A method for rapidly detecting the fatty acid content of paddy rice based on near-infrared spectroscopy

Abstract

The present invention discloses a method for rapidly detecting the fatty acid content of paddy rice based on near-infrared spectroscopy, which relates to the technical field of rapid detection methods for the components of food crops. In the present invention, a batch of paddy rice samples with different fatty acid contents are prepared into powders as a calibration set. By collecting the near-infrared diffuse reflection and diffuse transmission spectra of the calibration set, performing first-order derivative preprocessing on them respectively, and then splicing the head and tail to obtain the diffuse reflection-diffuse transmission fusion spectrum of the samples, and measuring the fatty acid values of the calibration set by chemical methods, a regression model of the fusion spectrum and fatty acids is constructed using partial least squares method. Finally, when detecting unknown samples, the samples are prepared into powders, the near-infrared diffuse reflection and diffuse transmission spectra are collected, the fusion spectrum is obtained by the same treatment, and the fatty acid prediction result is obtained by calling the corresponding model. The beneficial effects of the present invention are as follows: it can improve the detection accuracy and reduce the detection limit, and this method does not consume reagents, is objective, rapid and simple.

Description

Technical Field

[0001] The present invention relates to the technical field of rapid detection methods for the components of food crops, and particularly to a method for rapidly detecting the fatty acid content of paddy rice based on near-infrared spectroscopy. Background Art

[0002] Rice is one of the most important food crops in China. The stored paddy rice is an important strategic material for the country to cope with natural disasters and emergencies, ensure people's needs, and regulate market supply and demand. With the increase of storage time, paddy rice will gradually deteriorate, affecting its edibility and vitality, and resulting in a decrease in its use value. To ensure the quality and safety of stored grain, it is necessary to regularly detect the storage quality of paddy rice at multiple time nodes such as during grain storage and in and out of storage. The current national standard relies on color, smell, taste score value, and fatty acid value to comprehensively evaluate the storage quality of paddy rice. Among them, the fatty acid value is the only quantitative index. Since the traditional detection method for fatty acids in paddy rice is chemical titration, which has the disadvantages of consuming reagents, slow detection speed, poor environmental protection, and strong subjectivity of detection results, it has long affected the timeliness and accuracy of paddy rice purchase and storage. Therefore, there is an urgent need for a new rapid, accurate, and objective detection method to replace the current method and meet the quality detection requirements of paddy rice storage.

[0003] Near-infrared spectroscopy technology (NIRS) is a modern analytical technology that mainly detects hydrogen-containing groups in products, and has the advantages of rapidity, non-destructiveness, objectivity, and environmental protection. For example, the patent application with the publication number CN113484270A discloses a method for detecting the fat content of single-grain rice using near-infrared spectroscopy. It is feasible to detect the fatty acid value of paddy rice using NIRS. However, since the fatty acid content in paddy rice is in the range of dozens to hundreds of mg / 100g, which is close to or lower than the conventional detection limit (one-thousandth) of NIRS, new methods need to be developed to improve the detection accuracy and lower the detection limit. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method for rapidly detecting the fatty acid content of paddy rice based on near-infrared spectroscopy to overcome the problems of reagent consumption, slow detection speed, poor environmental protection, and poor objectivity of detection results that have long existed in the detection of fatty acid content during paddy rice purchase and storage.

[0005] The present invention solves the above technical problems through the following technical means:

[0006] A method for rapidly detecting the fatty acid content of paddy rice based on near-infrared spectroscopy includes the following steps:

[0007] S1. Collect several rice samples with different fatty acid contents as the near-infrared calibration set, and make each calibration set sample into powder;

[0008] S2. Collect the near-infrared diffuse reflectance and diffuse transmittance spectra of each powder sample by means of diffuse reflection and diffuse transmittance;

[0009] S3. Spectral fusion: Perform first derivative processing on the diffuse reflectance and diffuse transmittance spectra of each powder sample respectively, and then splice the two spectra end to end to obtain the diffuse reflectance-diffuse transmittance fusion spectrum of each sample;

[0010] S4. Detect the fatty acids in each powder sample by means of chemometrics to obtain the fatty acid reference value of each sample;

[0011] S5. Construct a near-infrared spectral calibration model for fatty acids: For the diffuse reflectance-diffuse transmittance fusion spectrum of each sample, further intercept the fragment in the diffuse reflectance range of 4636 - 7583 cm -1 and the fragment in the diffuse transmittance range of 8733 - 11672 cm -1 , then splice them end to end and process the spectrum using the vector normalization algorithm; For the processed diffuse reflectance-diffuse transmittance fusion spectrum, use the partial least squares method to construct a regression model between the spectrum and the chemical determination value of fatty acids, and the number of latent variables used in the model is 12;

[0012] S6. Predict the fatty acid content of the sample to be tested using the constructed regression model.

[0013] Advantageous effects: The method of the present invention performs data fusion and analysis on the diffuse reflectance and diffuse transmittance spectra of rice flour samples, which can improve the detection accuracy and reduce the detection limit. At the same time, it can be used in combination with other near-infrared-based rice quality detection technologies to detect other quality indicators, such as moisture, while detecting the fatty acid content of rice, realizing the simultaneous determination of multiple indicators, and providing technical support for the quality evaluation of the storage quality of rice during grain purchase and storage.

[0014] The detection process of the present invention is more objective than the traditional chemical titration method, does not consume reagents, and is not affected by the subjectivity of the testers; The sample treatment required for detection is less, and the detection process is fast and environmentally friendly.

[0015] In step S5 of the present invention, the band selection and corresponding modeling parameters are screened, thereby improving the accuracy of modeling and prediction. If the modeling parameters such as the selected diffuse reflectance range and diffuse transmittance range are not within the scope recorded in the present invention, the accuracy of modeling and prediction in the present invention is lower than that of the near-infrared diffuse reflectance method and the near-infrared diffuse transmittance method.

[0016] Preferably, in step S1, the powder particle size is not greater than 0.150 mm.

[0017] Preferably, first use a hulling machine to dehull the rice sample into brown rice, and then use a cyclone mill to grind the sample into powder and then sieve it.

[0018] Beneficial effects: To ensure the consistency of the prepared powder.

[0019] Preferably, in step S2, when using diffuse reflection for collection, the sample is placed in a cylindrical glass dish, ensuring that the rice powder in the dish is compact, has a uniform thickness, and the powder thickness is more than 5 mm. Then, the glass dish is placed on the collection window of the spectrometer, and the spectrum is collected using the diffuse reflection mode. The collection range of the spectrum is 11988 - 3996 cm -1 , and the resolution is 7.7 cm -1 . When using diffuse transmission for collection, the sample is placed in a cylindrical glass dish, keeping the rice powder in the dish compact and having a uniform thickness, and the powder thickness is about 2 mm. Then, the glass dish is placed on the collection window of the spectrometer, and the spectrum is collected using the diffuse transmission mode. The collection range of the spectrum is 12489 - 5793 cm -1 , and the resolution is 7.7 cm -1 .

[0020] Preferably, in the said step S2, the stoichiometry method is the potassium hydroxide titration method.

[0021] Preferably, the potassium hydroxide titration method includes the following steps:

[0022] a. Analysis of the sample to be tested: Weigh the test sample and put it into a conical flask. Add 95% ethanol with a pipette. The added volume of 95% ethanol is recorded as V1. Place it on a shaker and shake. Transfer the extract to a centrifuge tube, centrifuge and take the supernatant. Add 1% phenolphthalein - ethanol solution dropwise to the supernatant. The volume of the supernatant is recorded as V2. Titrate with potassium hydroxide - ethanol titrant until it turns slightly pink and remains so for 30 seconds without fading. Record the milliliters of potassium hydroxide - ethanol titrant consumed as V3;

[0023] b. Blank test: Take 95% ethanol and titrate with potassium hydroxide - ethanol titrant until it turns slightly pink and remains so for 30 seconds without fading. Record the milliliters of potassium hydroxide - ethanol titrant consumed as V0;

[0024] c. Result calculation: The fatty acid value M of the test sample is calculated by the following formula: In the formula, M is the fatty acid value of the test sample, with the unit of mg / 100g, c is the molar concentration of the potassium hydroxide - ethanol solution (unit: mol / L), m is the mass of the test sample (unit: gram), and w is the moisture mass fraction of the test sample (the mass of moisture contained in every 100 grams of the test sample, unit: gram).

[0025] Preferably, in step a, 5.0 g of the sample is weighed and placed into a conical flask, 30 ml of 95% ethanol is added using a pipette, and it is shaken on an oscillator for 1 hour; the extract is transferred to a centrifuge tube, and 20 ml of the supernatant is taken by centrifugation and placed into a conical flask. 5 drops of 1% phenolphthalein-ethanol solution are added, and it is titrated with 0.01 mol / L potassium hydroxide-ethanol titrant until it turns slightly pink.

[0026] Preferably, the number of smoothing points for the first derivative processing used in step S3 is 17 points.

[0027] Preferably, in step S6, the sample to be measured is successively subjected to steps S1, S2, and S3 to obtain a diffuse reflection-diffuse transmission fusion spectrum, and then the regression model constructed in step S5 is used to predict the diffuse reflection-diffuse transmission fusion spectrum of the powder to be measured collected, so as to obtain the fatty acid content of the sample to be measured.

[0028] The advantages of the present invention are as follows: The method of the present invention performs data fusion and analysis on the diffuse reflection and diffuse transmission spectra of rice flour samples, and can improve the detection accuracy and reduce the detection limit. At the same time, it can be used in combination with other near-infrared-based rice quality detection technologies to detect other quality indicators, such as moisture, while detecting the fatty acid content of rice, so as to realize the simultaneous determination of multiple indicators, and provide technical support for the quality evaluation of the storage quality of rice during grain purchase and storage.

[0029] The detection process of the present invention is more objective than the traditional chemical titration method, does not consume reagents, and is not affected by the subjectivity of the detection personnel; the sample treatment required for detection is less, and the detection process is fast and environmentally friendly.

[0030] In step S5 of the present invention, the parameters of band selection and corresponding modeling are screened, so as to improve the accuracy of modeling and prediction. If the modeling parameters such as the selected diffuse reflection range and diffuse transmission range are not within the scope recorded in the present invention, the accuracy of modeling and prediction in the present invention is lower than that of the near-infrared diffuse reflection method and the near-infrared diffuse transmission method. Description of the Drawings

[0031] Figure 1 is a flowchart of a rapid detection method for the fatty acid content of rice based on near-infrared spectroscopy technology disclosed in an embodiment of the present invention;

[0032] Figure 2 is a diffuse reflection-diffuse transmission fusion spectrum of a rice flour calibration set of a rapid detection method for the fatty acid content of rice based on near-infrared spectroscopy technology disclosed in an embodiment of the present invention;

[0033] Figure 3 is a scatter plot of the leave-one-out cross-validation results of a diffuse reflection-diffuse transmission fusion model of rice flour of a rapid detection method for the fatty acid content of rice based on near-infrared spectroscopy technology disclosed in an embodiment of the present invention;

[0034] Figure 4 It is a scatter plot of the verification results of the rice flour diffuse reflection-diffuse transmission fusion model for a rapid detection method of the fatty acid content in paddy rice based on near-infrared spectroscopy disclosed in the embodiments of the present invention. Specific embodiments

[0035] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0036] The test materials and reagents used in the following embodiments can be obtained from commercial channels without special instructions.

[0037] For those not specifying specific techniques or conditions in the embodiments, they can all be carried out according to the techniques or conditions described in the literature in this field or according to the product instructions.

[0038] As Figure 1 shown, a method for rapidly detecting the fatty acid content in paddy rice based on near-infrared spectroscopy specifically includes the following steps:

[0039] S1: Sample collection and processing:

[0040] Collect 30 and 16 paddy rice samples with different fatty acid contents, a total of 46 samples, which are used as the calibration set (denoted as the rice flour calibration set) for near-infrared method calibration and the verification set (denoted as the rice flour verification set) for verifying the results of this method respectively; process each of the above samples, use a rice huller to hull the samples into brown rice, and then use a cyclone mill to grind the samples into powder and pass through a 100-mesh sieve (the powder particle size is not greater than 0.150 mm).

[0041] S2: Spectrum collection:

[0042] Collect the spectra of the rice flour diffuse reflection calibration set samples in a diffuse reflection manner on a Bruker MPA Fourier transform near-infrared spectrometer in Germany. During collection, place the sample in a cylindrical glass dish, ensure that the rice flour in the dish is compact, has a uniform thickness, and the powder thickness is more than 5 mm. Then place the glass dish on the collection window of the spectrometer and collect the spectra using the diffuse reflection mode. The collection range of the spectra is 11988 - 3996 cm -1 , with a resolution of 7.7 cm -1 , and each sample is collected 2 times and the average value is taken.

[0043] The spectra of the diffuse transmission calibration set samples of rice flour were collected in the diffuse transmission mode on a Bruker MPA Fourier transform near-infrared spectrometer in Germany. When collecting in the diffuse transmission mode, the sample was placed in a cylindrical glass dish, ensuring that the rice flour in the dish was compact, with a uniform thickness of about 2 mm. Then, the glass dish was placed on the collection window of the spectrometer, and the spectra were collected using the diffuse transmission mode. The collection range of the spectra was 12489 - 5793 cm -1 , with a resolution of 7.7 cm -1 .

[0044] S3: Spectral fusion: The diffuse reflectance and diffuse transmission spectra of each sample were processed using the first derivative (17-point smoothing) respectively. Then, the processed diffuse reflectance and diffuse transmission spectra were spliced end to end to obtain the diffuse reflectance-diffuse transmission fusion spectrum of rice flour, as Figure 2 shown. As Figure 2 can be seen, the absorbances of the two range bands of diffuse reflectance and diffuse transmission that make up the fusion spectrum fluctuate around 0. The peaks and valleys in the figure represent the absorption peaks of various components. Among them, the peaks and valleys at positions such as 1021 nm (9795 cm -1 ), 1210 nm (8264 cm -1 ), 1406 nm (7112 cm -1 ), 1760 nm (5681 cm -1 ), 2114 nm (4730 cm -1 ) are related to the absorption of fat, fatty hydrocarbons, and fatty acids, indicating that the fused spectrum carries sufficient information about the fatty acids in rice flour.

[0045] S4: Fatty acid detection: The fatty acids in each powder were detected using the potassium hydroxide titration method to obtain the fatty acid reference value for each sample. The specific steps of this method are as follows:

[0046] a. Analysis of the sample to be measured: Weigh 5.0 g of rice flour and put it into a 150 ml conical flask. Add 30 ml (V1) of 95% ethanol using a pipette, and shake it on an oscillator for 1 hour. Transfer the extract to a 50 ml centrifuge tube, centrifuge it at a speed of 2000 revolutions per minute for 5 minutes, take 20 ml (V2) of the supernatant and put it into a conical flask. Add 5 drops of 1% phenolphthalein-ethanol (95%) solution, and titrate it with 0.01 mol / L potassium hydroxide-ethanol titrant until it turns slightly pink and remains unchanged for 30 seconds. Record the milliliters (V3) of the potassium hydroxide-ethanol titrant consumed;

[0047] b. Blank test: Take 30 ml of 95% ethanol, titrate it with 0.01 mol / L potassium hydroxide-ethanol titrant until it turns slightly pink and remains unchanged for 30 seconds. Record the milliliters (V0) of the potassium hydroxide-ethanol titrant consumed;

[0048] c. Calculation of results: The fatty acid value M of the sample is calculated using the following formula: Where M is the fatty acid value of the sample, in mg / 100g, c is the molar concentration of potassium hydroxide-ethanol solution (in mol / L), m is the mass of the sample (in grams), and w is the water mass fraction of the sample (the mass of water per 100g of the sample, in grams). The water mass fraction w of each sample is detected by drying and weighing method. The steps are: weigh 1.5g rice flour sample (m1), put it in an aluminum box (the weight of the aluminum box is recorded as m2) and dry it in an oven at 135°C for 3 hours, then take it out and restore it to room temperature in a dryer, then weigh the total weight of the aluminum box and the dried rice flour, and record its weight as m3. Calculate the water mass fraction according to the following formula: w = (m 1+ m 2- m3) / m1.

[0049] The descriptive statistics of the fatty acid content of rice determined by chemical method are shown in Table 1. As can be seen from Table 1, the fatty acid content of the calibration sample set has a wide distribution, and its range can cover the content range of the validation set, and can cover the fatty acid content of most new and old grains, which is well representative.

[0050] Table 1 Descriptive statistics of fatty acid content in rice determined by chemical method

[0051]

[0052]

[0053] S5: Construction of near-infrared spectroscopy model of rice fatty acids:

[0054] The diffuse reflection-diffuse transmission fusion spectrum of each rice flour sample was further used to screen the spectral range for modeling so that the constructed model could be most relevant to the measured fatty acid content. The diffuse reflection range was intercepted at 4636-7583cm -1 The fragment and diffuse transmission range is 8733-11672cm -1 The fragments were then spliced ​​end to end and the spectra were processed using a vector normalization algorithm. Both ranges contain the characteristic peaks related to fatty acids described in S3 (4730 cm -1 5681cm -1 、7112cm -1 and 9795cm -1 etc.), of which the diffuse transmission range is 1210nm (8264cm -1 ) and 1406nm(7112cm -1 ), diffuse reflection range of 1760nm (5681cm -1 ) and 2114nm(4730cm -1) The peaks and valleys are related to the absorption of fat and fatty acids, indicating that the fused spectrum carries sufficient information about the fatty acids in rice flour. For the processed diffuse reflectance-diffuse transmittance fused spectrum, the partial least squares method was used to construct a regression model between the spectrum and the chemical determination values of fatty acids. The number of latent variables used in the model was 12.

[0055] As a control, using the diffuse reflectance spectrum and diffuse transmittance spectrum of each rice flour sample, the partial least squares method was used to construct a diffuse reflectance detection model and a diffuse transmittance detection model for rice flour fatty acids. Among them, the spectral range of the rice flour fatty acid diffuse reflectance model was 4798 - 3996 cm -1 and 9589 - 10391 cm -1 , the spectral preprocessing was vector normalization, and the number of latent variables was 4; while the spectral range of the rice flour fatty acid diffuse transmittance model was 7128 - 7807 cm -1 , 8470 - 9813 cm -1 , and 11147 - 11826 cm -1 , the spectral preprocessing was vector normalization, and the number of latent variables was 10. Both groups of models were the optimal models selected by the interval partial least squares method (siPLS) in different spectral ranges and spectral preprocessing combinations.

[0056] To verify the calibration performance of the model constructed in step S5, leave-one-out cross-validation was performed on the three groups of rice flour fatty acid models constructed. Among them, the leave-one-out cross-validation results of the fusion model are as Figure 3 shown, while the leave-one-out cross-validation results of the fusion model and the control models (diffuse reflectance model and diffuse transmittance model) are shown in Table 2.

[0057] As can be Figure 3 seen, the cross-validation correlation coefficient (R cv ) of the rice paddy fatty acid diffuse reflectance-diffuse transmittance fusion model was 0.873, and the root mean square error of cross-validation (RMSECV) was 6.95. This result indicates that the constructed model has good calibration effect and can accurately predict the calibration set samples.

[0058] As can be seen from Table 2, the leave-one-out cross-validation results of the control models are all inferior to those of the fusion model, with lower R cv and higher RMSECV, indicating that the fusion model has superiority in predicting rice flour fatty acids compared to only using the near-infrared diffuse reflectance and diffuse transmittance methods.

[0059] Table 2 Leave-one-out cross-validation results of the fusion model and the control models for predicting rice flour fatty acids

[0060] Rice flour diffuse reflection-diffuse transmission fusion model Rice flour diffuse reflection model Rice flour diffuse transmission model <![CDATA[R cv > 0.873 0.489 0.8518 RMSECV 6.95 12.4 7.46

[0061] S6: Use the model constructed in steps S1 - S5 to predict the grain sample to be tested. In this embodiment, 16 rice flour validation set samples are used for testing, including the following steps:

[0062] a. Perform the same processing on the sample to be tested as in step S1, and prepare it into powder.

[0063] b. Perform near-infrared spectrum collection and spectrum fusion on the powder to be tested to obtain diffuse reflection - diffuse transmission fusion spectrum. The spectrum collection and fusion steps are the same as S2 and S3 respectively.

[0064] c. Use the model constructed in step S5 to predict the diffuse reflection - diffuse transmission fusion spectrum of the powder to be tested collected, and obtain the fatty acid content of the sample to be tested.

[0065] To verify the accuracy of the measured fatty acid content, taking the potassium hydroxide titration method used in step S4 as the standard, measure the fatty acid content of the validation set samples. The distribution of the measured fatty acid content is shown in Table 1. Evaluate the spectral prediction values and chemical determination values of the fatty acid content of the validation set. The evaluation results are as Figure 4 shown.

[0066] As Figure 4 can be seen, when predicting the fatty acid of paddy using the near-infrared diffuse reflection - diffuse transmission model method, the prediction correlation coefficient (R p ) between the predicted value of fatty acid and the chemical detection value is 0.827, and the root mean square error of prediction (RMSEP) is 7.64. This result indicates that the method used has a high correlation and low error between the prediction result of paddy fatty acid and the chemical determination result, that is, this method can achieve relatively accurate prediction for external samples.

[0067] Through the above technical solutions, a rapid detection method for the fatty acid content of paddy provided by the present invention has the advantages of objective and accurate detection results and rapid detection process, and can effectively improve the quality detection efficiency during paddy storage and purchase. This method can also be extended to the rapid quality detection of other agricultural products, such as the detection of the fatty acid content of corn, and is expected to better serve grain storage and provide technical support for the national grain quality safety.

[0068] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for rapidly detecting the fatty acid content of paddy rice based on near-infrared spectroscopy, characterized in that: It includes the following steps: S1. Collect several rice samples with different fatty acid contents as the near-infrared calibration set, and make each calibration set sample into powder; S2. Collect the near-infrared diffuse reflection and diffuse transmission spectra of each powder sample in the ways of diffuse reflection and diffuse transmission. The collection range of the diffuse reflection spectrum is 11988 - 3996 cm -1 , and the collection range of the diffuse transmission spectrum is 12489 - 5793 cm -1 ; S3. Spectral fusion: Perform first derivative processing on the diffuse reflection and diffuse transmission spectra of each powder sample respectively, and then splice the two spectra head to tail to obtain the diffuse reflection-diffuse transmission fusion spectrum of each sample; The smoothing points for the first derivative processing used are 17 points; S4. Use chemometrics to detect the fatty acids in each powder sample to obtain the fatty acid reference value of each sample; S5. Construct a near-infrared spectroscopy calibration model for fatty acids: For the diffuse reflectance-diffuse transmittance fusion spectrum of each sample, further intercept the fragment in the diffuse reflectance range of 4636 - 7583 cm -1 and the fragment in the diffuse transmittance range of 8733 - 11672 cm -1 Then splice the head and tail, and process the spectrum using the vector normalization algorithm; for the processed diffuse reflectance-diffuse transmittance fusion spectrum, use partial least squares method to construct a regression model between the spectrum and the chemical determination value of fatty acids, and the number of latent variables used in the model is 12; S6. Use the constructed regression model to predict the fatty acid content of the test sample.

2. The method for rapidly detecting the fatty acid content of paddy rice based on near-infrared spectroscopy according to claim 1, wherein: In the step S1, the powder particle size is not greater than 0.150 mm.

3. The method for rapidly detecting the fatty acid content of paddy rice based on near-infrared spectroscopy according to claim 2, wherein: First, use a rice huller to hull the paddy sample into brown rice, and then use a cyclone mill to grind the sample into powder and then sieve it.

4. The method for rapidly detecting the fatty acid content of paddy rice based on near-infrared spectroscopy according to claim 1, wherein: In step S2, when using diffuse reflection for collection, the sample is placed in a cylindrical glass dish, ensuring that the rice powder in the dish is compact, has a uniform thickness, and the powder thickness is more than 5 mm. Then, the glass dish is placed on the collection window of the spectrometer, and the spectrum is collected using the diffuse reflection mode with a resolution of 7.7 cm -1 , and each sample is collected 2 times and the average value is taken; when using diffuse transmission for collection, the sample is placed in a cylindrical glass dish, keeping the rice powder in the dish compact, having a uniform thickness, and the powder thickness is about 2 mm. Then, the glass dish is placed on the collection window of the spectrometer, and the spectrum is collected using the diffuse transmission mode with a resolution of 7.7 cm -1 .

5. The method for rapidly detecting the fatty acid content of paddy rice based on near-infrared spectroscopy according to claim 1, wherein: In the step S2, the chemometrics is the potassium hydroxide titration method.

6. The method for rapidly detecting the fatty acid content of paddy rice based on near-infrared spectroscopy according to claim 5, wherein: The potassium hydroxide titration method includes the following steps: a. Test sample analysis: Weigh the test sample and put it into a conical flask, add 95% ethanol with a pipette, and record the added amount of 95% ethanol as V1. Place it on an oscillator and shake; Transfer the extract to a centrifuge tube, centrifuge and take the supernatant, add 1% phenolphthalein-ethanol solution dropwise to the supernatant, record the volume of the supernatant as V2, and titrate with potassium hydroxide-ethanol titrant until it turns slightly red and remains unchanged for 30 seconds, and record the milliliters of potassium hydroxide-ethanol titrant consumed as V3; b. Blank test: Take 95% ethanol and titrate it with potassium hydroxide-ethanol titrant until it turns slightly red and remains unchanged for 30 seconds, and record the milliliters of potassium hydroxide-ethanol titrant consumed as V0; c. Result calculation: The fatty acid value M of the sample is calculated using the following formula: Where M is the fatty acid value of the sample, in mg / 100g; c is the molar concentration of the potassium hydroxide-ethanol solution, in mol / L; m is the mass of the sample, in grams; and w is the moisture mass fraction of the sample, that is, the mass of moisture contained in every 100 grams of the test sample, in grams.

7. The method for rapidly detecting the fatty acid content of paddy rice based on near-infrared spectroscopy according to claim 6, wherein: In the step a, weigh 5.0 g of the test sample and put it into a conical flask, add 30 ml of 95% ethanol with a pipette, place it on an oscillator and shake for 1 hour; Transfer the extract to a centrifuge tube, centrifuge and take 20 ml of the supernatant into a conical flask, add 5 drops of 1% phenolphthalein-ethanol solution, and titrate with 0.01 mol / L potassium hydroxide-ethanol titrant until it turns slightly red.

8. The method for rapidly detecting the fatty acid content of paddy rice based on near-infrared spectroscopy according to claim 1, wherein: In the step S6, the test sample is successively subjected to steps S1, S2, and S3 to obtain the diffuse reflection-diffuse transmission fusion spectrum, and then the regression model constructed in step S5 is used to predict the diffuse reflection-diffuse transmission fusion spectrum of the collected test powder to obtain the fatty acid content of the test sample.

Images