sugar and heavy white content of sweet potato by the analysis method of the present invention vs. general analysis method Ingredient name Analysis value of the present invention General analysis value Moisture content 55.02
(55.026354) 55.12
(55.123722) Sugar content 2.18
(2.17892) 2.18
(2.181996) Heavy white content 0.21
(0.208105492) 0.21
(0.207918828)

As a result, it was confirmed that the values measured by using NIR from sweet potatoes according to the present invention can be measured at an error range level within ±0.5 (% or ㎖) for all components compared to the existing general analysis values.

So far, the present invention has been focused on the preferred embodiments. Those skilled in the art to which the present invention pertains will understand that the present invention can be implemented in a modified form without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments should be considered in terms of explanation, not limitation. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the equivalent range should be interpreted as being included in the present invention.

Claims (11)

As a method of evaluating the quality of sweet potatoes,
The method comprises: a) measuring a spectrum value by irradiating near-infrared light on a sweet potato sample; And b) calculating moisture, sugar and heavy white contents of sweet potato using the measured spectral values.
Each of the near-infrared radiation wavelength for calculating the moisture, sugar and heavy white content of the sweet potato is 500 to 600 nm, 800 to 1,000 nm of sugar, and 870 to 1,000 nm of white sugar,
In step b), the measured spectral values are substituted into the calibration curves of the moisture, sugar, and heavy protein contents of sweet potato to calculate moisture, sugar, and heavy protein contents of the sweet potato sample, respectively.
The moisture content calibration curve of the sweet potato is represented by Equation 1 below, and when x is the spectral value of the sweet potato sample in Equation 1, the moisture content measurement value y(%) of the sweet potato sample is calculated,
The sugar content calibration curve of the sweet potato is represented by Equation 2, and in the following Equation 2, when x′ is the spectral value of the sweet potato sample, the sugar content measurement value of the sweet potato sample y′(%) is calculated,
The sweet potato content calibration curve of the sweet potato is represented by Equation (3), and when x″ in Equation 3 is the spectral value of the sweet potato sample, the sweet potato content measurement value y″ (%) of the sweet potato sample is calculated. How to evaluate the quality of the
[Equation 1]
y = 0.5944 x + 22.36
[Equation 2]
y′= 0.5792 x′+ 0.9183
[Equation 3]
y″= 0.7418 x″+ 0.05369 delete delete delete delete delete delete delete According to claim 1,
c) further comprising determining whether the storage conditions are suitable. The method of claim 9,
The method of the irradiation wavelength of the near-infrared to determine the suitability of the storage conditions is 0 to 2000 nm. The method of claim 9,
The suitability of the above storage conditions is determined by multiplying the near-infrared spectral value and the eigenvalue to determine a suitability class when the value is greater than -0.15, and a non-conformance class when the value is less than -0.15.
The near-infrared spectrum value is a spectrum value measured by irradiating a near-infrared ray on a sweet potato sample,
The eigenvalue is characterized in that the variance derived by putting Equation 4 below in a matlab program is the largest eigenvalue.
[Equation 4]

In the above equation, W is a transformation matrix, S _B is a variance matrix between classes, and S _w is a variance matrix within a class.

Images