CN112748085A - Method for establishing near-infrared model for predicting cadmium content in rice and method for predicting cadmium content in rice - Google Patents
Method for establishing near-infrared model for predicting cadmium content in rice and method for predicting cadmium content in rice Download PDFInfo
- Publication number
- CN112748085A CN112748085A CN202011525736.2A CN202011525736A CN112748085A CN 112748085 A CN112748085 A CN 112748085A CN 202011525736 A CN202011525736 A CN 202011525736A CN 112748085 A CN112748085 A CN 112748085A
- Authority
- CN
- China
- Prior art keywords
- rice
- cadmium
- infrared
- cadmium content
- infrared information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 241000209094 Oryza Species 0.000 title claims abstract description 129
- 235000007164 Oryza sativa Nutrition 0.000 title claims abstract description 129
- 235000009566 rice Nutrition 0.000 title claims abstract description 129
- 238000000034 method Methods 0.000 title claims abstract description 102
- 229910052793 cadmium Inorganic materials 0.000 title claims abstract description 95
- BDOSMKKIYDKNTQ-UHFFFAOYSA-N cadmium atom Chemical compound [Cd] BDOSMKKIYDKNTQ-UHFFFAOYSA-N 0.000 title claims abstract description 95
- 230000029087 digestion Effects 0.000 claims description 43
- VLTRZXGMWDSKGL-UHFFFAOYSA-N perchloric acid Chemical compound OCl(=O)(=O)=O VLTRZXGMWDSKGL-UHFFFAOYSA-N 0.000 claims description 20
- 238000001228 spectrum Methods 0.000 claims description 14
- 238000010521 absorption reaction Methods 0.000 claims description 13
- GRYLNZFGIOXLOG-UHFFFAOYSA-N Nitric acid Chemical compound O[N+]([O-])=O GRYLNZFGIOXLOG-UHFFFAOYSA-N 0.000 claims description 10
- 229910017604 nitric acid Inorganic materials 0.000 claims description 10
- 230000003595 spectral effect Effects 0.000 claims description 10
- 239000003153 chemical reaction reagent Substances 0.000 claims description 9
- 238000012216 screening Methods 0.000 claims description 7
- 239000002253 acid Substances 0.000 claims description 6
- 239000007788 liquid Substances 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 5
- 238000010438 heat treatment Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000002790 cross-validation Methods 0.000 claims description 3
- 238000002156 mixing Methods 0.000 claims description 3
- 238000004380 ashing Methods 0.000 claims description 2
- 238000000889 atomisation Methods 0.000 claims description 2
- 238000001035 drying Methods 0.000 claims description 2
- 239000003795 chemical substances by application Substances 0.000 claims 1
- 238000001514 detection method Methods 0.000 abstract description 7
- 238000004458 analytical method Methods 0.000 abstract description 3
- 239000000243 solution Substances 0.000 description 13
- 238000004497 NIR spectroscopy Methods 0.000 description 5
- 238000002329 infrared spectrum Methods 0.000 description 5
- 238000012937 correction Methods 0.000 description 4
- 229910001385 heavy metal Inorganic materials 0.000 description 4
- 235000013305 food Nutrition 0.000 description 3
- 238000012569 chemometric method Methods 0.000 description 2
- 238000005034 decoration Methods 0.000 description 2
- 238000009795 derivation Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000000862 absorption spectrum Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000007865 diluting Methods 0.000 description 1
- 238000010790 dilution Methods 0.000 description 1
- 239000012895 dilution Substances 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000011049 filling Methods 0.000 description 1
- 229910052500 inorganic mineral Inorganic materials 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000011068 loading method Methods 0.000 description 1
- 239000011707 mineral Substances 0.000 description 1
- 102000004169 proteins and genes Human genes 0.000 description 1
- 108090000623 proteins and genes Proteins 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 229910021642 ultra pure water Inorganic materials 0.000 description 1
- 239000012498 ultrapure water Substances 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 229940088594 vitamin Drugs 0.000 description 1
- 229930003231 vitamin Natural products 0.000 description 1
- 235000013343 vitamin Nutrition 0.000 description 1
- 239000011782 vitamin Substances 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
- 239000002023 wood Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/3563—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light for analysing solids; Preparation of samples therefor
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/35—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light
- G01N21/359—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light using near infrared light
Landscapes
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention provides a method for establishing a near-infrared model for predicting cadmium content in rice and a method for predicting cadmium content in rice, and relates to the technical field of analysis and detection. According to the method, the cadmium-related characteristic near-infrared information of the rice sample is screened by adopting a backward interval least square method, the screened cadmium-related characteristic near-infrared information is fitted with the cadmium content, the accuracy of the obtained near-infrared model is better, and the prediction capability is greatly improved. The invention also provides a method for predicting the cadmium content in the rice by using the near infrared model established by the establishing method in the technical scheme, the method can realize the prediction of the cadmium content in the rice only by measuring the near infrared information of the rice, and has the advantages of high accuracy, simple operation and wide application range.
Description
Technical Field
The invention relates to the technical field of analysis and detection, in particular to a method for establishing a near-infrared model for predicting cadmium content in rice and a method for predicting cadmium content in rice.
Background
Rice is a staple food for 50% of the world's population, providing people with major calories, vitamins, minerals and proteins. With the improvement of living standard of people, the safety problem of rice draws wide attention. Rice faces an increasing risk due to the severe threat of heavy metals in the environment. Research shows that the heavy metal cadmium in the environment can be transferred to rice along with polluted soil. Therefore, the method for rapidly detecting the content of the heavy metal cadmium in the rice is urgent and has important significance.
The near infrared spectroscopy (NIR) technology has the characteristics of rapidness, accuracy and no damage, is used for replacing the traditional chemometric analysis method in recent years, and is widely applied to the industries of food, medicine and wood. This is because the absorption spectrum information of the near infrared spectrum in different bands represents different chemical bond information. Therefore, near infrared spectroscopy can be combined with supervised chemometric methods to build quantitative predictive models. At present, the near infrared spectrum technology combined with a chemometric method has been successfully applied to quality detection of various foods.
Partial Least Squares (PLS) is though the most widely used classical modeling method in near infrared quantitative analysis. However, in the prior art, when the model established by the partial least square method is used for detecting the content of heavy metals in rice, the influence factors are more and the detection accuracy is low.
Disclosure of Invention
In view of the above, the present invention aims to provide a method for establishing a near-infrared model for predicting cadmium content in rice and a method for predicting cadmium content in rice. The near-infrared model established by the invention can realize accurate prediction of cadmium content in the rice sample, and the method is simple and has wide application range.
In order to achieve the above object, the present invention provides the following technical solutions:
the invention provides a method for establishing a near-infrared model for predicting cadmium content in rice, which comprises the following steps:
measuring the cadmium content of the rice sample;
acquiring cadmium-related characteristic near-infrared information of a rice sample;
fitting the cadmium content of the rice sample and the cadmium related characteristic near-infrared information by using a backward interval least square method to obtain a near-infrared model;
the method for acquiring the cadmium-related characteristic near-infrared information of the rice sample comprises the following steps:
measuring near infrared information of the rice sample;
performing spectrum pretreatment on the near-infrared information of the rice sample to obtain pretreated near-infrared information;
based on the cadmium content of the rice sample and the preprocessed near-infrared information, performing pre-modeling by using a backward interval least square method to obtain a series of pre-models; during the pre-modeling, parameters of a backward interval least square method comprise: the number n of the divided intervals is 15-25, and the number LVs of the main factors is 1-10;
screening the pre-model with the minimum cross validation root mean square error (RMESCV) in the series of pre-models;
and extracting the spectral information of the pre-model with the minimum RMESCV as the cadmium related characteristic near infrared information of the rice sample.
Preferably, the method for determining the cadmium content of the rice sample comprises the following steps:
mixing a rice sample and a digestion reagent, and digesting to obtain a digestion solution;
and determining the digestion solution by adopting an atomic absorption spectrometer to obtain the cadmium content of the rice sample.
Preferably, the digestion reagent is a mixed acid of nitric acid and perchloric acid; the mass concentration of the nitric acid is 65-68%; the mass concentration of the perchloric acid is 70-72%; the volume ratio of the nitric acid to the perchloric acid is 9: 1; the material-liquid ratio of the rice sample to the digestion reagent is (0.3-0.5) g: (8-10) mL.
Preferably, the digestion temperature is 180 ℃ and the digestion time is 2-2.5 h.
Preferably, after digestion, the method further comprises heating the obtained digestion system to 220 ℃ to drive acid for 1-1.5 h until the digestion system is clear and transparent.
Preferably, the parameters of the atomic absorption spectrometer include: light source: a hollow cathode lamp; lamp current: 50 percent; wavelength: 228.8 nm; pass band width: 0.5 nm; drying temperature: 120 ℃/20 s; ashing temperature: 800 ℃/20 s; atomization temperature: 1300 ℃/3 s; clearing: 2500 deg.C/3 s.
Preferably, the parameters for determining the near infrared information of the rice sample include: the sampling mode is diffuse reflection of an integrating sphere; the scanning range is 4000-12000 cm-1(ii) a Resolution was 16cm-1(ii) a The number of scans was 64; the number of sample points is 1154.
Preferably, the method of spectral preprocessing is a first derivative method.
The invention also provides a method for predicting the cadmium content in rice, which comprises the following steps:
measuring near infrared information of rice to be measured;
and substituting the near-infrared information of the rice to be detected into the near-infrared model obtained by the establishing method in the technical scheme to obtain the cadmium content of the rice to be detected.
The invention provides a method for establishing a near-infrared model for predicting cadmium content in rice, which comprises the following steps: measuring the cadmium content of the rice sample; acquiring cadmium-related characteristic near-infrared information of a rice sample; fitting the cadmium content of the rice sample and the cadmium related characteristic near-infrared information by using a backward interval least square method to obtain a near-infrared model; the method for acquiring the cadmium-related characteristic near-infrared information of the rice sample comprises the following steps: measuring near infrared information of the rice sample; performing spectrum pretreatment on the near-infrared information of the rice sample to obtain pretreated near-infrared information; based on the cadmium content of the rice sample and the preprocessed near-infrared information, performing pre-modeling by using a backward interval least square method to obtain a series of pre-models; during the pre-modeling, parameters of a backward interval least square method comprise: the number n of the divided intervals is 15-25, and the number LVs of the main factors is 1-10; screening the pre-model with the minimum RMESCV in the series of pre-models; and extracting the spectral information of the pre-model with the minimum RMESCV as the cadmium related characteristic near infrared information of the rice sample. According to the method, the cadmium-related characteristic near-infrared information of the rice sample is screened by adopting a backward interval least square method, the screened cadmium-related characteristic near-infrared information is fitted with the cadmium content, the accuracy of the obtained near-infrared model is better, and the prediction capability is greatly improved.
The invention also provides a method for predicting the cadmium content in the rice by using the near infrared model established by the establishing method in the technical scheme, the method can realize the prediction of the cadmium content in the rice only by measuring the near infrared information of the rice, and has the advantages of high accuracy, simple operation and wide application range.
Drawings
FIG. 1 shows raw near infrared information for 825 samples;
FIG. 2 is a diagram of near infrared information pre-processed by a first-order derivation method;
FIG. 3 shows the near infrared spectrum of cadmium in rice, which is the most preferred rice for BiPLS.
Detailed Description
The invention provides a method for establishing a near-infrared model for predicting cadmium content in rice, which comprises the following steps:
measuring the cadmium content of the rice sample;
acquiring cadmium-related characteristic near-infrared information of a rice sample;
fitting the cadmium content of the rice sample and the cadmium related characteristic near-infrared information by using a backward interval least square method to obtain a near-infrared model;
the method for acquiring the cadmium-related characteristic near-infrared information of the rice sample comprises the following steps:
measuring near infrared information of the rice sample;
performing spectrum pretreatment on the near-infrared information of the rice sample to obtain pretreated near-infrared information;
based on the cadmium content of the rice sample and the preprocessed near-infrared information, performing pre-modeling by using a backward interval least square method to obtain a series of pre-models; during the pre-modeling, parameters of a backward interval least square method comprise: the number n of the divided intervals is 15-25, and the number LVs of the main factors is 1-10;
screening the pre-model with the minimum RMESCV in the series of pre-models;
and extracting the spectral information of the pre-model with the minimum RMESCV as the cadmium related characteristic near infrared information of the rice sample.
The method is used for measuring the cadmium content of the rice sample. In the invention, the method for measuring the cadmium content of the rice sample is preferably GB 5009.15-2014 method; said GB 5009.15-2014 preferably comprises the following steps:
mixing a rice sample and a digestion reagent, and digesting to obtain a digestion solution;
and determining the digestion solution by adopting an atomic absorption spectrometer to obtain the cadmium content of the rice sample.
According to the invention, a rice sample and a digestion reagent are mixed and digested to obtain digestion liquid.
In the present invention, the digestion reagent is preferably a mixed acid of nitric acid and perchloric acid; the mass concentration of the nitric acid is preferably 65-68%; the mass concentration of the perchloric acid is preferably 70-72%; the volume ratio of the nitric acid to the perchloric acid is preferably 9: 1; the material-liquid ratio of the rice sample to the digestion reagent is preferably (0.3-0.5) g: (8-10) mL, more preferably 0.3 g: 8 mL. In the invention, the digestion temperature is preferably 180 ℃, the digestion time is preferably 2-2.5 h, and the digestion time is more preferably 2 h. After digestion is finished, the method preferably further comprises the step of heating the digestion system to 220 ℃ to remove acid for 1-1.5 h until the digestion system is clear and transparent. In the present invention, the volume of the final digestion solution is preferably 1mL, and the digestion solution is preferably diluted appropriately before the on-line detection, and the dilution factor in the present invention is not particularly limited, and may be selected according to actual conditions.
After the digestion solution is obtained, the invention adopts an atomic absorption spectrometer to measure the digestion solution, and the cadmium content of the rice sample is obtained. In the present invention, the parameters of the atomic absorption spectrometer are shown in table 1.
TABLE 1 operating parameters of atomic absorption spectrometer
The method obtains the cadmium-related characteristic near-infrared information of the rice sample. In the invention, the method for acquiring the cadmium-related characteristic near-infrared information of the rice sample comprises the following steps:
measuring near infrared information of the rice sample;
performing spectrum pretreatment on the near-infrared information of the rice sample to obtain pretreated near-infrared information;
based on the cadmium content of the rice sample and the preprocessed near-infrared information, performing pre-modeling by using a backward interval least square method to obtain a series of pre-models; during the pre-modeling, parameters of a backward interval least square method comprise: the number n of the divided intervals is 15-25, and the number LVs of the main factors is 1-10;
screening the pre-model with the minimum RMESCV in the series of pre-models;
and extracting the spectral information of the pre-model with the minimum RMESCV as the cadmium related characteristic near infrared information of the rice sample.
The invention measures the near infrared information of the rice sample. In the invention, the method for measuring the near infrared information of the rice sample comprises the following steps: directly carrying out near-infrared detection on the rice sample by adopting an atomic absorption spectrophotometer; the parameters of the atomic absorption spectrophotometer preferably include: the sampling mode is diffuse reflection of an integrating sphere; the scanning range is 4000-12000 cm-1(ii) a Resolution was 16cm-1(ii) a The number of scans was 64; the number of sample points is 1154.
After the near-infrared information of the rice sample is obtained, the invention carries out spectrum pretreatment on the near-infrared information of the rice sample to obtain the pretreated near-infrared information. In the present invention, the method of spectral preprocessing is preferably a first derivative method.
After the cadmium content of the rice sample and the preprocessed near-infrared information are obtained, the rice sample cadmium content and the preprocessed near-infrared information are used for performing pre-modeling by using a Backward interval least square method (BiPLS) to obtain a series of pre-models. In the present invention, in the pre-modeling, parameters of the backward interval least square method include: the number n of the divided sections is 15-25, and the number LVs of the main factors is 1-10.
After a series of pre-models are obtained, the pre-model with the minimum RMESCV in the series of pre-models is screened. The screening method of the present invention is not particularly limited, and a screening means known to those skilled in the art may be used.
After the minimal RMESCV pre-model is obtained, the spectral information of the minimal RMESCV pre-model is extracted to be used as the cadmium characteristic near-infrared information of the rice sample. The extraction method is not particularly limited, and the extraction method known to those skilled in the art can be adopted.
After the cadmium content and the cadmium related characteristic near-infrared information of the rice sample are obtained, fitting the cadmium content and the cadmium related characteristic near-infrared information of the rice sample by using a back interval least square method to obtain a near-infrared model.
The fitting method is not particularly limited, and fitting means well known to those skilled in the art may be used.
According to the method, the cadmium-related characteristic near-infrared information of the rice sample is screened by adopting a backward interval least square method, the screened cadmium-related characteristic near-infrared information is fitted with the cadmium content, the accuracy of the obtained near-infrared model is better, and the prediction capability is greatly improved.
The invention also provides a method for predicting the cadmium content in rice, which comprises the following steps:
measuring near infrared information of rice to be measured;
and substituting the near-infrared information of the rice to be detected into the near-infrared model obtained by the establishing method in the technical scheme to obtain the cadmium content of the rice to be detected.
The invention determines the near infrared information of the rice to be measured. In the invention, the method and parameters for determining the near infrared information of the rice to be detected are preferably consistent with the technical scheme, and are not described again.
After the near-infrared information of the rice to be detected is obtained, the near-infrared information of the rice to be detected is substituted into the model obtained by the establishing method in the technical scheme, and the cadmium content of the rice to be detected is obtained.
The prediction method provided by the invention can realize the prediction of the cadmium content in the rice only by measuring the near infrared information of the rice, and has the advantages of high accuracy, simple operation and wide application range.
The method for establishing the near-infrared model for predicting the cadmium content in the rice and the method for predicting the cadmium content in the rice provided by the invention are described in detail with reference to the following embodiments, but the method and the method are not to be construed as limiting the scope of the invention.
Example 1
Firstly, determining the cadmium content of a rice sample
825 parts of rice sample are obtained from northern mountain town (198 samples) in Changsha county, new spring town (210 samples) in Xiangyin county, Riboda urban (203 samples) and Changli urban (214 samples) by using a method in national standard GB 5009.15-2014 to measure the content of cadmium in rice.
Weighing 0.3g of rice sample into a digestion bottle, adding 8mL of nitric acid (superior pure, with the mass concentration of 65-68%) and perchloric acid (superior pure, with the mass concentration of 70-72%) according to the volume ratio of 9: 1, placing a digestion bottle on an electric hot plate for digestion at 180 ℃ for 2 hours, then heating to 220 ℃ for digestion for 1 hour until the digestion solution is clear and transparent, wherein the final volume of the digestion solution is 1 mL; diluting the digestion solution to 50mL with ultrapure water to obtain a solution to be detected;
performing on-machine detection on the liquid to be detected by using an atomic absorption spectrophotometer;
the operating parameters of the atomic absorption spectrometer are shown in table 1.
Secondly, acquiring near infrared information of the rice sample
Uniformly filling a rice sample into a sample cup, and placing the sample cup on an atomic absorption spectrophotometer for scanning: the sampling mode is integrating sphere diffuse reflection, and the spectrum scanning range is 4000-12000 cm-1Resolution of 16cm-1Scanning times are 64 times, and sampling points are 1154; and (3) repeating sample loading scanning for each sample, and averagely obtaining a near infrared spectrum so as to eliminate interference caused by non-uniformity of the samples.
Preprocessing of near infrared information
And performing spectrum pretreatment on the original near-infrared information of the rice sample by using matlab software and adopting a first derivative method to obtain the pretreated near-infrared information.
FIG. 1 shows the NIR information of 825 samples, and FIG. 2 shows the NIR information pre-processed by first-order derivation.
Fourthly, pre-modeling
By the Hotelling T-square method (Hotelling T)2) The outlier samples were removed by 18 and, to divide The samples evenly into a representative calibration set and a prediction set, The remaining 807 samples were taken using The Kennard Stone method (K-S algorithm) according to The calibration set: prediction set 2: 1, 538 correction sets and 269 prediction sets are divided, and the results are shown in Table 2. As can be seen from table 2: the division of the sample set is more uniform and reasonable.
Table 2 sample set partitioning results
Based on the cadmium content of the rice sample and the preprocessed near-infrared information, performing pre-modeling by using a backward interval least square method; and equally dividing the full spectrum into n subintervals by a directional interval least square method, sequentially removing one subinterval, performing combined modeling by using the remaining n-1 subintervals, and calculating to obtain n RMSECV values in total. The interval corresponding to the minimum RMSECV value is the first excluded interval, and so on, and the calculation is carried out until the last interval is left.
And dividing the preprocessed spectral data into n subintervals by a backward interval least square method, wherein the value range of n is 15-25. When n takes different values, the feature subintervals selected using BiPLS are shown in table 3.
TABLE 3BiPLS subinterval optimization results
Wherein R is2 cCorrelation coefficient, R, representing correction set2 pThe correlation coefficient represents the prediction set, and RMSEP is the prediction root mean square error.
As can be seen from table 3: when the spectrum is divided into 23 (i.e., n-23), the corresponding rmesccv is minimal.
When the spectrum was divided into 23 intervals and 17 intervals were selected, the pre-model was most effective (i.e., the pre-model with the smallest rmesccv), and the results are shown in table 4. FIG. 3 shows the near infrared spectrum of cadmium in rice selected by BiPLS as an optimum.
TABLE 4 concrete optimization results of BiPLS interval divided into 23 subintervals
And extracting the cadmium related characteristic near-infrared information of the rice sample based on the optimal pre-model.
Fifth, the establishment of the model
And fitting the cadmium content of the rice sample and the cadmium related characteristic near-infrared information by using a backward interval least square method to obtain a near-infrared model.
Sixthly, evaluation of model
In order to check the reliability and effectiveness of the established near-infrared model, when the correction model is optimized, the near-infrared model needs to be analyzed and checked.
The robustness of the model is firstly checked by an internal interactive verification method, and the quality of the model is judged through the correlation coefficient (Rc) and RMSECV of a correction set. In order to test the prediction capability of the model, namely the accuracy and the feasibility of the model, a correlation coefficient (Rp) and a prediction Root Mean Square Error (RMSEP) of a verification set are used as judgment bases, and the larger the correlation coefficient is, the smaller the RMSEP is, and the better the prediction performance of the model is. The final models were evaluated and compared using the RMSEP indices of the validation set, with the results shown in table 5.
The PLS method employs full spectrum modeling: the main factor number (LVs) is selected to be 10;
the iPLS method divides the full spectrum into 5 subintervals, and selects the 3 rd interval (the selected waveband is 8925.14-7150.91) for modeling;
BiPLS divides the full spectrum into 23 sub-intervals and selects 17 sub-intervals for modeling (the selected wavelength bands are: 4744.1-3980.4, 5901.2-5137.6, 7829.8-6680.4, 12096-8223.2).
BiPLS, PLS and ipss modeling methods were compared and the results are shown in table 5. As can be seen from table 5: the correlation coefficients Rc and Rp of the BiPLS are both obviously improved, and the RMSECV and RMSEP are both reduced, which shows that the performance of the BiPLS model, including the accuracy and the prediction capability, are both obviously improved.
TABLE 5 comparison of different modeling methods
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.
Claims (9)
1. A method for establishing a near-infrared model for predicting cadmium content in rice is characterized by comprising the following steps:
measuring the cadmium content of the rice sample;
acquiring cadmium-related characteristic near-infrared information of a rice sample;
fitting the cadmium content of the rice sample and the cadmium related characteristic near-infrared information by using a backward interval least square method to obtain a near-infrared model;
the method for acquiring the cadmium-related characteristic near-infrared information of the rice sample comprises the following steps:
measuring near infrared information of the rice sample;
performing spectrum pretreatment on the near-infrared information of the rice sample to obtain pretreated near-infrared information;
based on the cadmium content of the rice sample and the preprocessed near-infrared information, performing pre-modeling by using a backward interval least square method to obtain a series of pre-models; during the pre-modeling, parameters of a backward interval least square method comprise: the number n of the divided intervals is 15-25, and the number LVs of the main factors is 1-10;
screening the pre-model with the minimum cross validation root mean square error in the series of pre-models;
and extracting spectral information of the pre-model with the minimum cross validation root mean square error as cadmium related characteristic near-infrared information of the rice sample.
2. The method of establishing according to claim 1, wherein the determining the cadmium content of the rice sample comprises the steps of:
mixing a rice sample and a digestion reagent, and digesting to obtain a digestion solution;
and determining the digestion solution by adopting an atomic absorption spectrometer to obtain the cadmium content of the rice sample.
3. The establishing method according to claim 2, characterized in that the digesting agent is a mixed acid of nitric acid and perchloric acid; the mass concentration of the nitric acid is 65-68%; the mass concentration of the perchloric acid is 70-72%; the volume ratio of the nitric acid to the perchloric acid is 9: 1; the material-liquid ratio of the rice sample to the digestion reagent is (0.3-0.5) g: (8-10) mL.
4. The establishing method according to claim 2 or 3, characterized in that the digestion temperature is 180 ℃ and the digestion time is 2-2.5 h.
5. The establishing method according to claim 4, characterized by further comprising heating the obtained digestion system to 220 ℃ to expel acid for 1-1.5 h after digestion is finished until the digestion system is clear and transparent.
6. The method of establishing as claimed in claim 2, wherein the parameters of the atomic absorption spectrometer comprise: light source: a hollow cathode lamp; lamp current: 50 percent; wavelength: 228.8 nm; pass band width: 0.5 nm; drying temperature: 120 ℃/20 s; ashing temperature: 800 ℃/20 s; atomization temperature: 1300 ℃/3 s; clearing: 2500 deg.C/3 s.
7. The method of establishing of claim 1, wherein determining the parameters of the near infrared information of the rice sample comprises: the sampling mode is diffuse reflection of an integrating sphere; the scanning range is 4000-12000 cm-1(ii) a Resolution was 16cm-1(ii) a The number of scans was 64; the number of sample points is 1154.
8. The method of creating claim 1, wherein the spectral preprocessing method is a first derivative method.
9. A method for predicting cadmium content in rice is characterized by comprising the following steps:
measuring near infrared information of rice to be measured;
substituting the near-infrared information of the rice to be detected into the near-infrared model obtained by the establishing method of any one of claims 1-8 to obtain the cadmium content of the rice to be detected.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011525736.2A CN112748085A (en) | 2020-12-22 | 2020-12-22 | Method for establishing near-infrared model for predicting cadmium content in rice and method for predicting cadmium content in rice |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011525736.2A CN112748085A (en) | 2020-12-22 | 2020-12-22 | Method for establishing near-infrared model for predicting cadmium content in rice and method for predicting cadmium content in rice |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112748085A true CN112748085A (en) | 2021-05-04 |
Family
ID=75648649
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011525736.2A Pending CN112748085A (en) | 2020-12-22 | 2020-12-22 | Method for establishing near-infrared model for predicting cadmium content in rice and method for predicting cadmium content in rice |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112748085A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102830072A (en) * | 2012-08-13 | 2012-12-19 | 中国计量学院 | Identification method for rice leaves contaminated by heavy metals based on near infrared spectroscopy |
CN104089926A (en) * | 2014-07-18 | 2014-10-08 | 湖南省食品测试分析中心 | NIR (Near Infrared Ray) spectral analysis model and method for identifying excessive content of cadmium in rice |
CN104374731A (en) * | 2014-11-18 | 2015-02-25 | 湖南省食品测试分析中心 | MIR spectral analysis model and method for identifying excessive cadmium content of rice. |
CN108519339A (en) * | 2018-03-26 | 2018-09-11 | 江苏大学 | A kind of blade cadmium content Vis-NIR spectral signature modeling methods based on WT-LSSVR |
-
2020
- 2020-12-22 CN CN202011525736.2A patent/CN112748085A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102830072A (en) * | 2012-08-13 | 2012-12-19 | 中国计量学院 | Identification method for rice leaves contaminated by heavy metals based on near infrared spectroscopy |
CN104089926A (en) * | 2014-07-18 | 2014-10-08 | 湖南省食品测试分析中心 | NIR (Near Infrared Ray) spectral analysis model and method for identifying excessive content of cadmium in rice |
CN104374731A (en) * | 2014-11-18 | 2015-02-25 | 湖南省食品测试分析中心 | MIR spectral analysis model and method for identifying excessive cadmium content of rice. |
CN108519339A (en) * | 2018-03-26 | 2018-09-11 | 江苏大学 | A kind of blade cadmium content Vis-NIR spectral signature modeling methods based on WT-LSSVR |
Non-Patent Citations (4)
Title |
---|
PENGCHENG NIE ET AL: "The Effects of Drying Temperature on Nitrogen Concentration Detection in Calcium Soil Studied by NIR Spectroscopy", 《APPL. SCI.》 * |
中华人民共和国国家卫生和计划生育委员会: "GB5009.15-2014食品安全国家标准食品中镉的测定", 《中华人民共和国国家标准》 * |
朱向荣等: "基于近红外光谱与组合间隔偏最小二乘法的稻米镉含量快速检测", 《食品与机械》 * |
肖朝耿等: "基于DR-FTIR技术结合BIPLS法", 《中国粮油学报》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101915744B (en) | Near infrared spectrum nondestructive testing method and device for material component content | |
Cuadrado et al. | Comparison and joint use of near infrared spectroscopy and Fourier transform mid infrared spectroscopy for the determination of wine parameters | |
CN109085136B (en) | Method for measuring content of oxide components in cement raw material by near-infrared diffuse reflection spectrum | |
Pan et al. | In situ monitoring of total polyphenols content during tea extract oxidation using a portable spectroscopy system with variables selection algorithms | |
Wang et al. | Onsite nutritional diagnosis of tea plants using micro near-infrared spectrometer coupled with chemometrics | |
Chen et al. | A novel variable selection method based on stability and variable permutation for multivariate calibration | |
Pozza et al. | Integration of vis-NIR and pXRF spectroscopy for rapid measurement of soil lead concentrations | |
Wang et al. | Multielement Principal Component Analysis and Origin Traceability of Rice Based on ICP‐MS/MS | |
CN103743705A (en) | Rapid detection method for sorghum halepense and similar species | |
Zhao et al. | Laser-induced breakdown spectroscopy (LIBS) for the detection of exogenous contamination of metal elements in lily bulbs | |
CN103134770B (en) | Eliminate moisture detects total nitrogen content of soil impact method near infrared spectrum | |
CN103760130A (en) | Method for measuring tween-80 content in compound musk injection by using near-infrared spectrum | |
CN112630180B (en) | Ultraviolet/visible light absorption spectrum model for detecting concentration of organophosphorus pesticide in water body | |
Xie et al. | Prediction of acrylamide content in potato chips using near-infrared spectroscopy | |
CN112748085A (en) | Method for establishing near-infrared model for predicting cadmium content in rice and method for predicting cadmium content in rice | |
Fan et al. | Near-infrared spectroscopy and chemometric modelling for rapid diagnosis of kidney disease | |
Liu et al. | Research on the online rapid sensing method of moisture content in famous green tea spreading | |
La et al. | Fusion of spectral and electrochemical sensor data for estimating soil macronutrients | |
CN101140225B (en) | Method for detecting lead in scenting agent with AOTF near-infrared spectrometer | |
CN104181125A (en) | Method for rapidly determining Kol-bach value of beer malt | |
CN109520950A (en) | The insensitive chemical component spectral method of detection of a kind of pair of spectral shift | |
CN115615976A (en) | Water quality chemical oxygen demand detection method based on laser spectrum technology | |
CN113075197A (en) | Method for rapidly detecting adulteration content of corn flour in white pepper powder | |
CN115078327A (en) | Rapid detection method for dangerous chemicals based on neural network architecture search | |
Tang et al. | Combining Near‐Infrared Spectroscopy and Chemometrics for Rapid Recognition of an Hg‐Contaminated Plant |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210504 |