CN109765196A - A kind of method for quick identification of atlantic salmon and rainbow trout - Google Patents
A kind of method for quick identification of atlantic salmon and rainbow trout Download PDFInfo
- Publication number
- CN109765196A CN109765196A CN201910126179.8A CN201910126179A CN109765196A CN 109765196 A CN109765196 A CN 109765196A CN 201910126179 A CN201910126179 A CN 201910126179A CN 109765196 A CN109765196 A CN 109765196A
- Authority
- CN
- China
- Prior art keywords
- rainbow trout
- atlantic salmon
- sample
- chilled
- quick identification
- 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
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses the method for quick identification of a kind of atlantic salmon and rainbow trout, belong to meat products technical field of quality detection.Method for quick identification of the present invention to atlantic salmon and rainbow trout specifically: collect atlantic salmon and rainbow trout sample and number, with the progress data acquisition of NIR technology;Data are pre-processed;Dimensionality reduction is carried out to pretreated data using PCA;The quick identification model of atlantic salmon and rainbow trout is established using chemometrics method;Data acquisition is carried out to unknown flesh of fish sample, the classification of unknown flesh of fish sample to be measured is predicted using the quick model that identifies.Method for quick identification of the invention is not limited by the subjectivity of sensory evaluation, the complexity of common detection methods, is suitble to large batch of quick detection.Atlantic salmon and rainbow trout are identified using method of the invention, when number of principal components is 10, training set and test set discrimination are respectively 98.33%, 95%, it can be achieved that quickly identifying.
Description
Technical field
The invention belongs to meat products technical field of quality detection, and in particular to a kind of atlantic salmon and rainbow trout it is quick
Discrimination method.
Background technique
In China market, salmon can be divided into the atlantic salmon of seawater and the rainbow trout of fresh water, and usually we are used as
The salmon of sashimi is all the atlantic salmon from the import of the states such as Denmark, Norway.Since the market price of rainbow trout is than big west
Low one third of foreign salmon or so, and the demand of salmon is continuously increased in the market, it is mixed just to there is fish commodity sign
Disorderly, the phenomenon that adulterating, the case where especially pretending to be atlantic salmon with rainbow trout, are in the majority.The not only serious damage of this way
Consumer's interests have also upset market order, have affected economic development.
Currently used meat discrimination method includes the technologies such as traditional organoleptic detection, immunology, molecular biology.Sense organ inspection
Survey is primarily referred to as through the eye of sense organ personnel, mouth, nose, hand etc. to items such as the color of the flesh of fish, mouthfeel, smell, shape, viscoelasticities
Index carries out comprehensive evaluation.Immunological method mainly includes the test tube precipitation method, agar diffusion method, countercurrent immunoelectrophoresis, puts
Penetrate immunization, Immunohistochemical Method and Enzyme Linked Immunoadsorbent Assay (enzyme-linked immunosorbnent assay,
ELISA), wherein ELISA is more common, and the principle for meat identification is to distinguish meat product to the identification of albumen using antibody
Kind.In Protocols in Molecular Biology most classic technology be polymerase chain reaction (polymerase chain reaction,
PCR), by designing specific primer, specific purpose DNA fragmentation is amplified, is separated through agarose gel electrophoresis different size of
DNA fragmentation, for identifying and detecting.Round pcr identifies for meat, and key is that all kinds of species specificities can be found
Genetic fragment, and design primer expands target fragment.Organoleptic detection is easy to be limited and by reviewer's training quality with master
The characteristics of property seen, one-sided;Elisa technique is sensitiveer reliable compared with organoleptic detection technology, but there are some drawbacks, such as mesh
Mark molecule content is low, and various meat specific antibodies are difficult to prepare and there are cross reactions etc.;Conventional PCR method is compared with organoleptic detection
Method is sensitive, and specificity is high, time saving compared with immunological method, but lacks comparison, needs to be compared to each other between multiple groups.Although conventional inspection
Survey method can effectively identify chilled and freeze thawing atlantic salmon, but be difficult to meet the real-time quick detection of high-volume sample.Cause
It is the necessity for identifying atlantic salmon and rainbow trout, ensuring food safety that this, which finds easy, quick, objective modern analytical technique,
Condition.
Near infrared spectrum (Near Infrared Spectrometry, NIR) is mainly the disresonance due to molecular vibration
Property makes molecular vibration, and from ground state to what is generated when high energy order transition, mainly C-H, O-H, N-H, S-H, P-H etc. of record are hydrogeneous
The frequency multiplication and sum of fundamental frequencies of group vibration absorb.The sum of fundamental frequencies of hydric group vibration and frequency multiplication at different levels near infrared spectrum and organic molecule
Uptake zone it is consistent, by the near infrared spectrum of scanned samples, in available sample organic molecule hydric group feature letter
Breath, the different groups such as near infrared absorption wavelength of methyl, methylene, phenyl ring or same group in different chemical environments and strong
Degree has significant difference, and the frequency multiplication of same group and sum of fundamental frequencies information can often be obtained in multiple wave bands in near-infrared spectra area.
Compared with common detection methods, NIR technology is a kind of indirect analysis technology, a large amount of and representative by collecting
Master sample, necessary data are measured by stringent careful chemical analysis, then by computer founding mathematical models, i.e., it is fixed
Then mark predicts unknown sample by the mathematical model again to reflect tested sample group normal distribution rule to greatest extent
Required data.It, which has, does not destroy that sample, not consume chemical reagent, free from environmental pollution etc. excellent easily and fast, efficiently, accurately and
Point is widely used in food, quality and security of agricultural products detection.In detection, because containing a large amount of albumen in meat
The organic matters such as matter, fat, organic acid, so a large amount of information can be obtained by the spectrum analysis to meat.
Domestic and foreign scholars have benefited our pursuits, including meat it is adulterated identify, microorganism quickly measure, meat be classified etc.
Primary Evaluation, and the research done to atlantic salmon and rainbow trout identification etc. is less.
Summary of the invention
It is an object of the invention to overcome defect existing in the prior art, such as: organoleptic examination is by reviewer's training element
The limitation of matter and have the characteristics that subjectivity, one-sided;Various meat specific antibodies are difficult to prepare and exist in elisa technique
Cross reaction etc.;Conventional PCR method needs to be compared to each other between multiple groups.The present invention provides a kind of atlantic salmon and rainbow trouts
Method for quick identification.
Specifically, the invention is realized by the following technical scheme: a kind of quick identification side of atlantic salmon and rainbow trout
Method carries out as steps described below:
(1) prepare atlantic salmon and rainbow trout sample and number, specially the chilled atlantic salmon of different batches, jelly
Melt atlantic salmon, chilled rainbow trout, freeze thawing rainbow trout sample;
(2) spectrum data gathering is carried out to the flesh of fish sample in step (1) near infrared spectrum (NIR) technology;
(3) spectral information for collecting step (2) utilizes first derivative (first by rows at spectrum matrix
Derivative, 1st Der), second dervative (second derivative, 2nd Der), mean value centralization (mean
Centering), multiplicative scatter correction (multiplicative scatter correction, MSC), standard normal variable become
Change (standard normal variate transformation, SNVT) and 6 kinds of normalization (Normalization) pre- places
Reason method pre-processes spectroscopic data;
(4) using principal component analysis (principal component analysis, PCA) method to pre- in step (3)
Treated, and spectroscopic data carries out dimension-reduction treatment;
(5) based on the spectroscopic data after step (4) dimension-reduction treatment, the chilled Atlantic Ocean is established using the method for Chemical Measurement
Salmon, freeze thawing atlantic salmon, chilled rainbow trout, freeze thawing rainbow trout quick identification model;
(6) spectral information acquisition is carried out to unknown flesh of fish sample to be measured, the quick identification model pair established using step (5)
The classification of unknown flesh of fish sample to be measured is quickly identified.
Wherein in above-mentioned steps (1), chilled atlantic salmon sample is the Faeroe Islands atlantic salmon of different batches;Freeze
Melt atlantic salmon sample and refer to -18 DEG C of freezing processings carried out the chilled atlantic salmon of every batch of 7 days, is then solved
Freeze and obtains;Chilled rainbow trout sample is the rainbow trout of the Qinghai cultivation of different batches;Freeze thawing rainbow trout sample refers to every batch of ice
Fresh rainbow trout carries out 7 days -18 DEG C of freezing processings, is then carried out defrosting acquisition.
Wherein above-mentioned steps (2) spectrum data gathering method particularly includes: closely red using Antaris II type Fourier transformation
External spectrum instrument (Thermo Fisher, the U.S.) carries out spectral scan, scanning range 10000- using diffusing reflection mode
4000cm-1, scanning times are 16 times, resolution ratio 8cm-1。
The wherein pretreatment of above-mentioned steps (3) spectroscopic data specifically: using 1st Der, 2nd Der, MC, MSC,
SNVT, Normalization method pre-process spectroscopic data using matlab software.
Wherein above-mentioned steps (4) carry out dimension-reduction treatment to data pretreated in step (3) using PCA method: first
PCA processing is carried out to data, then chooses input of the different number of principal components as model.
Wherein in above-mentioned steps (5), the method for the Chemical Measurement is linear discriminant analysis (linear
Discriminant analysis, LDA), support vector machines (support vector machine, SVM) or backpropagation
Artificial neural network (back-propagation artificial neural network, BPANN), preferably backpropagation people
Artificial neural networks (BPANN) method.
Wherein above-mentioned steps (6) quickly identify the classification of unknown flesh of fish sample to be measured, using NIR technology to be measured
Flesh of fish sample carries out spectrum data gathering, brings into after the data of flesh of fish sample to be measured are then first passed through step (3) and (4) processing
In the established quick identification model of step (5), the identification of atlantic salmon and rainbow trout is completed using Matlab processing software.
The object of the method for quick identification of the atlantic salmon true and false of the invention are as follows: chilled atlantic salmon, the big west of freeze thawing
Foreign salmon, chilled rainbow trout, freeze thawing rainbow trout.
Compared with prior art, beneficial effects of the present invention embody as follows:
(1) if NIR technology used in the present invention makes molecular vibration from ground state to height due to the anharmonicity of molecular vibration
It is generated when energy level transition, the frequency multiplication and sum of fundamental frequencies of mainly C-H, O-H, N-H, S-H, P-H etc. hydric groups vibration of record are inhaled
It receives.Near infrared spectrum is consistent with the uptake zone of sum of fundamental frequencies and frequency multiplication at different levels that hydric group in organic molecule vibrates, and passes through scanning
The near infrared spectrum of sample, the characteristic information of organic molecule hydric group, different groups such as methyl, methylene in available sample
The near infrared absorption wavelength and intensity of base, phenyl ring etc. or same group in different chemical environments have significant difference, and same
The frequency multiplication of group and sum of fundamental frequencies information can often be obtained in multiple wave bands in near-infrared spectra area, thus analyst coverage can almost cover it is all
Organic compound and mixture;And containing organic matters such as a large amount of protein, fat, organic acids in meat, by meat
Spectrum analysis can obtain a large amount of information.Have compared with other technologies and does not destroy easily and fast, efficiently, accurately and
Sample, the advantages that not consuming chemical reagent, is free from environmental pollution.
(2) present invention utilizes 1st Der, 6 kinds of 2nd Der, MC, MSC, SNVT, Normalization preprocess methods
Original spectral data is pre-processed, optimal preprocess method is selected according to model discrimination height.
(3) present invention identify with rainbow trout and be ground using NIR technology combination chemometrics method to atlantic salmon
Study carefully.Because NIR technology is a kind of indirect analysis technology, it is necessary to realize qualitative point to unknown sample by establishing calibration model
Analysis.Therefore using chilled atlantic salmon, freeze thawing atlantic salmon, chilled rainbow trout, freeze thawing rainbow trout as research object, first with
The spectral information of NIR technology collecting sample, and the spectroscopic data three times of each piece of sample collected is averaging conduct
Then final data are established using chemometrics method at spectrum matrix by rows and identify model.Compare linear discriminant
Three kinds of (LDA), support vector machines (SVM) and back-propagation artificial neural network (BPANN) model recognition effects are analyzed, according to knowledge
Not rate height discovery BPANN modelling effect is best.At this point, number of principal components is 10, the discrimination of training set is 98.33%, there is 4
Sample is identified mistake, and test set discrimination is 95% at this time, has 6 samples to be identified mistake.Therefore BPANN modelling effect compared with
Good, the present invention identifies atlantic salmon and rainbow trout using NIR technology combination BPANN model.
Detailed description of the invention
Fig. 1 is atlantic salmon of the present invention and the original atlas of near infrared spectra of rainbow trout and average atlas of near infrared spectra;
Fig. 2 is the pretreated atlantic salmon of distinct methods of the present invention and rainbow trout atlas of near infrared spectra, wherein F-AS
For chilled atlantic salmon, FT-AS is freeze thawing atlantic salmon, and F-RT is chilled rainbow trout, and FT-RT is freeze thawing rainbow trout.
Specific embodiment
The present invention is further described below by specific embodiment and in conjunction with attached drawing, but is not intended to limit the present invention.
(1) prepare chilled atlantic salmon, freeze thawing atlantic salmon, chilled rainbow trout, freeze thawing rainbow trout sample:
Choose the Faeroe Islands atlantic salmon of 3 batches (different dates of manufacture);
Chilled atlantic salmon sample is that every batch of atlantic salmon is chosen the middle section flesh of fish, divides in sterile refrigerating chamber and grows up
The fish block of × wide × Gao Yuewei 3 × 3 × 1, amounts to 90 pieces by 30 pieces of every batch of.
Freeze thawing atlantic salmon sample is that the chilled atlantic salmon of every batch of is chosen the middle section flesh of fish, after freezing 7d that it is straight
Connect be put into 4 DEG C of refrigerating chambers thaw to meat sample central temperature be 4 DEG C.Then being divided into length × width × height in sterile refrigerating chamber is about 3
× 3 × 1 fish block, amounts to 90 pieces by 30 pieces of every batch of.
Chilled rainbow trout sample is that every batch of rainbow trout is chosen the middle section flesh of fish, is divided into length × width × height in sterile refrigerating chamber
About 3 × 3 × 1 fish block, amounts to 90 pieces by 30 pieces of every batch of.
Freeze thawing rainbow trout sample is that the chilled rainbow trout of every batch of is chosen the middle section flesh of fish, is directly placed into 4 after freezing 7d
It is 4 DEG C that DEG C refrigerating chamber, which thaws to meat sample central temperature,.Then being divided into length × width × height in sterile refrigerating chamber is about 3 × 3 × 1
Fish block, amounts to 90 pieces by 30 pieces of every batch of.
(2) spectrum data gathering is carried out to flesh of fish sample with NIR technology:
Using II type Fourier Transform Near Infrared instrument of Antaris (ThermoFisher, the U.S.), using diffusing reflection side
Formula carries out spectral scan, scanning range 10000-4000cm-1, scanning times are 16 times, resolution ratio 8cm-1.Each sample
20min at room temperature is placed before acquisition, acquisition Shi Xianyong filter paper draws the moisture of sample surface, to prevent moisture remained on surface
The curve of spectrum is had an impact.Spectrum is acquired in 3 different parts respectively to each sample surface.
Fig. 1 is atlantic salmon of the present invention and the original atlas of near infrared spectra of rainbow trout and average atlas of near infrared spectra;From flat
In equal atlas of near infrared spectra, cold fresh fish is similar with the spectrum tendency of the freeze thawing flesh of fish, mainly the difference of absorbance value,
8277,6900,5164cm-1 nearby has absorption peak.The reason is that protein, rouge of the flesh of fish after freeze-thaw, in the flesh of fish
Fat, moisture etc. change and influence the absorptivity of the flesh of fish, scattering coefficient, and then spectrally show this variation
Come.The SPECTRAL DIVERSITY of rainbow trout and atlantic salmon is not only the difference of absorbance value, in 5789-5670cm-1Big west nearby
Foreign salmon sample has narrow absorption peak.It can be seen that near infrared spectrum from original atlas of near infrared spectra to be overlapped on the whole more
Seriously, bands of a spectrum are wider, and intuitive differentiation is more difficult.Therefore it needs to be identified by chemometrics method.
(3) spectroscopic data is pre-processed using preprocess method:
The spectroscopic data for obtaining flesh of fish sample by NIR technology is arranged first, with spectral region 10000-
4000cm-1Corresponding 1557 absorbance values are variable, and the spectroscopic data three times of each sample collected is averaging and is made
For the spectroscopic data of the sample, and by rows at spectrum matrix.Then using 1st Der, 2nd Der, MC, MSC,
SNVT, Normalization method pre-process spectrum.
Fig. 2 be different pretreatments method of the present invention after atlantic salmon and rainbow trout atlas of near infrared spectra, 1st Der with
2nd Der method belongs to derivation denoising, and 1st Der can eliminate the translation of baseline, and 2nd Der can eliminate the rotation of baseline.
MC method is that each data matrix is subtracted average value, and treated in this way, and spectroscopic data sufficiently reflects change information, and makes
All data distributions simplify follow-up data operation in zero point two sides.MSC method has the function of scattering and the offset of eliminating light
Energy.SNVT method is pre-processed to a spectrum, and interference information is greatly reduced.Normalization method be for
By all data all in an identical data area, [- 1,1] is set by data area in this research, can be made
The distribution of variable and average value is more balanced.In order to obtain preferably modeling effect, before modeling, the present invention using 6 kinds not
Same preprocess method pre-processes original spectral data, determines every kind of model most according to the height of lift scheme performance
Good preprocess method.
(4) based on pretreated spectroscopic data in step (3), it is dropped using principal component analysis (PCA) method
Dimension processing:
Because spectroscopic data is more, if can directly reduce the speed and efficiency of identification using model, therefore step (3) is located in advance
The data application PCA method obtained after reason carries out dimension-reduction treatment, reduces operand, then chooses different number of principal components as mould
The input of type.PCA is a kind of a kind of unsupervised statistical method for multiple indexs being converted into several overall targets, it is along association
From multi-dimensional data space to low-dimensional data space projection, each principal component is the linear combination of original variable in variance maximum direction,
And it is irrelevant between each principal component.
(5) be based on step (4) dimension-reduction treatment spectroscopic data, using the method for Chemical Measurement establish atlantic salmon with
Rainbow trout quickly identifies model:
The present invention has chosen three kinds of chemometrics methods altogether and distinguishes to atlantic salmon and rainbow trout, specially
Linear discriminant analysis (LDA), back-propagation artificial neural network (BPANN), support vector machines (SVM) method;According to different masters
The corresponding training set discrimination of component number carrys out the effect of judgment models, and the discrimination of training set is higher, and modelling effect is better;
Wherein, LDA is a kind of conventional pattern-recognition and sample classification method, focuses on the distribution of sample in space
And distance analysis each other, data are projected into a direction by operation method, so that the projection between group and group to the greatest extent may be used
It can separate, and the relationship in same group is closer, then classifies in new space to sample;BPANN model is a kind of
By the Multi-layered Feedforward Networks of Back Propagation Algorithm training, it can learn and store a large amount of input-output mode map relationship, have
There is stronger operational capability, the more complicated classification problem of data can be handled;SVM model is based on structural risk minimization
Model is established, the input space is transformed to by a higher dimensional space by the nonlinear transformation defined with interior Product function, it will be to be solved
Pattern recognition problem transform into a quadratic programming optimization problem.
Choose total sample 2/3 is used as training set (chilled atlantic salmon, freeze thawing atlantic salmon, chilled rainbow trout, jelly
Melt each 60 of rainbow trout sample), LDA, BPANN and SVM model are established respectively, and the results are shown in Table 1.
The identification result of table 1 LDA, BPANN and SVM model training set and test set under different number of principal components
From table 1 it follows that original spectrum is pre-processed by SNVT, Normalization method in LDA model
Afterwards, training set discrimination is 89.58%.In BPANN model, pass through 1st using initial data modeling or initial data
It is modeled after the pretreatment of Der, MC method, training set discrimination reaches 98.33%.In SVM model, model recognition performance is lower,
Only after the pretreatment of Normalization method, the training set discrimination of model reaches 97.5%.
(6) quick predict is carried out to the classification of unknown flesh of fish sample to be measured:
Using NIR technology to atlantic salmon sample to be measured (chilled atlantic salmon, freeze thawing atlantic salmon, chilled rainbow
Each 30 of trout, freeze thawing rainbow trout sample) spectrum data gathering is carried out, it is then that the spectroscopic data of unknown flesh of fish sample to be measured is first
Pretreated spectra is carried out, recycles PCA method to carry out dimension-reduction treatment, then inputs established LDA, BPANN and SVM model
In, flesh of fish classification, which is completed, using Matlab processing software identifies.
Identification result is as shown in table 1, in LDA model, original spectrum by Normalization method pretreatment after,
Spectroscopic data balances the distribution of variable and average value more all in [- 1,1] range.At this point, model performance is best, survey
Examination collection discrimination reaches 86.67%.In BPANN model, after the pretreatment of MC method, test set discrimination reaches original spectrum
To 95%.Each data matrix is subtracted average value by MC method, simplifies the calculating of follow-up data processing, to the pre- of raising model
It is helpful to survey collection discrimination.In SVM model, model recognition performance is lower, only after the pretreatment of SNVT method, model
The discrimination of test set reaches 80.33%.
Higher according to discrimination, the better principle of modelling effect show that BPANN modelling effect is preferable, identification result with it is corresponding
The actual flesh of fish classification of sample is consistent substantially, this shows that BPANN model can be used for practical application.
In order to keep sample representative, selection is most easily obscured on domestic market, the big west of the two class flesh of fish-similar in color
Foreign salmon and rainbow trout;In order to make in the detection of model more accurately and reliably, convenient for using actual market, the embodiment of the present invention is protected
The diversity of sample is demonstrate,proved, the atlantic salmon for not only choosing 3 batches and rainbow trout are as sample, due to the main storage of the flesh of fish
Hiding mode is divided into refrigeration and cold storage, has also selected chilled and freeze thawing flesh of fish sample.In order to eliminate sample uneven and baseline drift,
Reduce system noise, improve signal-to-noise ratio etc. and influence, the present invention using 1st Der, 2nd Der, MC, MSC, SNVT,
Normalization6 kind preprocess method pre-processes original spectral data.In order to improve the speed and effect of model identification
Rate, reduces operand, and the present invention carries out dimension-reduction treatment to spectroscopic data using PCA method.In order to choose best discriminant technique model, this
Invention is chosen LDA, BPANN and SVM model and is identified respectively to atlantic salmon and rainbow trout, the results showed that BPANN model
It is more suitable for the identification of atlantic salmon and rainbow trout.In conjunction with the above advantage, the present patent application utilizes NIR technology combination BPANN
Model identifies atlantic salmon and rainbow trout.
Claims (7)
1. the method for quick identification of a kind of atlantic salmon and rainbow trout, it is characterised in that carry out as steps described below:
(1) prepare atlantic salmon and rainbow trout sample and number, the specially chilled atlantic salmon of different batches, freeze thawing is big
Western salmon, chilled rainbow trout, freeze thawing rainbow trout sample;
(2) spectrum data gathering is carried out to the flesh of fish sample in step (1) near infrared spectrum (NIR) technology;
(3) spectral information for collecting step (2) utilizes first derivative (first by rows at spectrum matrix
Derivative, 1st Der), second dervative (second derivative, 2nd Der), mean value centralization (mean
Centering), multiplicative scatter correction (multiplicative scatter correction, MSC), standard normal variable become
Change (standard normal variate transformation, SNVT) and 6 kinds of normalization (Normalization) pre- places
Reason method pre-processes spectroscopic data;
(4) using principal component analysis (principal component analysis, PCA) method to pretreatment in step (3)
Spectroscopic data afterwards carries out dimension-reduction treatment;
(5) based on the spectroscopic data after step (4) dimension-reduction treatment, chilled Atlantic salmon is established using the method for Chemical Measurement
Fish, freeze thawing atlantic salmon, chilled rainbow trout, freeze thawing rainbow trout quick identification model;
(6) spectral information acquisition is carried out to unknown flesh of fish sample to be measured, the quick identification model established using step (5) is to unknown
The classification of flesh of fish sample to be measured is quickly identified.
2. the method for quick identification of a kind of atlantic salmon and rainbow trout according to claim 1, it is characterised in that step
(1) in, chilled atlantic salmon sample is the Faeroe Islands atlantic salmon of different batches;Freeze thawing atlantic salmon sample refers to
- 18 DEG C of freezing processings that the chilled atlantic salmon of every batch of is carried out to 7 days, are then carried out defrosting acquisition;Chilled rainbow trout sample
It originally is the rainbow trout of the Qinghai cultivation of different batches;Freeze thawing rainbow trout sample refers to-the 18 of every batch of chilled rainbow trout progress 7 days
DEG C freezing processing, is then carried out defrosting acquisition.
3. the method for quick identification of a kind of atlantic salmon and rainbow trout according to claim 1, it is characterised in that step
(2) spectrum data gathering method particularly includes: utilize Antaris II type Fourier Transform Near Infrared instrument (Thermo
Fisher, the U.S.), spectral scan, scanning range 10000-4000cm are carried out using diffusing reflection mode-1, scanning times 16
It is secondary, resolution ratio 8cm-1。
4. the method for quick identification of a kind of atlantic salmon and rainbow trout according to claim 1, it is characterised in that step
(3) pretreatment of spectroscopic data specifically: 1st Der, 2nd Der, MC, MSC, SNVT, Normalization method are used,
Spectroscopic data is pre-processed using Matlab software.
5. the method for quick identification of a kind of atlantic salmon and rainbow trout according to claim 1, it is characterised in that step
(4) dimension-reduction treatment is carried out to data pretreated in step (3) using PCA method: PCA processing is carried out to data first, so
Input of the different number of principal components as model is chosen afterwards.
6. the method for quick identification of a kind of atlantic salmon and rainbow trout according to claim 1, it is characterised in that step
(5) in, the method for the Chemical Measurement is linear discriminant analysis (linear discriminant analysis, LDA), props up
Hold vector machine (support vector machine, SVM) or back-propagation artificial neural network (back-propagation
Artificial neural network, BPANN), preferably back-propagation artificial neural network (back-propagation
Artificial neural network, BPANN) method.
7. the method for quick identification of a kind of atlantic salmon and rainbow trout according to claim 1, it is characterised in that step
(6) classification of unknown flesh of fish sample to be measured is quickly identified, spectroscopic data is carried out to flesh of fish sample to be measured using NIR technology
The data of flesh of fish sample to be measured, are then first passed through that bring step (5) into after step (3) and (4) processing established quick by acquisition
Identify in model, the identification of atlantic salmon and rainbow trout is completed using Matlab processing software.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910126179.8A CN109765196A (en) | 2019-02-20 | 2019-02-20 | A kind of method for quick identification of atlantic salmon and rainbow trout |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910126179.8A CN109765196A (en) | 2019-02-20 | 2019-02-20 | A kind of method for quick identification of atlantic salmon and rainbow trout |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109765196A true CN109765196A (en) | 2019-05-17 |
Family
ID=66456870
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910126179.8A Pending CN109765196A (en) | 2019-02-20 | 2019-02-20 | A kind of method for quick identification of atlantic salmon and rainbow trout |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109765196A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110596078A (en) * | 2019-08-14 | 2019-12-20 | 暨南大学 | Method for tracing and identifying origin of geographical marked mandarin fish |
CN111678973A (en) * | 2020-07-18 | 2020-09-18 | 上海海洋大学 | Method for rapidly identifying Atlantic salmon and rainbow trout based on mass spectrometry technology |
CN111735806A (en) * | 2020-06-18 | 2020-10-02 | 中国海洋大学 | Rapid fish product identification method based on laser-induced breakdown spectroscopy technology |
CN112730410A (en) * | 2020-12-25 | 2021-04-30 | 上海海洋大学 | Method for quickly distinguishing seafood by using spectrometry |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106596453A (en) * | 2016-12-16 | 2017-04-26 | 中国水产科学研究院黄海水产研究所 | Method for discriminating wild and cultured sea bass based on near infrared spectroscopy technology |
CN109001146A (en) * | 2018-07-26 | 2018-12-14 | 江苏大学 | A kind of method for quick identification of chilled beef and the fresh beef of jellyization |
-
2019
- 2019-02-20 CN CN201910126179.8A patent/CN109765196A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106596453A (en) * | 2016-12-16 | 2017-04-26 | 中国水产科学研究院黄海水产研究所 | Method for discriminating wild and cultured sea bass based on near infrared spectroscopy technology |
CN109001146A (en) * | 2018-07-26 | 2018-12-14 | 江苏大学 | A kind of method for quick identification of chilled beef and the fresh beef of jellyization |
Non-Patent Citations (2)
Title |
---|
N.B. TITO等: "Use of near infrared spectroscopy to predict microbial numbers on Atlantic salmon", 《FOOD MICROBIOLOGY》 * |
吴浩等: "近红外光谱分析技术在动物源性食品检测中的应用进展", 《食品工业科技》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110596078A (en) * | 2019-08-14 | 2019-12-20 | 暨南大学 | Method for tracing and identifying origin of geographical marked mandarin fish |
CN111735806A (en) * | 2020-06-18 | 2020-10-02 | 中国海洋大学 | Rapid fish product identification method based on laser-induced breakdown spectroscopy technology |
CN111735806B (en) * | 2020-06-18 | 2021-12-17 | 中国海洋大学 | Rapid fish product identification method based on laser-induced breakdown spectroscopy technology |
CN111678973A (en) * | 2020-07-18 | 2020-09-18 | 上海海洋大学 | Method for rapidly identifying Atlantic salmon and rainbow trout based on mass spectrometry technology |
CN112730410A (en) * | 2020-12-25 | 2021-04-30 | 上海海洋大学 | Method for quickly distinguishing seafood by using spectrometry |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Xu et al. | Raman spectroscopy coupled with chemometrics for food authentication: A review | |
CN109765196A (en) | A kind of method for quick identification of atlantic salmon and rainbow trout | |
Cheng et al. | Heterospectral two-dimensional correlation analysis with near-infrared hyperspectral imaging for monitoring oxidative damage of pork myofibrils during frozen storage | |
CN108875913B (en) | Tricholoma matsutake rapid nondestructive testing system and method based on convolutional neural network | |
Ye et al. | Detecting and classifying minor bruised potato based on hyperspectral imaging | |
Pang et al. | Rapid vitality estimation and prediction of corn seeds based on spectra and images using deep learning and hyperspectral imaging techniques | |
CN103278464B (en) | Flesh of fish detection method and device | |
Morsy et al. | Robust linear and non-linear models of NIR spectroscopy for detection and quantification of adulterants in fresh and frozen-thawed minced beef | |
Zhang et al. | Identification of corn seeds with different freezing damage degree based on hyperspectral reflectance imaging and deep learning method | |
Karoui et al. | Mid‐infrared spectroscopy as a new tool for the evaluation of fish freshness | |
CN109001146A (en) | A kind of method for quick identification of chilled beef and the fresh beef of jellyization | |
CN106124445A (en) | A kind of quick, Undamaged determination genetically engineered soybean method | |
CN109142269A (en) | A kind of method for quick identification of chilled beef difference storage time | |
Agyekum et al. | Rapid and nondestructive quantification of trimethylamine by FT-NIR coupled with chemometric techniques | |
Xu et al. | A novel hyperspectral microscopic imaging system for evaluating fresh degree of pork | |
Peijin et al. | Rapid identification of the variety and geographical origin of Wuyou No. 4 rice by fourier transform near-infrared spectroscopy coupled with chemometrics | |
Zhang et al. | Classification detection of saccharin jujube based on hyperspectral imaging technology | |
Qi et al. | Moldy peanut kernel identification using wavelet spectral features extracted from hyperspectral images | |
Li et al. | Fourier transform near-infrared spectroscopy coupled with variable selection methods for fast determination of salmon fillets storage time | |
CN106018321A (en) | Establishment method for single corn seed protein detection model and application of establishment method | |
Jiang et al. | Rapid nondestructive detecting of wheat varieties and mixing ratio by combining hyperspectral imaging and ensemble learning | |
Liu et al. | Fast detection of diarrhetic shellfish poisoning toxins in mussels using NIR spectroscopy and improved twin support vector machines | |
US20230089466A1 (en) | Establishment of Identification and Screening Method of Cows with A2 Beta-Casein Genotype of Producing A2 Milk and Applications Thereof | |
CN109765197A (en) | A kind of method for quick identification of chilled atlantic salmon and freeze thawing atlantic salmon | |
CN113049526B (en) | Corn seed moisture content determination method based on terahertz attenuated total reflection |
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 |
Application publication date: 20190517 |
|
RJ01 | Rejection of invention patent application after publication |