CN108362659B - Edible oil type rapid identification method based on multi-source spectrum parallel fusion - Google Patents
Edible oil type rapid identification method based on multi-source spectrum parallel fusion Download PDFInfo
- Publication number
- CN108362659B CN108362659B CN201810120412.7A CN201810120412A CN108362659B CN 108362659 B CN108362659 B CN 108362659B CN 201810120412 A CN201810120412 A CN 201810120412A CN 108362659 B CN108362659 B CN 108362659B
- Authority
- CN
- China
- Prior art keywords
- edible oil
- raman
- spectrogram
- infrared
- training set
- 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.)
- Active
Links
- 239000008157 edible vegetable oil Substances 0.000 title claims abstract description 175
- 238000000034 method Methods 0.000 title claims abstract description 73
- 238000001228 spectrum Methods 0.000 title claims abstract description 22
- 230000004927 fusion Effects 0.000 title claims abstract description 14
- 238000001069 Raman spectroscopy Methods 0.000 claims abstract description 142
- 238000007781 pre-processing Methods 0.000 claims abstract description 6
- 238000012549 training Methods 0.000 claims description 108
- 230000003287 optical effect Effects 0.000 claims description 38
- 238000000605 extraction Methods 0.000 claims description 24
- 241000894007 species Species 0.000 claims description 20
- 238000001237 Raman spectrum Methods 0.000 claims description 15
- 230000002860 competitive effect Effects 0.000 claims description 15
- 238000005070 sampling Methods 0.000 claims description 15
- 238000002790 cross-validation Methods 0.000 claims description 14
- 238000002329 infrared spectrum Methods 0.000 claims description 12
- 238000005457 optimization Methods 0.000 claims description 12
- 238000012614 Monte-Carlo sampling Methods 0.000 claims description 11
- 230000006870 function Effects 0.000 claims description 10
- 230000003595 spectral effect Effects 0.000 claims description 9
- 230000003044 adaptive effect Effects 0.000 claims description 8
- 238000009499 grossing Methods 0.000 claims description 8
- 238000010438 heat treatment Methods 0.000 claims description 8
- 230000010354 integration Effects 0.000 claims description 8
- 238000000354 decomposition reaction Methods 0.000 claims description 7
- 239000003921 oil Substances 0.000 claims description 6
- 235000019198 oils Nutrition 0.000 claims description 6
- 230000002452 interceptive effect Effects 0.000 claims description 5
- 239000002245 particle Substances 0.000 claims description 5
- 238000012706 support-vector machine Methods 0.000 claims description 5
- 238000012795 verification Methods 0.000 claims description 5
- 238000012937 correction Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 3
- 240000007594 Oryza sativa Species 0.000 claims description 2
- 235000007164 Oryza sativa Nutrition 0.000 claims description 2
- 235000019483 Peanut oil Nutrition 0.000 claims description 2
- 235000019484 Rapeseed oil Nutrition 0.000 claims description 2
- 239000010495 camellia oil Substances 0.000 claims description 2
- 235000005687 corn oil Nutrition 0.000 claims description 2
- 239000002285 corn oil Substances 0.000 claims description 2
- 239000004006 olive oil Substances 0.000 claims description 2
- 235000008390 olive oil Nutrition 0.000 claims description 2
- 239000000312 peanut oil Substances 0.000 claims description 2
- 235000009566 rice Nutrition 0.000 claims description 2
- 239000003549 soybean oil Substances 0.000 claims description 2
- 235000012424 soybean oil Nutrition 0.000 claims description 2
- 235000020238 sunflower seed Nutrition 0.000 claims description 2
- IPCSVZSSVZVIGE-UHFFFAOYSA-N hexadecanoic acid Chemical compound CCCCCCCCCCCCCCCC(O)=O IPCSVZSSVZVIGE-UHFFFAOYSA-N 0.000 description 2
- WRIDQFICGBMAFQ-UHFFFAOYSA-N (E)-8-Octadecenoic acid Natural products CCCCCCCCCC=CCCCCCCC(O)=O WRIDQFICGBMAFQ-UHFFFAOYSA-N 0.000 description 1
- LQJBNNIYVWPHFW-UHFFFAOYSA-N 20:1omega9c fatty acid Natural products CCCCCCCCCCC=CCCCCCCCC(O)=O LQJBNNIYVWPHFW-UHFFFAOYSA-N 0.000 description 1
- ZCYVEMRRCGMTRW-UHFFFAOYSA-N 7553-56-2 Chemical compound [I] ZCYVEMRRCGMTRW-UHFFFAOYSA-N 0.000 description 1
- QSBYPNXLFMSGKH-UHFFFAOYSA-N 9-Heptadecensaeure Natural products CCCCCCCC=CCCCCCCCC(O)=O QSBYPNXLFMSGKH-UHFFFAOYSA-N 0.000 description 1
- OYHQOLUKZRVURQ-HZJYTTRNSA-N Linoleic acid Chemical compound CCCCC\C=C/C\C=C/CCCCCCCC(O)=O OYHQOLUKZRVURQ-HZJYTTRNSA-N 0.000 description 1
- ZQPPMHVWECSIRJ-UHFFFAOYSA-N Oleic acid Natural products CCCCCCCCC=CCCCCCCCC(O)=O ZQPPMHVWECSIRJ-UHFFFAOYSA-N 0.000 description 1
- 239000005642 Oleic acid Substances 0.000 description 1
- 235000021314 Palmitic acid Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 235000005911 diet Nutrition 0.000 description 1
- 230000037213 diet Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 235000013675 iodine Nutrition 0.000 description 1
- 229910052740 iodine Inorganic materials 0.000 description 1
- 239000011630 iodine Substances 0.000 description 1
- QXJSBBXBKPUZAA-UHFFFAOYSA-N isooleic acid Natural products CCCCCCCC=CCCCCCCCCC(O)=O QXJSBBXBKPUZAA-UHFFFAOYSA-N 0.000 description 1
- OYHQOLUKZRVURQ-IXWMQOLASA-N linoleic acid Natural products CCCCC\C=C/C\C=C\CCCCCCCC(O)=O OYHQOLUKZRVURQ-IXWMQOLASA-N 0.000 description 1
- 235000020778 linoleic acid Nutrition 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- WQEPLUUGTLDZJY-UHFFFAOYSA-N n-Pentadecanoic acid Natural products CCCCCCCCCCCCCCC(O)=O WQEPLUUGTLDZJY-UHFFFAOYSA-N 0.000 description 1
- 235000015097 nutrients Nutrition 0.000 description 1
- 235000016709 nutrition Nutrition 0.000 description 1
- ZQPPMHVWECSIRJ-KTKRTIGZSA-N oleic acid Chemical compound CCCCCCCC\C=C/CCCCCCCC(O)=O ZQPPMHVWECSIRJ-KTKRTIGZSA-N 0.000 description 1
- 235000021313 oleic acid Nutrition 0.000 description 1
- 238000011160 research Methods 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
- G01N21/3577—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry using infrared light for analysing liquids, e.g. polluted water
-
- 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
-
- 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
- G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/65—Raman scattering
Abstract
The invention discloses a method for rapidly identifying the type of edible oil based on multi-source spectrum parallel fusion, which comprises the following steps: selecting a plurality of edible oil samples of at least one kind; respectively collecting a near infrared spectrogram and a Raman spectrogram of a plurality of edible oil samples; respectively preprocessing the near-infrared spectrogram and the Raman spectrogram of a plurality of edible oil samples to obtain preprocessed near-infrared spectrogram and preprocessed Raman spectrogram of the edible oil samples; respectively extracting characteristic variables of the preprocessed near-infrared spectrogram and the preprocessed Raman spectrogram of the edible oil samples to obtain characteristic variables of the near-infrared spectrogram and the Raman spectrogram of the edible oil samples; establishing a qualitative model; and (5) identifying the species. The method for rapidly identifying the type of the edible oil based on the multi-source spectrum parallel fusion is safe and rapid, is convenient and fast to detect, has high identification accuracy, and has high practical value and popularization value.
Description
Technical Field
The invention relates to the technical field of edible oil rapid identification, in particular to an edible oil type rapid identification method based on multi-source spectrum parallel fusion.
Background
The edible oil contains a plurality of important nutrient components required by human body, which are indispensable in our daily diet life, and researches show that part of the edible oil is rich in a plurality of characteristic components, including palmitic acid, oleic acid, linoleic acid and iodine value, and different kinds of edible oil have different contents of a plurality of characteristic components; in addition, the content of various characteristic components mainly determines the nutritional value of the edible oil, so the content is generally used as an important component index for evaluating the quality of the edible oil and is also an important basis for determining the commercial value of the edible oil. With the continuous increase of the price of the edible oil, in order to gain violence, many illegal merchants launch low-value oil as high-value oil to market or blend the low-value oil into the high-value oil, so that the benefits of consumers and legal production and sales enterprises are seriously harmed. Therefore, there is a need to develop a method for rapidly identifying the type of edible oil, which is important for maintaining the benefits of consumers and legal operators and maintaining the normal order of the edible oil market.
Disclosure of Invention
Aiming at the defects in the technology, the invention provides a safe, reliable, convenient and efficient edible oil type rapid identification method based on multi-source spectrum parallel fusion.
The technical scheme adopted by the invention for solving the technical problems is as follows: a method for rapidly identifying the type of edible oil based on multi-source spectrum parallel fusion comprises the following steps: step one, selecting a sample: selecting a plurality of edible oil samples of at least one kind; step two, spectrum collection: respectively collecting a near infrared spectrogram and a Raman spectrogram of a plurality of edible oil samples; step three, spectrum pretreatment: respectively preprocessing the near-infrared spectrogram and the Raman spectrogram of a plurality of edible oil samples to obtain preprocessed near-infrared spectrogram and preprocessed Raman spectrogram of the edible oil samples; step four, extracting characteristic variables: respectively extracting characteristic variables of the preprocessed near-infrared spectrogram and the preprocessed Raman spectrogram of the edible oil samples to obtain characteristic variables of the near-infrared spectrogram and the Raman spectrogram of the edible oil samples; step five, establishing a qualitative model: dividing a plurality of edible oil samples into a training set and a testing set, and sequentially establishing a near-infrared training set qualitative model and a Raman training set qualitative model according to the near-infrared spectrogram characteristic variable and the Raman spectrogram characteristic variable of the edible oil samples in the training set; step six, species identification: respectively adopting a near infrared training set qualitative model and a Raman training set qualitative model to identify the types of the edible oil samples in the prediction set; putting the edible oil samples with the same type identification and correct type identification of the edible oil samples in the prediction set by the near-infrared training set qualitative model and the Raman training set qualitative model into the training set in the fifth step, and rebuilding the near-infrared training set qualitative model and the Raman training set qualitative model; performing species identification on the residual edible oil samples in the prediction set by using the reestablished near-infrared training set qualitative model and Raman training set qualitative model; and circulating the steps until the final near-infrared training set qualitative model and the final Raman training set qualitative model are different in type identification and incorrect in type identification of the edible oil sample finally remaining in the prediction set.
Preferably, the acquisition process of the near infrared spectrum in the second step is as follows: placing an edible oil sample to be collected in a sample pool, setting the measuring range of a near-infrared spectrometer to be 1350-1800 nm, in the first stage, heating the edible oil sample to the temperature of 20 ℃, and collecting to obtain a first near-infrared spectrogram; in the second stage, the edible oil sample is heated to 40 ℃, and a second near-infrared spectrogram is acquired; in the third stage, the edible oil sample is heated to 60 ℃, and a third near-infrared spectrogram is obtained through collection; and finally, taking the average value of the first near-infrared spectrogram, the second near-infrared spectrogram and the third near-infrared spectrogram as the near-infrared spectrogram of the edible oil sample.
Preferably, in the second step, in the first stage of near infrared spectrum acquisition, the scanning frequency of the near infrared spectrometer is set to be 32 times, and the resolution is set to be 3.5cm-1Placing an optical path insert with an optical path of 5mm in a sample cell; the scanning times of the near infrared spectrometer are set to be 64 times in the second stage of near infrared spectrum acquisitionResolution of 3cm-1Placing an optical path insert with an optical path of 15mm in a sample cell; in the third stage of near infrared spectrum collection, the scanning times of the near infrared spectrometer are set to be 16 times, and the resolution is set to be 4cm-1An optical path insert with an optical path of 10mm is arranged in the sample cell.
Preferably, the raman spectrum acquisition process in the second step is as follows: placing an edible oil sample to be collected in a sample cell, and setting the measuring range of a Raman spectrometer to be 780-1780 cm-1In the first stage, the edible oil sample is heated to 20 ℃, and a first Raman spectrogram is acquired; in the second stage, the edible oil sample is heated to 40 ℃, and a second Raman spectrogram is obtained through collection; in the third stage, the edible oil sample is heated to 60 ℃, and a third Raman spectrogram is obtained by collection; and finally, taking the average value of the first Raman spectrogram, the second Raman spectrogram and the third Raman spectrogram as the Raman spectrogram of the edible oil sample.
Preferably, in the first stage of the raman spectrum collection in the second step, the integration time of the raman spectrometer is set to be 10 seconds, the laser power is 220MW, and an optical path insert with an optical path of 5mm is placed in the sample cell; in the second stage of Raman spectrum collection, setting the integration time of a Raman spectrometer to be 30 seconds, setting the laser power to be 320MW, and placing an optical path insert with an optical path of 15mm in a sample cell; in the third stage of Raman spectrum collection, the integration time of a Raman spectrometer is set to be 20 seconds, the laser power is 270MW, and an optical path insert with an optical path of 10mm is arranged in a sample cell.
Preferably, the spectral noise is eliminated by carrying out a moving average 11-point smoothing method on the near-infrared spectrogram of the edible oil sample, and baseline correction is carried out by adopting a self-adaptive iterative reweighed penalty least square algorithm to obtain a preprocessed near-infrared spectrogram of the edible oil sample; carrying out Savitzky-Golay filtering on the Raman spectrogram of the edible oil sample in sequence, carrying out 9-point smoothing to eliminate spectral noise, and carrying out 1445cm-1And performing normalization treatment by taking the characteristic peak intensity as a reference to obtain a pretreatment Raman spectrogram of the edible oil sample.
Preferably, the method for extracting the feature variables in the fourth step is as follows: firstly, performing primary characteristic variable extraction on a preprocessed near-infrared spectrogram and a preprocessed Raman spectrogram of the edible oil sample by adopting a competitive self-adaptive re-weighting sampling method to obtain primary near-infrared spectrogram characteristic variable and primary Raman spectrogram characteristic variable; and then, performing secondary characteristic variable extraction on the primary near-infrared spectrogram characteristic variable and the primary Raman spectrogram characteristic variable by adopting sparse dictionary learning to obtain the near-infrared spectrogram characteristic variable and the Raman spectrogram characteristic variable of the edible oil sample.
Preferably, a competitive self-adaptive re-weighting sampling method is adopted to extract characteristic variables of the preprocessed near-infrared spectrogram and the preprocessed Raman spectrogram of the edible oil sample for one time, a 10-fold partial least square method cross validation modeling is adopted when a wavelength variable subset is selected, Monte Carlo sampling times are set to be 1-50 times, and when the root mean square error value of a partial least square method cross validation model is minimum, the characteristic variables of the preprocessed near-infrared spectrogram and the characteristic variables of the preprocessed Raman spectrogram are obtained.
Preferably, the first near-infrared spectrogram characteristic variable and the first raman spectrogram characteristic variable are subjected to secondary characteristic variable extraction by sparse dictionary learning, a redundant dictionary parameter k is set to be 16 × 16 × 3, and a balance error parameter is set
And setting a weight parameter lambda as 30/sigma, sequentially adopting an orthogonal matching pursuit method and a K-singular value decomposition method to carry out iterative optimization, and setting the iteration times to be 2-20 times to obtain the near-infrared spectrogram characteristic variable and the Raman spectrogram characteristic variable of the edible oil sample.
Preferably, the establishing process of the qualitative model in the step five is as follows: respectively taking near-infrared spectrogram characteristic variables and Raman spectrogram characteristic variables of the edible oil sample in the training set as input variables of a qualitative model, and sequentially establishing a near-infrared training set qualitative model and a Raman training set qualitative model by a support vector machine classification method; performing qualitative model on near infrared training set and Raman training set by adopting particle swarm optimization algorithmThe penalty factor C and the kernel function parameter g in the model are optimized, and the ranges of the penalty factor C and the kernel function parameter g are both set to be 2-20～220And setting the range of the interactive verification parameter V to be 2-10.
Compared with the prior art, the invention has the beneficial effects that: according to the edible oil type rapid identification method based on multi-source spectrum parallel fusion, the influence of scanning times, resolution, temperature and optical path on a near-infrared spectrogram can be effectively eliminated by adopting staged near-infrared spectrum acquisition; by adopting staged Raman spectrum collection, the influence of temperature, optical path and self performance of a Raman spectrometer on a Raman spectrogram can be effectively eliminated; extracting characteristic variables by a competitive self-adaptive re-weighting sampling method and sparse dictionary learning in sequence, and effectively compressing useless variables and interference information in near infrared spectrum and Raman spectrum; by sequentially establishing the near-infrared training set qualitative model and the Raman training set qualitative model and adopting the parallel fusion thought to identify the species, the identification accuracy is greatly improved.
Drawings
FIG. 1 is a flow chart of the edible oil type rapid identification method based on multi-source spectrum parallel fusion according to the invention;
FIG. 2 is a near infrared spectrum of a 157 portion edible oil sample according to the present invention;
FIG. 3 is a Raman spectrum of a 157 parts edible oil sample according to the present invention.
Detailed Description
The present invention is further described in detail below with reference to the attached drawings so that those skilled in the art can implement the invention by referring to the description text.
As shown in FIG. 1, the invention provides a method for rapidly identifying the type of edible oil based on multi-source spectrum parallel fusion, which comprises the following steps:
step one, selecting a sample: several samples of edible oils of at least one kind are selected.
Step two, spectrum collection: respectively collecting a near infrared spectrogram and a Raman spectrogram of a plurality of edible oil samples;
wherein, it is near redThe external spectrum was collected as follows: placing an edible oil sample to be collected in a sample cell, setting the measuring range of a near-infrared spectrometer to be 1350-1800 nm, and in the first stage, setting the scanning times of the near-infrared spectrometer to be 32 times and the resolution to be 3.5cm-1Placing an optical path insert with an optical path of 5mm in a sample pool, heating the edible oil sample to the temperature of 20 ℃, and collecting to obtain a first near-infrared spectrogram; in the second stage, the scanning times of the near infrared spectrometer are set to be 64 times, and the resolution is set to be 3cm-1Placing an optical path insert with an optical path of 15mm in a sample pool, heating the edible oil sample to 40 ℃, and collecting to obtain a second near-infrared spectrogram; in the third stage, the scanning times of the near-infrared spectrometer are set to be 16 times, and the resolution is set to be 4cm-1Placing an optical path insert with an optical path of 10mm in a sample pool, heating the edible oil sample to 60 ℃, and collecting to obtain a third near-infrared spectrogram; and finally, taking the average value of the first near-infrared spectrogram, the second near-infrared spectrogram and the third near-infrared spectrogram as the near-infrared spectrogram of the edible oil sample.
Wherein, the Raman spectrum acquisition process is as follows: placing an edible oil sample to be collected in a sample cell, and setting the measuring range of a Raman spectrometer to be 780-1780 cm-1In the first stage, setting the integration time of a Raman spectrometer to be 10 seconds, setting the laser power to be 220MW, placing an optical path insert with an optical path of 5mm in a sample pool, heating an edible oil sample to the temperature of 20 ℃, and collecting to obtain a first Raman spectrogram; in the second stage, setting the integration time of the Raman spectrometer to be 30 seconds, setting the laser power to be 320MW, placing an optical path insert with an optical path of 15mm in a sample pool, heating the edible oil sample to the temperature of 40 ℃, and collecting to obtain a second Raman spectrogram; in the third stage of Raman spectrum collection, setting the integration time of a Raman spectrometer to be 20 seconds, setting the laser power to be 270MW, placing an optical path insert with an optical path of 10mm in a sample pool, heating the edible oil sample to the temperature of 60 ℃, and collecting to obtain a third Raman spectrum; and finally, taking the average value of the first Raman spectrogram, the second Raman spectrogram and the third Raman spectrogram as the Raman spectrogram of the edible oil sample.
Step three, spectrum pretreatment: respectively preprocessing the near-infrared spectrogram and the Raman spectrogram of a plurality of edible oil samples, eliminating spectral noise by a moving average 11-point smoothing method on the near-infrared spectrogram of the edible oil samples, and performing baseline correction by adopting a self-adaptive iterative reweighed punishment least square algorithm to obtain the preprocessed near-infrared spectrogram of the edible oil samples; carrying out Savitzky-Golay filtering on the Raman spectrogram of the edible oil sample in sequence, carrying out 9-point smoothing to eliminate spectral noise, and carrying out 1445cm-1And performing normalization treatment by taking the characteristic peak intensity as a reference to obtain a pretreatment Raman spectrogram of the edible oil sample.
Step four, extracting characteristic variables: respectively extracting characteristic variables of the preprocessed near-infrared spectrogram and the preprocessed Raman spectrogram of a plurality of edible oil samples, and firstly, extracting the characteristic variables of the preprocessed near-infrared spectrogram and the preprocessed Raman spectrogram of the edible oil samples for one time by adopting a competitive adaptive re-weighting sampling method to obtain characteristic variables of the first near-infrared spectrogram and the first Raman spectrogram; and then, performing secondary characteristic variable extraction on the primary near-infrared spectrogram characteristic variable and the primary Raman spectrogram characteristic variable by adopting sparse dictionary learning to obtain the near-infrared spectrogram characteristic variable and the Raman spectrogram characteristic variable of the edible oil sample.
The method comprises the steps of performing primary characteristic variable extraction on a preprocessed near-infrared spectrogram and a preprocessed Raman spectrogram of an edible oil sample by adopting a competitive self-adaptive re-weighting sampling method, selecting a wavelength variable subset, performing cross validation modeling by adopting a 10-fold partial least square method, setting the Monte Carlo sampling frequency to be 1-50 times, and obtaining the primary near-infrared spectrogram characteristic variable and the primary Raman spectrogram characteristic variable when the root mean square error value of a partial least square method cross validation model is minimum.
Performing secondary characteristic variable extraction on the primary near-infrared spectrogram characteristic variable and the primary Raman spectrogram characteristic variable by adopting sparse dictionary learning, setting a redundant dictionary parameter k to be 16 multiplied by 3, and setting a balance error parameter
And setting a weight parameter lambda as 30/sigma, sequentially adopting an orthogonal matching pursuit method and a K-singular value decomposition method to carry out iterative optimization, and setting the iteration times to be 2-20 times to obtain the near-infrared spectrogram characteristic variable and the Raman spectrogram characteristic variable of the edible oil sample.
Step five, establishing a qualitative model: dividing a plurality of edible oil samples into a training set and a testing set, respectively taking near-infrared spectrogram characteristic variables and Raman spectrogram characteristic variables of the edible oil samples in the training set as input variables of a qualitative model, and sequentially establishing a near-infrared training set qualitative model and a Raman training set qualitative model by a support vector machine classification method; and then, optimizing a penalty factor C and a kernel function parameter g in the near-infrared training set qualitative model and the Raman training set qualitative model by adopting a particle swarm optimization algorithm, wherein the ranges of the penalty factor C and the kernel function parameter g are both set to be 2 during optimization-20～220And setting the range of the interactive verification parameter V to be 2-10.
Step six, species identification: respectively adopting a near infrared training set qualitative model and a Raman training set qualitative model to identify the types of the edible oil samples in the prediction set; putting the edible oil samples with the same type identification and correct type identification of the edible oil samples in the prediction set by the near-infrared training set qualitative model and the Raman training set qualitative model into the training set in the fifth step, and rebuilding the near-infrared training set qualitative model and the Raman training set qualitative model; performing species identification on the residual edible oil samples in the prediction set by using the reestablished near-infrared training set qualitative model and Raman training set qualitative model; and circulating the steps until the final near-infrared training set qualitative model and the final Raman training set qualitative model are different in type identification and incorrect in type identification of the edible oil sample finally remaining in the prediction set.
Examples
1. Sample selection
Collecting 157 parts of soybean oil, peanut oil, rapeseed oil, tea seed oil, rice oil, corn oil, sunflower seed oil and olive oil 8-class edible oil samples, and selecting 118 parts of training set edible oil samples and 39 parts of prediction set edible oil samples according to a ratio of 3:1 by adopting an SPXY algorithm; wherein, the number distribution of the prediction set samples and the training set samples of the 8 types of edible oil samples is shown in the following table 1.
TABLE 1
2. Spectrum collection
And respectively adopting a near-infrared spectrometer and a Raman spectrometer to collect a near-infrared spectrogram and a Raman spectrogram of 157 parts of the edible oil sample, wherein the spectrum collection process is as described in the step two, the near-infrared spectrogram of the 157 parts of the edible oil sample is as shown in figure 2, and the Raman spectrogram is as shown in figure 3.
3. Spectral preprocessing
Respectively preprocessing a near-infrared spectrogram and a Raman spectrogram of 157 parts of edible oil samples, eliminating spectral noise by a moving average 11-point smoothing method on the near-infrared spectrogram of the edible oil samples, and performing baseline correction by adopting a self-adaptive iterative reweighed punishment least square algorithm to obtain a preprocessed near-infrared spectrogram of the edible oil samples; carrying out Savitzky-Golay filtering on the Raman spectrogram of the edible oil sample in sequence, carrying out 9-point smoothing to eliminate spectral noise, and carrying out 1445cm-1And performing normalization treatment by taking the characteristic peak intensity as a reference to obtain a pretreatment Raman spectrogram of the edible oil sample.
4. Feature variable extraction
Performing feature extraction on the near-infrared spectrogram of 157 parts of edible oil samples by combining competitive adaptive re-weighted sampling and sparse dictionary learning: firstly, extracting characteristic variables of a near-infrared spectrogram of an edible oil sample by a competitive adaptive re-weighting sampling method for one time, selecting a wavelength variable subset, performing cross validation modeling by a 10-fold partial least square method, setting the Monte Carlo sampling frequency to be 1-50 times, and when the Monte Carlo sampling frequency is 39 times, minimizing the root mean square error value of the cross validation model by the partial least square method to be 1.4839, and extracting characteristic variables of the spectrogram of 56 times(ii) a And then, performing secondary characteristic variable extraction on the 56 primary spectrogram characteristic variables by adopting sparse dictionary learning, setting a redundant dictionary parameter k to be 16 multiplied by 3, and setting a balance error parameter
Setting a weight parameter lambda to be 30/sigma, sequentially adopting an orthogonal matching pursuit method and a K-singular value decomposition method to carry out iterative optimization, setting the iteration frequency to be 2-20 times, obtaining 0.2572 with the minimum root mean square error value when the sigma is 25 and the iteration frequency is 10 times, and obtaining 28 spectrogram characteristic variables obtained by secondary extraction.
Performing feature extraction on the Raman spectrogram of 157 parts of edible oil samples by combining competitive adaptive re-weighted sampling and sparse dictionary learning: firstly, performing primary characteristic variable extraction on a Raman spectrogram of an edible oil sample by adopting a competitive adaptive reweighting sampling method, selecting a wavelength variable subset, performing cross validation modeling by adopting a 10-fold partial least square method, setting the Monte Carlo sampling frequency to be 1-50 times, and when the Monte Carlo sampling frequency is 36 times, minimizing the root mean square error value of the cross validation model by adopting the partial least square method to be 0.5872 and extracting 68 primary spectrogram characteristic variables; and then, performing secondary characteristic variable extraction on the 68 primary spectrogram characteristic variables by adopting sparse dictionary learning, setting a redundant dictionary parameter k to be 16 multiplied by 3, and setting a balance error parameter
Setting a weight parameter lambda to be 30/sigma, sequentially adopting an orthogonal matching pursuit method and a K-singular value decomposition method to carry out iterative optimization, setting the iteration frequency to be 2-20 times, obtaining 0.3684 with the minimum root mean square error value when the sigma is 25 and the iteration frequency is 16 times, and obtaining 32 spectrogram characteristic variables obtained by secondary extraction.
5. Modeling model establishment
Respectively taking near-infrared spectrogram characteristic variable and Raman spectrogram characteristic variable of 118 edible oil samples in the training set as input variables of a qualitative model, and sequentially establishing the near-infrared training set by a support vector machine classification methodA qualitative model and a Raman training set qualitative model; the punishment factor C and the kernel function parameter g in the near-infrared training set qualitative model and the Raman training set qualitative model are optimized by adopting a particle swarm optimization algorithm, and the ranges of the punishment factor C and the kernel function parameter g are both set to be 2 during optimization-20～220And setting the range of the interactive verification parameter V to be 2-10.
6. Species discrimination
Respectively adopting a near infrared training set qualitative model and a Raman training set qualitative model to perform species identification on 39 edible oil samples in the prediction set; the obtained near-infrared training set qualitative model and the Raman training set qualitative model identify the same and correct types of the 32 edible oil samples in the prediction set, the 32 edible oil samples are put into the training set in the fifth step to obtain 150 new edible oil samples in the training set, and the near-infrared training set qualitative model and the Raman training set qualitative model are reestablished;
performing species identification on the remaining 7 edible oil samples in the prediction set by using the reestablished near-infrared training set qualitative model and the reestablished Raman training set qualitative model to obtain that the species identification of the 4 edible oil samples in the prediction set is the same and correct, putting the 4 edible oil samples into the training set in the fifth step to obtain 154 new edible oil samples in the training set, and reestablishing the near-infrared training set qualitative model and the Raman training set qualitative model;
and performing species identification on the remaining 3 edible oil samples in the prediction set by utilizing the re-established near-infrared training set qualitative model and the Raman training set qualitative model to obtain that the species identification of the 3 edible oil samples in the prediction set is different and incorrect, putting the 3 edible oil samples into a singular sample, and taking the near-infrared training set qualitative model and the Raman training set qualitative model finally established by 154 edible oil samples as the edible oil species identification model fused in parallel.
While embodiments of the invention have been described above, it is not limited to the applications set forth in the description and the embodiments, which are fully applicable in various fields of endeavor to which the invention pertains, and further modifications may readily be made by those skilled in the art, it being understood that the invention is not limited to the details shown and described herein without departing from the general concept defined by the appended claims and their equivalents.
Claims (3)
1. A method for rapidly identifying the type of edible oil based on multi-source spectrum parallel fusion is characterized by comprising the following steps:
step one, selecting a sample: selecting a plurality of edible oil samples of at least one kind;
wherein 157 parts of soybean oil, peanut oil, rapeseed oil, tea seed oil, rice oil, corn oil, sunflower seed oil and olive oil 8-class edible oil samples are collected, 118 parts of training collection edible oil samples and 39 parts of prediction collection edible oil samples are selected according to the ratio of 3:1 by adopting an SPXY algorithm;
step two, spectrum collection: respectively collecting a near infrared spectrogram and a Raman spectrogram of a plurality of edible oil samples;
wherein, a near-infrared spectrum chart and a Raman spectrum chart of 157 parts of edible oil samples are respectively collected by a near-infrared spectrometer and a Raman spectrometer;
the acquisition process of the near infrared spectrum in the second step is as follows: placing an edible oil sample to be collected in a sample pool, setting the measuring range of a near-infrared spectrometer to be 1350-1800 nm, in the first stage, heating the edible oil sample to the temperature of 20 ℃, and collecting to obtain a first near-infrared spectrogram; in the second stage, the edible oil sample is heated to 40 ℃, and a second near-infrared spectrogram is acquired; in the third stage, the edible oil sample is heated to 60 ℃, and a third near-infrared spectrogram is obtained through collection; finally, taking the average value of the first near-infrared spectrogram, the second near-infrared spectrogram and the third near-infrared spectrogram as the near-infrared spectrogram of the edible oil sample; in the second step, in the first stage of near infrared spectrum acquisition, the scanning frequency of the near infrared spectrometer is set to be 32 times, and the resolution is set to be 3.5cm-1Placing an optical path insert with an optical path of 5mm in a sample cell; in the second stage of near infrared spectrum collection, the scanning times of the near infrared spectrometer are set to be 64 times, and the resolution is set to be 3cm-1Placing an optical path insert with an optical path of 15mm in a sample cell; in the third stage of near infrared spectrum collection, the scanning times of the near infrared spectrometer are set to be 16 times, and the resolution is set to be 4cm-1Placing an optical path insert with an optical path of 10mm in a sample cell;
the Raman spectrum acquisition process in the second step is as follows: placing an edible oil sample to be collected in a sample cell, and setting the measuring range of a Raman spectrometer to be 780-1780 cm-1In the first stage, the edible oil sample is heated to 20 ℃, and a first Raman spectrogram is acquired; in the second stage, the edible oil sample is heated to 40 ℃, and a second Raman spectrogram is obtained through collection; in the third stage, the edible oil sample is heated to 60 ℃, and a third Raman spectrogram is obtained by collection; finally, taking the average value of the first Raman spectrogram, the second Raman spectrogram and the third Raman spectrogram as the Raman spectrogram of the edible oil sample; in the second step, in the first stage of Raman spectrum collection, the integral time of a Raman spectrometer is set to be 10 seconds, the laser power is 220MW, and an optical path insert with an optical path of 5mm is arranged in a sample cell; in the second stage of Raman spectrum collection, setting the integration time of a Raman spectrometer to be 30 seconds, setting the laser power to be 320MW, and placing an optical path insert with an optical path of 15mm in a sample cell; in the third stage of Raman spectrum collection, setting the integration time of a Raman spectrometer to be 20 seconds, setting the laser power to be 270MW, and placing an optical path insert with an optical path of 10mm in a sample cell;
step three, spectrum pretreatment: respectively preprocessing the near-infrared spectrogram and the Raman spectrogram of a plurality of edible oil samples to obtain preprocessed near-infrared spectrogram and preprocessed Raman spectrogram of the edible oil samples;
step four, extracting characteristic variables: respectively extracting characteristic variables of the preprocessed near-infrared spectrogram and the preprocessed Raman spectrogram of the edible oil samples to obtain characteristic variables of the near-infrared spectrogram and the Raman spectrogram of the edible oil samples; firstly, adopting a competitive self-adaptive re-weighting sampling method to carry out pretreatment near infrared spectrogram and pretreatment Raman spectrogram on the edible oil samplePerforming primary characteristic variable extraction to obtain primary near-infrared spectrogram characteristic variable and primary Raman spectrogram characteristic variable; secondly, performing secondary characteristic variable extraction on the primary near-infrared spectrogram characteristic variable and the primary Raman spectrogram characteristic variable by adopting sparse dictionary learning to obtain the near-infrared spectrogram characteristic variable and the Raman spectrogram characteristic variable of the edible oil sample; performing primary characteristic variable extraction on the preprocessed near-infrared spectrogram and the preprocessed Raman spectrogram of the edible oil sample by adopting a competitive self-adaptive re-weighting sampling method, selecting a wavelength variable subset, performing cross validation modeling by adopting a 10-fold partial least square method, setting the Monte Carlo sampling frequency to be 1-50 times, and obtaining the primary near-infrared spectrogram characteristic variable and the primary Raman spectrogram characteristic variable when the root mean square error value of the partial least square method cross validation model is minimum; performing secondary characteristic variable extraction on the primary near-infrared spectrogram characteristic variable and the primary Raman spectrogram characteristic variable by adopting sparse dictionary learning, setting a redundant dictionary parameter k to be 16 multiplied by 3, and setting a balance error parameter
Setting a weight parameter lambda as 30/sigma, sequentially adopting an orthogonal matching pursuit method and a K-singular value decomposition method to carry out iterative optimization, and setting the iteration times to be 2-20 times to obtain the near-infrared spectrogram characteristic variable and the Raman spectrogram characteristic variable of the edible oil sample;
the method comprises the following steps of performing feature extraction on a near-infrared spectrogram of 157 parts of edible oil samples by combining competitive adaptive re-weighted sampling and sparse dictionary learning: firstly, extracting characteristic variables of a near-infrared spectrogram of an edible oil sample by adopting a competitive self-adaptive re-weighting sampling method for one time, selecting a wavelength variable subset, adopting a 10-fold partial least square method for cross validation modeling, setting the Monte Carlo sampling frequency to be 1-50 times, and when the Monte Carlo sampling frequency is 39 times, minimizing the root mean square error value of the partial least square method cross validation model to be 1.4839, and extracting 56 characteristic variables of the primary spectrogram; then, performing secondary feature on the 56 primary spectrogram feature variables by adopting sparse dictionary learningExtracting variables, setting a redundant dictionary parameter k to be 16 multiplied by 3, and setting a balance error parameter
Setting a weight parameter lambda to be 30/sigma, sequentially adopting an orthogonal matching pursuit method and a K-singular value decomposition method to carry out iterative optimization, setting the iteration frequency to be 2-20 times, obtaining 0.2572 with the minimum root mean square error value when the sigma is 25 and the iteration frequency is 10 times, and obtaining 28 spectrogram characteristic variables obtained by secondary extraction;
performing feature extraction on the Raman spectrogram of 157 parts of edible oil samples by combining competitive adaptive re-weighted sampling and sparse dictionary learning: firstly, performing primary characteristic variable extraction on a Raman spectrogram of an edible oil sample by adopting a competitive adaptive reweighting sampling method, selecting a wavelength variable subset, performing cross validation modeling by adopting a 10-fold partial least square method, setting the Monte Carlo sampling frequency to be 1-50 times, and when the Monte Carlo sampling frequency is 36 times, minimizing the root mean square error value of the cross validation model by adopting the partial least square method to be 0.5872 and extracting 68 primary spectrogram characteristic variables; and then, performing secondary characteristic variable extraction on the 68 primary spectrogram characteristic variables by adopting sparse dictionary learning, setting a redundant dictionary parameter k to be 16 multiplied by 3, and setting a balance error parameter
Setting a weight parameter lambda to be 30/sigma, sequentially adopting an orthogonal matching pursuit method and a K-singular value decomposition method to carry out iterative optimization, setting the iteration frequency to be 2-20 times, obtaining 0.3684 with the minimum root mean square error value when sigma is 25 and the iteration frequency is 16 times, and obtaining 32 spectrogram characteristic variables obtained by secondary extraction;
step five, establishing a qualitative model: dividing a plurality of edible oil samples into a training set and a testing set, and sequentially establishing a near-infrared training set qualitative model and a Raman training set qualitative model according to the near-infrared spectrogram characteristic variable and the Raman spectrogram characteristic variable of the edible oil samples in the training set;
wherein the training is concentratedRespectively taking near-infrared spectrogram characteristic variables and Raman spectrogram characteristic variables of 118 edible oil samples as input variables of a qualitative model, and sequentially establishing a near-infrared training set qualitative model and a Raman training set qualitative model by a support vector machine classification method; the punishment factor C and the kernel function parameter g in the near-infrared training set qualitative model and the Raman training set qualitative model are optimized by adopting a particle swarm optimization algorithm, and the ranges of the punishment factor C and the kernel function parameter g are both set to be 2 during optimization-20～220Setting the range of the interactive verification parameter V to be 2-10;
step six, species identification:
respectively adopting a near infrared training set qualitative model and a Raman training set qualitative model to identify the types of the edible oil samples in the prediction set; putting the edible oil samples with the same type identification and correct type identification of the edible oil samples in the prediction set by the near-infrared training set qualitative model and the Raman training set qualitative model into the training set in the fifth step, and rebuilding the near-infrared training set qualitative model and the Raman training set qualitative model; performing species identification on the residual edible oil samples in the prediction set by using the reestablished near-infrared training set qualitative model and Raman training set qualitative model; the steps are circulated until the final near infrared training set qualitative model and the final Raman training set qualitative model are different in type identification and incorrect in type identification of the edible oil sample finally remaining in the prediction set;
the method comprises the following steps of (1) performing species identification on 39 edible oil samples in a prediction set by respectively adopting a near-infrared training set qualitative model and a Raman training set qualitative model; the obtained near-infrared training set qualitative model and the Raman training set qualitative model identify the same and correct types of the 32 edible oil samples in the prediction set, the 32 edible oil samples are put into the training set in the fifth step to obtain 150 new edible oil samples in the training set, and the near-infrared training set qualitative model and the Raman training set qualitative model are reestablished; performing species identification on the remaining 7 edible oil samples in the prediction set by using the reestablished near-infrared training set qualitative model and the reestablished Raman training set qualitative model to obtain that the species identification of the 4 edible oil samples in the prediction set is the same and correct, putting the 4 edible oil samples into the training set in the fifth step to obtain 154 new edible oil samples in the training set, and reestablishing the near-infrared training set qualitative model and the Raman training set qualitative model; and performing species identification on the remaining 3 edible oil samples in the prediction set by utilizing the re-established near-infrared training set qualitative model and the Raman training set qualitative model to obtain that the species identification of the 3 edible oil samples in the prediction set is different and incorrect, putting the 3 edible oil samples into a singular sample, and taking the near-infrared training set qualitative model and the Raman training set qualitative model finally established by 154 edible oil samples as the edible oil species identification model fused in parallel.
2. The method for rapidly identifying the type of the edible oil based on the multi-source spectrum parallel fusion of claim 1, wherein spectral noise is eliminated by performing a moving average 11-point smoothing method on the near-infrared spectrogram of the edible oil sample, and baseline correction is performed by adopting a self-adaptive iterative reweighted punishment least square algorithm to obtain a preprocessed near-infrared spectrogram of the edible oil sample; carrying out Savitzky-Golay filtering on the Raman spectrogram of the edible oil sample in sequence, carrying out 9-point smoothing to eliminate spectral noise, and carrying out 1445cm-1And performing normalization treatment by taking the characteristic peak intensity as a reference to obtain a pretreatment Raman spectrogram of the edible oil sample.
3. The method for rapidly identifying the type of the edible oil based on the multi-source spectrum parallel fusion as claimed in claim 1, wherein the establishment process of the qualitative model in the fifth step is as follows: respectively taking near-infrared spectrogram characteristic variables and Raman spectrogram characteristic variables of the edible oil sample in the training set as input variables of a qualitative model, and sequentially establishing a near-infrared training set qualitative model and a Raman training set qualitative model by a support vector machine classification method; the punishment factor C and the kernel function parameter g in the near-infrared training set qualitative model and the Raman training set qualitative model are optimized by adopting a particle swarm optimization algorithm, and the ranges of the punishment factor C and the kernel function parameter g are both set to be 2 during optimization-20～220And setting the range of the interactive verification parameter V to be 2-10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810120412.7A CN108362659B (en) | 2018-02-07 | 2018-02-07 | Edible oil type rapid identification method based on multi-source spectrum parallel fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810120412.7A CN108362659B (en) | 2018-02-07 | 2018-02-07 | Edible oil type rapid identification method based on multi-source spectrum parallel fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108362659A CN108362659A (en) | 2018-08-03 |
| CN108362659B true CN108362659B (en) | 2021-03-30 |
Family
ID=63004951
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810120412.7A Active CN108362659B (en) | 2018-02-07 | 2018-02-07 | Edible oil type rapid identification method based on multi-source spectrum parallel fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108362659B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108645841B (en) * | 2018-08-24 | 2020-11-03 | 武汉轻工大学 | Sesame oil multi-component adulteration detection method based on Raman spectrum wavelet fusion |
| CN110865046B (en) * | 2019-11-28 | 2022-01-07 | 浙江农林大学 | Method for rapidly detecting content of trans-fatty acid isomer of edible oil |
| CN112304922A (en) * | 2020-10-29 | 2021-02-02 | 辽宁石油化工大学 | Method for quantitatively analyzing crude oil by Raman spectrum based on partial least square method |
| CN114113035B (en) * | 2021-11-18 | 2024-02-02 | 北京理工大学 | Identification method of transgenic soybean oil |
| CN116952923B (en) * | 2023-07-27 | 2024-01-23 | 南京大学 | Machine learning-based micro-plastic on-site high-precision monitoring method and system |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1734359A1 (en) * | 2005-06-18 | 2006-12-20 | Roche Diagnostics GmbH | RAMAN spectroscopic analysis method and system therefor |
| CN101504363A (en) * | 2009-03-18 | 2009-08-12 | 哈尔滨商业大学 | Edible fatty acid value detection method based on near-infrared spectrum analysis |
| CN103398970A (en) * | 2013-07-24 | 2013-11-20 | 骆驰 | Method for qualitatively and quantitatively analyzing edible oil and further detecting hogwash oil |
| CN103959292A (en) * | 2011-09-23 | 2014-07-30 | 陶氏益农公司 | Chemometrics for near infrared spectral analysis |
| CN104374738A (en) * | 2014-10-30 | 2015-02-25 | 中国科学院半导体研究所 | Qualitative analysis method for improving identification result on basis of near-infrared mode |
| CN105021535A (en) * | 2015-08-07 | 2015-11-04 | 中南林业科技大学 | Nondestructive testing method and system of fatty acid content of rice |
| CN105588817A (en) * | 2015-12-16 | 2016-05-18 | 新希望双喜乳业（苏州）有限公司 | Milk freshness detecting method based on multisource spectroscopic data fusion |
| CN106404743A (en) * | 2016-11-01 | 2017-02-15 | 北京华泰诺安技术有限公司 | Raman spectrum and near infrared spectrum combined detection method and detection device |
| CN106706546A (en) * | 2016-12-28 | 2017-05-24 | 中山市腾创贸易有限公司 | Analysis method for artificial intelligence learning materials on basis of infrared and Raman spectrum data |
| CN107646089A (en) * | 2015-03-06 | 2018-01-30 | 英国质谱公司 | Spectrum analysis |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110178420A1 (en) * | 2010-01-18 | 2011-07-21 | Trent Ridder | Methods and apparatuses for improving breath alcohol testing |
| CN101059425A (en) * | 2007-05-29 | 2007-10-24 | 浙江大学 | Method and device for identifying different variety green tea based on multiple spectrum image texture analysis |
| CN101738383A (en) * | 2008-11-07 | 2010-06-16 | 中国农业科学院农业环境与可持续发展研究所 | Mid-infrared spectrum-based method for quickly detecting contents of heavy metal elements in soil |
| CN104048941B (en) * | 2014-06-25 | 2017-02-15 | 常熟雷允上制药有限公司 | Method for quickly measuring content of multiple index components in radix ophiopogonis through near infrared spectroscopy |
| CN104807803B (en) * | 2015-04-20 | 2017-09-29 | 武汉轻工大学 | Peanut oil based on multi-source optical spectrum data fusion mixes pseudo- quantitative detecting method |
| CN105117734B (en) * | 2015-07-28 | 2018-04-13 | 江南大学 | Corn seed classification hyperspectral imagery recognition methods based on model online updating |
| CN105823752B (en) * | 2016-03-22 | 2018-10-12 | 武汉轻工大学 | The method that near infrared spectroscopy quickly differentiates edible oil type |
| CN105806824B (en) * | 2016-03-22 | 2019-03-19 | 武汉轻工大学 | Edible oil type method for quick identification based on multi-source optical spectrum Fusion Features |
| CN107563448B (en) * | 2017-09-11 | 2020-06-23 | 广州讯动网络科技有限公司 | Sample space clustering division method based on near infrared spectrum analysis |
-
2018
- 2018-02-07 CN CN201810120412.7A patent/CN108362659B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1734359A1 (en) * | 2005-06-18 | 2006-12-20 | Roche Diagnostics GmbH | RAMAN spectroscopic analysis method and system therefor |
| CN101504363A (en) * | 2009-03-18 | 2009-08-12 | 哈尔滨商业大学 | Edible fatty acid value detection method based on near-infrared spectrum analysis |
| CN103959292A (en) * | 2011-09-23 | 2014-07-30 | 陶氏益农公司 | Chemometrics for near infrared spectral analysis |
| CN103398970A (en) * | 2013-07-24 | 2013-11-20 | 骆驰 | Method for qualitatively and quantitatively analyzing edible oil and further detecting hogwash oil |
| CN104374738A (en) * | 2014-10-30 | 2015-02-25 | 中国科学院半导体研究所 | Qualitative analysis method for improving identification result on basis of near-infrared mode |
| CN107646089A (en) * | 2015-03-06 | 2018-01-30 | 英国质谱公司 | Spectrum analysis |
| CN105021535A (en) * | 2015-08-07 | 2015-11-04 | 中南林业科技大学 | Nondestructive testing method and system of fatty acid content of rice |
| CN105588817A (en) * | 2015-12-16 | 2016-05-18 | 新希望双喜乳业（苏州）有限公司 | Milk freshness detecting method based on multisource spectroscopic data fusion |
| CN106404743A (en) * | 2016-11-01 | 2017-02-15 | 北京华泰诺安技术有限公司 | Raman spectrum and near infrared spectrum combined detection method and detection device |
| CN106706546A (en) * | 2016-12-28 | 2017-05-24 | 中山市腾创贸易有限公司 | Analysis method for artificial intelligence learning materials on basis of infrared and Raman spectrum data |
Non-Patent Citations (4)
| Title |
|---|
| Identification of Edible Oil Based on Multi-source Spectra Data Fusion;YU Yaru等;《Proceedings of the 2017 5th International Conference on Frontiers of Manufacturing Science and Measuring Technology (FMSMT 2017)》;20170430;全文 * |
| NIR-FT/Raman Spectroscopy for Nutritional Classification of Cereal Foods;Miryeong Sohn等;《Cereal Chemistry》;20051115;第82卷(第6期);全文 * |
| 基于多源光谱特征融合技术的花生油掺伪检测;涂斌等;《食品发酵与工业》;20160425;第42卷(第4期);全文 * |
| 基于近红外_拉曼光谱融合的食用油品质和品种快速检测方法研究;涂斌;《中国优秀硕士论文全文数据库 工程科技Ⅰ辑》;20170615(第06期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108362659A (en) | 2018-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108362659B (en) | Edible oil type rapid identification method based on multi-source spectrum parallel fusion | |
| Li et al. | Review of NIR spectroscopy methods for nondestructive quality analysis of oilseeds and edible oils | |
| CN105806824B (en) | Edible oil type method for quick identification based on multi-source optical spectrum Fusion Features | |
| Li et al. | A combination of chemometrics methods and GC–MS for the classification of edible vegetable oils | |
| CN102735642B (en) | Method for quickly and losslessly identifying virgin olive oil and olive-residue oil | |
| CN112782114B (en) | Method for identifying aging years of dried orange peel | |
| Li et al. | Grade identification of tieguanyin tea using fluorescence hyperspectra and different statistical algorithms | |
| CN109001181B (en) | Method for rapidly identifying type of edible oil by combining Raman spectrum typical correlation analysis | |
| Insausti et al. | Screening analysis of biodiesel feedstock using UV–vis, NIR and synchronous fluorescence spectrometries and the successive projections algorithm | |
| Ranulfi et al. | Nutritional characterization of healthy and Aphelenchoides besseyi infected soybean leaves by laser-induced breakdown spectroscopy (LIBS) | |
| Nogales-Bueno et al. | Assessment of total fat and fatty acids in walnuts using near-infrared hyperspectral imaging | |
| CN102841072A (en) | Method for identifying transgenic rice and non-transgenic rice based on NIR (Near Infrared Spectrum) | |
| CN104865222A (en) | Nondestructive testing method of content of fatty acid in peony seeds | |
| Lu et al. | Fluorescence hyperspectral image technique coupled with HSI method to predict solanine content of potatoes | |
| Liu et al. | Near-infrared prediction of edible oil frying times based on Bayesian Ridge Regression | |
| CN108303406B (en) | Camellia seed oil adulteration detection method based on Raman spectrum | |
| CN108051394B (en) | Sesame oil adulteration detection method based on near infrared spectrum | |
| CN113033066A (en) | Method for establishing near infrared spectrum identification model of sargassum fusiforme production area, strain and cultivation mode and identification method | |
| CN109142265B (en) | Method for rapidly identifying type of edible oil by near infrared spectrum wavelet fusion | |
| CN108267424B (en) | Method for rapidly identifying type of edible oil based on multiple characteristic components | |
| CN107121408A (en) | The quick nondestructive discrimination method of edible vegetable oil kind | |
| Monferrere et al. | Chemometric characterization of sunflower seeds | |
| Teye et al. | Nondestructive authentication of cocoa bean cultivars by FT-NIR spectroscopy and multivariate techniques | |
| CN112485238A (en) | Method for identifying turmeric essential oil producing area based on Raman spectrum technology | |
| de Almeida et al. | Direct NIR spectral determination of genetic improvement, light availability, and their interaction effects on chemically selected yerba-mate leaves |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |