CN117969425A - Method for identifying gadus with big head based on multispectral imaging system - Google Patents
Method for identifying gadus with big head based on multispectral imaging system Download PDFInfo
- Publication number
- CN117969425A CN117969425A CN202410161704.0A CN202410161704A CN117969425A CN 117969425 A CN117969425 A CN 117969425A CN 202410161704 A CN202410161704 A CN 202410161704A CN 117969425 A CN117969425 A CN 117969425A
- Authority
- CN
- China
- Prior art keywords
- model
- qda
- training
- data
- gadus
- 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
- 238000000034 method Methods 0.000 title claims abstract description 49
- 241000276435 Gadus Species 0.000 title claims abstract description 18
- 238000000701 chemical imaging Methods 0.000 title abstract description 8
- 241000251468 Actinopterygii Species 0.000 claims abstract description 37
- 230000003595 spectral effect Effects 0.000 claims abstract description 31
- 238000001228 spectrum Methods 0.000 claims abstract description 18
- 238000012549 training Methods 0.000 claims description 26
- 239000011159 matrix material Substances 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 abstract description 10
- 235000019688 fish Nutrition 0.000 description 30
- 239000000523 sample Substances 0.000 description 30
- 241000276457 Gadidae Species 0.000 description 17
- 241000276438 Gadus morhua Species 0.000 description 16
- 238000001514 detection method Methods 0.000 description 7
- 238000003709 image segmentation Methods 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 6
- 235000013372 meat Nutrition 0.000 description 6
- 241000894007 species Species 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 241001599589 Gadus macrocephalus Species 0.000 description 3
- 238000011529 RT qPCR Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000002372 labelling Methods 0.000 description 3
- 230000000877 morphologic effect Effects 0.000 description 3
- 108090000623 proteins and genes Proteins 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 101150001086 COB gene Proteins 0.000 description 2
- 241000282465 Canis Species 0.000 description 2
- 101150053771 MT-CYB gene Proteins 0.000 description 2
- 241001417902 Mallotus villosus Species 0.000 description 2
- 241000269821 Scombridae Species 0.000 description 2
- 101150006264 ctb-1 gene Proteins 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000004907 flux Effects 0.000 description 2
- 235000020640 mackerel Nutrition 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 101150088166 mt:Cyt-b gene Proteins 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 108020004465 16S ribosomal RNA Proteins 0.000 description 1
- 102100025287 Cytochrome b Human genes 0.000 description 1
- 102000015884 Cytochrome c oxidase subunit I Human genes 0.000 description 1
- 108050004212 Cytochrome c oxidase subunit I Proteins 0.000 description 1
- 108010075028 Cytochromes b Proteins 0.000 description 1
- 241001313700 Gadus chalcogrammus Species 0.000 description 1
- 241000276495 Melanogrammus aeglefinus Species 0.000 description 1
- 241000276498 Pollachius virens Species 0.000 description 1
- 241000785681 Sander vitreus Species 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000012472 biological sample Substances 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000004090 dissolution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 210000003470 mitochondria Anatomy 0.000 description 1
- 230000002438 mitochondrial effect Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 239000013642 negative control Substances 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 102000004169 proteins and genes Human genes 0.000 description 1
- 238000003753 real-time PCR Methods 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000004611 spectroscopical analysis Methods 0.000 description 1
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
-
- 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
- G01N21/84—Systems specially adapted for particular applications
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a method for identifying gadus with big head based on a multispectral imaging system, and relates to the technical field of fish identification. The invention discloses an identification method of gadus, which comprises the following steps: (1) acquiring spectral data information of a fish sample; (2) Bringing the spectrum data information into a trained QDA model, namely a database; (3) And (3) bringing the fish sample of the unknown variety into the database obtained in the step (1) for comparison to obtain the specific variety. The identification method of the gadus with the advantages of simple operation, short operation time, no need of complex instruments and complicated comparison process, and capability of identifying the authenticity of the gadus with the advantages of simple operation.
Description
Technical Field
The invention relates to the technical field of fish identification, in particular to a method for identifying gadus with big head based on a multispectral imaging system.
Background
The head cod (Gadus macrocephalus), also known as Pacific cod, belongs to the order of the Gadus (Gabiformes), the family of the Gadus (Gadidae), the genus Gadus (Gadus). The headedness cod is mainly distributed in the North Pacific region. The big head cod is prolonged in shape, slightly flattened at side and gradually tapered backwards at tail, 25-40cm long, big head, big mouth, slightly longer upper jaw than lower jaw, neck with one tentacle, length equal to or slightly longer than eye diameter, grey brown at head, back and body side, grey white ventral surface, pale yellow chest collaterals, and grey fins.
At present, a plurality of fish species identification methods mainly comprise a traditional morphological method, a protein analysis method and a molecular biological analysis method according to the technical principle. The morphological identification method is the most traditional identification method, and is mainly carried out according to morphological standards such as appearance, color, smell, taste and other sensory indexes, and has the advantages of simplicity, intuitiveness, low cost, strong practicability and the like. However, when used for identifying fish meat samples having very similar characteristics, the identification method becomes difficult, and the use of the method is limited.
The prior art discloses a real-time fluorescence PCR detection method for the components of the headedness codfish, the atlantic cods, the walleye cods and other 16 codiform fishes and mitochondria 16S rRNA gene sequences of miscible species thereof are compared, primers or probes with stronger specificity are selected, the real-time fluorescence PCR detection method for the components of the headedness codfish is established, the absolute sensitivity and the relative sensitivity of the detection of the components of the headedness codfish are stronger, and a powerful guarantee is provided for the trade of the codfish (the research of real-time fluorescence PCR detection technology for the components of the codfish and the oil codfish, xinhongmei, university of Dalian industry, 2023).
Literature: the research of 3 molecular identification technologies of common cods in codaceae discloses a detection method of commercial cods, which comprises the steps of selecting antarctic canine fish and capelin mackerel which are easy to confuse with cods as negative controls, selecting 16SrDNA genes to design primers for distinguishing cods from non-cods, selecting mitochondrial Cytb genes to respectively design species-specific primers for Atlantic cods, pacific cods, haddock, antarctic canine fish and capelin mackerel, establishing a qPCR system, and carrying out species identification analysis on the qPCR system. The detection result shows that the established qPCR system has good specificity and repeatability, wherein the 16S rDNAqPCR system can rapidly and effectively distinguish codaceae from non-codaceae species (3 molecular identification technical researches of common codaceae codfish, wang Yi, zhejiang university, 2020).
There have been studies in the prior art to identify cod species using DNA mini-barcodes in combination with high resolution dissolution profiles (HRM), the major identified species including atlantic cod, pacific cod, walleye pollock and pollock. The study was mainly computer analyzed for two barcode regions, cytochrome c oxidase subunit i (COI) and cytochrome b (cytb), to determine genetic differences between the four species and was used to develop a real-time PCR method in combination with HRM analysis. However, only cod species can be identified by using this method, and the detection method is complicated.
At present, how to establish an identification method suitable for the gadus with simple operation method so as to identify the authenticity of the gadus is one of the most urgent demands in the market at present.
Disclosure of Invention
The invention aims to provide an identification method of the gadus with simple method and relatively accurate identification.
Term interpretation:
The term "multispectral data" refers to data collected over a range of wavelengths of a particular spectrum, between 10 and 100 aliquots, referred to as multispectral data.
The term "image segmentation" refers to the technique and process of dividing an image into several specific regions of unique properties and presenting objects of interest, which are key steps in performing image processing. Image segmentation is the process of dividing a digital image into mutually disjoint regions. The process of image segmentation is a labeling process, i.e. pixels belonging to the same region are given the same number.
The term "multispectral imaging technology" is an emerging technology developed based on imaging and spectroscopy, and is an analysis tool, which is excited by monochromatic light on the basis of simultaneous marking of multiple kinds of fluorescence, so that multiple kinds of fluorescence signals are mixed together, light with a required wavelength can be filtered and collected, then the collected multiple kinds of mixed light are subjected to unmixed through a system, and different components or positions of biological samples marked by different colors can be intuitively observed through signal output and display. Multispectral imaging systems can obtain high resolution spectra for each pixel of each image.
The term Database refers to a collection of data stored in a manner that can be shared by multiple users, independent of the application. The database has the following characteristics: the method is not repeated as much as possible, and is used for managing and controlling various application services of a specific organization in an optimal mode, wherein the data structure is independent of application programs using the data structure, and the data addition, deletion, modification and check are managed and controlled by unified software.
The term "spectral reflectance value (SPECTRAL REFLECTANC)", also known as spectral reflectance factor or spectral reflectance, refers to the ratio of the spectral radiant flux at a wavelength reflected from an object to the spectral radiant flux at the same wavelength reflected from a fully diffuse reflecting surface under specific lighting conditions in a direction defined by a prescribed solid angle.
In order to achieve the above object, the present invention has the following technical scheme:
In one aspect, the invention provides a method for identifying gadus, comprising the following steps:
(1) Acquiring spectral data information of a fish sample;
(2) Bringing the spectrum data information into a trained QDA model, namely a database;
(3) And (3) bringing the fish sample of the unknown variety into the database obtained in the step (1) for comparison to obtain the specific variety.
Preferably, the wavelength range of the spectral data information acquisition in the step (1) is 19 wave bands from visible light to near infrared light, respectively 365nm,405nm,430nm,450nm,470nm,490nm,515nm,540nm,570nm,590nm,630nm,645nm,660nm,690nm,780nm,850nm,880nm,940nm,970nm.
Specifically, the extracting spectral data information of the fish sample in the step (1) includes: and (3) extracting a fish sample, photographing the fish sample by using a multispectral imager to obtain a multispectral image of the fish sample, and extracting spectral data of the fish sample by using a multispectral imaging system.
More specifically, the time for photographing the fish sample by using the multispectral imager is 5-20s;
still more specifically, the photographing time is 5s to 10s.
More specifically, the method for extracting the spectral data of the fish meat by using the multispectral imaging system comprises the following steps:
(1) Opening VideometerLab software, and importing multispectral images of the fish sample;
(2) Removing image background, and performing image segmentation or layer labeling;
(3) Spectral information extraction is performed on the selected region using VideometerLab software.
Preferably, the image segmentation in step (2) is performed using a fixed pixel size of 52×52 pixels.
In some embodiments of the present invention, the pixel size of the fixed pixel point can be adjusted according to actual requirements.
Preferably, the operation of the step (3) in VideometerLab software is to sequentially select Image Tools, measurement, and Statistics, so as to obtain the spectrum information of the sample.
Specifically, the spectral information is an average spectral reflectance value of the selected region.
Further preferably, the training of the QDA model comprises the steps of: and (3) bringing the spectrum data of the headedness cod into a QDA model, using a creation tool of the model, importing training data through a function of the model, loading the model and constructing a network, and training to obtain a training result.
Still further preferably, the model creation tool is RStudio.
More specifically, the QDA model is predicted by calling predict a method after loading the model with the MASS package in R.
Preferably, the training of the QDA model comprises a splitting step of spectrum data.
Further preferably, the spectral data is split into a training set of 70% of the spectral data and a test set of 30% of the spectral data.
Preferably, the QDA training code:
fit_qda_clsm<-qda(form_clsm,data＝traindata,method＝"t")。
Specifically, the data range at each spectrum of the headup cod: 365nm from 20.78 to 26.18, 405nm from 19.37 to 25.34, 430nm from 20.37 to 26.10, 450nm from 25.88 to 31.11, 470nm from 26.54 to 31.62, 490nm from 26.36 to 31.35, 515nm from 25.31 to 30.18, 540nm from 25.05 to 29.82, 570nm from 23.39 to 28.15, 590nm from 23.73 to 28.63, 630nm from 24.16 to 29.25, 645nm from 24.29 to 29.46, 660nm from 24.33 to 29.50, 690nm from 24.27 to 29.53, 780nm from 24.12 to 29.56, 850nm from 31.30 to 36.21, 880nm from 34.18 to 38.86, 940nm from 40.19 to 44.37, 970nm from 40.86 to 45.03.
In particular, the fish comprises frozen fish or fresh fish bodies.
Preferably, the cut meat pieces are selected from meat near the vertebrae of the fish for cutting.
In particular, sampling near the spine is primarily aimed at taking as much as possible a larger and more uniform piece of fish.
Specifically, the size of the fish sample is (0.5-3 cm).
Preferably, the training of the QDA model further includes a step of testing the training result in such a manner that Accuracy (Accuracy), recall (Macro-R), precision (Macro-P) and harmonic mean (Macro-F1) of the model are calculated using the confusion matrix.
The calculation formula is as follows:
Accuracy rate:
Accuracy rate:
Recall rate:
wherein: true Positive (TP): a positive sample is successfully predicted as positive.
True Negative (TN): negative samples were successfully predicted as negative.
False Positive (FP): the negative sample error is predicted as positive.
FALSE NEGATIVE (FN): the positive sample error is predicted negative.
F value:
The calculation formula of Macro:
The calculation result of the accuracy is an average value of the accuracy.
Preferably, in the step (3), the specific variety is compared by taking the sample data to be tested into a database model to obtain a predicted variety and a probability value, comparing predict whether the class result and the label are matched, if so, the prediction is correct, and if not, the prediction is incorrect.
The beneficial effects of the invention are as follows:
the identification method of the headedness cod constructed by the invention has the advantages of simple operation, short operation time, no need of complex instruments and complicated comparison process, and can identify the authenticity of the headedness cod.
Drawings
FIG. 1 is a graph of average spectra and error data for a headcod.
Fig. 2 is a confusion matrix diagram for the QDA model.
FIG. 3 is a graph of the prediction results of predicting unknown fish samples using the QDA model.
Detailed Description
In order to make the technical means, the creation features, the achievement of the purpose and the effect of the present invention easy to understand, the present invention will be further elucidated with reference to the specific embodiments, but the following embodiments are only preferred embodiments of the present invention, not all of them. Based on the examples in the embodiments, those skilled in the art can obtain other examples without making any inventive effort, which fall within the scope of the invention. In the following examples, unless otherwise specified, the methods of operation used were conventional, the equipment used was conventional, and the materials used in the examples were the same.
Example 1 sample acquisition and extraction of spectral information
The collection of the sample comprises the following steps:
(1) Selecting the headedness cod to be treated, cutting a 1cm x 2cm x 1cm fish sample near the spine of the headedness cod, placing the fish sample in a 50mL centrifuge tube, and placing the centrifuge tube in a temperature of-80 ℃ for storage.
(2) And (3) respectively scanning the fish meat sample in the step (1) by using a multispectral instrument (model number VideometerLab of the multispectral instrument), wherein the analysis time is 5s-10s, and obtaining multispectral images of the cod meat sample.
(3) After the scanning is finished, the sample image is segmented, and the image segmentation is carried out by adopting the fixed pixel point 52 by 52 pixel size.
The spectrum information extraction method is characterized by comprising the following steps of:
The extraction of spectral information is performed by using the software VideometerLab matched with the multispectral imager, and the process is as follows:
(1) Opening VideometerLab software, and importing multispectral images of the fish sample;
(2) Removing image background, and performing image segmentation or layer labeling;
(3) Extracting spectral information from the selected region by VideometerLab software, wherein the wavelength range of spectral information acquisition is 19 wave bands from visible light to near infrared light, which are respectively 365nm,405nm,430nm,450nm,470nm,490nm,515nm,540nm,570nm,590nm,630nm,645nm,660nm,690nm,780nm,850nm,880nm,940nm,970nm.
The operation of the step (3) in VideometerLab software is that Image Tools- & gt Measurement- & gt Statistics are sequentially selected, so that the spectrum information of the sample can be obtained, the average value and the error of the average spectrum reflection value of the gadus under different spectrum conditions are shown in figure 1, and specific values are shown in table 1.
Table 1.
| sp | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| Band365 | 27.85 | 22.45 | 23.65 | 25.68 | 22.51 | 23.79 | 25.62 | 24.63 |
| Band405 | 25.72 | 21.36 | 21.99 | 23.75 | 21.71 | 23.03 | 24.9 | 24.52 |
| Band430 | 26.77 | 22.24 | 22.83 | 24.8 | 22.58 | 23.76 | 25.57 | 25.2 |
| Band450 | 33.07 | 26.73 | 28.27 | 31.23 | 26.64 | 28.69 | 30.45 | 28.96 |
| Band470 | 33.87 | 27.26 | 28.93 | 32.01 | 26.93 | 29.04 | 30.73 | 29.07 |
| Band490 | 33.74 | 27.11 | 28.78 | 31.86 | 26.6 | 28.65 | 30.3 | 28.65 |
| Band515 | 32.61 | 26.25 | 27.78 | 30.7 | 25.44 | 27.31 | 28.94 | 27.41 |
| Band540 | 32.16 | 26.01 | 27.58 | 30.37 | 25.24 | 27.06 | 28.6 | 27.12 |
| Band570 | 30.51 | 24.51 | 25.85 | 28.73 | 23.57 | 25.03 | 26.72 | 25.24 |
| Band590 | 31.02 | 24.92 | 26.27 | 29.31 | 23.93 | 25.27 | 27.07 | 25.47 |
| Band630 | 31.67 | 25.49 | 26.76 | 29.99 | 24.47 | 25.62 | 27.56 | 25.86 |
| Band645 | 31.89 | 25.71 | 26.94 | 30.21 | 24.66 | 25.7 | 27.71 | 25.95 |
| Band660 | 31.91 | 25.77 | 26.92 | 30.26 | 24.7 | 25.64 | 27.74 | 25.96 |
| Band690 | 31.86 | 25.84 | 26.84 | 30.24 | 24.82 | 25.57 | 27.68 | 25.86 |
| Band780 | 31.58 | 26.05 | 26.69 | 30.11 | 25.09 | 25.23 | 27.53 | 25.79 |
| Band850 | 38.05 | 32.88 | 33.52 | 36.87 | 32.78 | 32.39 | 34.67 | 33.23 |
| Band880 | 40.55 | 35.64 | 36.33 | 39.56 | 35.93 | 35.27 | 37.46 | 36.23 |
| Band940 | 45.59 | 41.68 | 42.06 | 44.89 | 42.77 | 41.49 | 43.42 | 42.8 |
| Band970 | 45.94 | 42.52 | 42.7 | 45.31 | 44.01 | 42.34 | 44.19 | 43.92 |
Example 2 model training
(1) QDA model training
The spectral data in example 1 is imported into a QDA model, and training data is imported through a function carried by the model, and the model is loaded and built into a network, and training is performed by using a creation tool RStudio of the model, thereby obtaining training results.
The model training process is as follows:
① Loading spectral data;
② Splitting spectral data, wherein 70% of the spectral data is a training set, 30% of the spectral data is a test set, and seed=34;
③ Loading a QDA model;
④ The results after training are reflected using a confusion matrix.
In step ②, seed: seed values set for random sampling.
The training codes of QDA are:
fit_qda_clsm<-qda(form_clsm,data＝traindata,method＝"t")。
the model training results were as follows:
the confusion matrix for the QDA model is shown in figure 2.
The overall accuracy, recall, accuracy, and harmonic mean of the database model trained by the QDA model are shown in table 2 below.
Table 2.
| Model | QDA |
| Accuracy (Accuracy) | 0.9906 |
| Recall (Macro-R) | 0.9855 |
| Accuracy rate (Macro-P) | 0.9898 |
| Blend average (Macro-F1) | 0.9875 |
As can be seen from Table 2 above, the QDA model has an accuracy of 99.06%, a recall of 0.9855, an accuracy of 0.9898, and a harmonic mean of 0.9875.
EXAMPLE 3 identification of unknown samples
The identification process of unknown fish samples during the test comprises the following steps:
(1) Loading data to be predicted and a data tag;
(2) Loading a model, and calling predict functions (R language tools);
The data tag in step (1) is the tag of an unknown sample, and the main purpose is to distinguish different samples.
And verifying the experimental result to judge whether the class result in predict is matched with the label, if so, the prediction is correct, and if not, the prediction is incorrect.
The prediction accuracy of the 3 sets of data is shown in table 3 below:
Table 3.
As can be seen from table 3 above, the predictive accuracy of the QDA model is 98.74% when the data amount is 195. When the data amount is 1905, the prediction accuracy of the QDA model is 98.74%. When the data amount is 2045, the prediction accuracy of the QDA model is 98.63%.
The unknown fish sample prediction process comprises the following steps:
(1) Loading data to be predicted;
(2) Loading a model and predicting by using predict functions;
(3) And obtaining the predicted variety and probability value.
Experimental results:
And loading the unknown fish sample into the trained model to obtain a predicted result shown in figure 3. And 15, representing the head cod, wherein the spectrum data information of the unknown fish sample is indicated to fall into the constructed spectrum database of the head cod, and the unknown sample can be judged to be the head cod.
The above codes are shown in Table 4: encoding.
Table 4.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (10)
1. The identification method of the gadus is characterized by comprising the following steps of:
(1) Acquiring spectral data information of a fish sample;
(2) Bringing the spectrum data information into a trained QDA model, namely a database;
(3) And (3) bringing the fish sample of the unknown variety into the database obtained in the step (1) for comparison to obtain the specific variety.
2. The method according to claim 1, wherein the spectral data information collected in step (1) has a wavelength range from visible light to near infrared light of 19 wavelength bands, respectively 365nm,405nm,430nm,450nm,470nm,490nm,515nm,540nm,570nm,590nm,630nm,645nm,660nm,690nm,780nm,850nm,880nm,940nm,970nm.
3. The method of authentication of claim 1, wherein training of the QDA model comprises the steps of: and (3) bringing the spectrum data of the headedness cod into a QDA model, using a creation tool of the model, importing training data through a function of the model, loading the model and constructing a network, and training to obtain a training result.
4. The method of claim 3, wherein the model creation tool is RStudio.
5. The authentication method of claim 4 wherein the QDA model is predicted by calling predict a method after loading the model with the MASS package in R.
6. The authentication method of claim 3, wherein the training code of the QDA model: fit_ qda _ clsm < -qda (form_ clsm, data= traindata, method= "t").
7. A method of authenticating according to claim 3, wherein the data ranges at each spectrum of the gadus are: 365nm from 20.78 to 26.18, 405nm from 19.37 to 25.34, 430nm from 20.37 to 26.10, 450nm from 25.88 to 31.11, 470nm from 26.54 to 31.62, 490nm from 26.36 to 31.35, 515nm from 25.31 to 30.18, 540nm from 25.05 to 29.82, 570nm from 23.39 to 28.15, 590nm from 23.73 to 28.63, 630nm from 24.16 to 29.25, 645nm from 24.29 to 29.46, 660nm from 24.33 to 29.50, 690nm from 24.27 to 29.53, 780nm from 24.12 to 29.56, 850nm from 31.30 to 36.21, 880nm from 34.18 to 38.86, 940nm from 40.19 to 44.37, 970nm from 40.86 to 45.03.
8. The method of claim 7, wherein the spectral data is an average spectral reflectance value.
9. The method of claim 3, wherein the training of the QDA model comprises testing the training results by calculating the accuracy, recall, precision, and harmonic mean of the model using the confusion matrix.
10. The method of claim 1, wherein the comparison is performed by taking the unknown fish sample data into a database model to obtain a predicted variety and probability value, comparing predict whether the class result and the label match, if so, the prediction is correct, otherwise, the prediction is incorrect.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410161704.0A CN117969425A (en) | 2024-02-05 | 2024-02-05 | Method for identifying gadus with big head based on multispectral imaging system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410161704.0A CN117969425A (en) | 2024-02-05 | 2024-02-05 | Method for identifying gadus with big head based on multispectral imaging system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117969425A true CN117969425A (en) | 2024-05-03 |
Family
ID=90851292
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410161704.0A Pending CN117969425A (en) | 2024-02-05 | 2024-02-05 | Method for identifying gadus with big head based on multispectral imaging system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117969425A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105190261A (en) * | 2013-03-21 | 2015-12-23 | Viavi科技有限公司 | Spectroscopic characterization of seafood |
| CN109765196A (en) * | 2019-02-20 | 2019-05-17 | 江苏大学 | A kind of method for quick identification of atlantic salmon and rainbow trout |
| US20200309746A1 (en) * | 2019-03-29 | 2020-10-01 | Shimadzu Corporation | Method for determining food-product quality and food-product quality determination device |
| US20220364985A1 (en) * | 2019-07-30 | 2022-11-17 | Alifax S.R.L. | Method and system to identify microorganisms |
| US11727669B1 (en) * | 2022-03-17 | 2023-08-15 | Institute of Facility Agriculture, Guangdong Academy of Agricultural Science | Intelligent recognition method of hyperspectral image of parasites in raw fish |
-
2024
- 2024-02-05 CN CN202410161704.0A patent/CN117969425A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105190261A (en) * | 2013-03-21 | 2015-12-23 | Viavi科技有限公司 | Spectroscopic characterization of seafood |
| CN109765196A (en) * | 2019-02-20 | 2019-05-17 | 江苏大学 | A kind of method for quick identification of atlantic salmon and rainbow trout |
| US20200309746A1 (en) * | 2019-03-29 | 2020-10-01 | Shimadzu Corporation | Method for determining food-product quality and food-product quality determination device |
| US20220364985A1 (en) * | 2019-07-30 | 2022-11-17 | Alifax S.R.L. | Method and system to identify microorganisms |
| US11727669B1 (en) * | 2022-03-17 | 2023-08-15 | Institute of Facility Agriculture, Guangdong Academy of Agricultural Science | Intelligent recognition method of hyperspectral image of parasites in raw fish |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Qin et al. | Detection of fish fillet substitution and mislabeling using multimode hyperspectral imaging techniques | |
| Kamruzzaman et al. | Application of NIR hyperspectral imaging for discrimination of lamb muscles | |
| Xing et al. | Combination of chemometric tools and image processing for bruise detection on apples | |
| Sosik et al. | Automated taxonomic classification of phytoplankton sampled with imaging‐in‐flow cytometry | |
| Naganathan et al. | Three dimensional chemometric analyses of hyperspectral images for beef tenderness forecasting | |
| Li et al. | Machine vision technology for detecting the external defects of fruits—a review | |
| CN110490895B (en) | Hyperspectral image processing-based method for improving meat source authenticity identification accuracy | |
| Fuchs et al. | Computational pathology analysis of tissue microarrays predicts survival of renal clear cell carcinoma patients | |
| US12050179B2 (en) | Method for extracting raman characteristic peaks employing improved principal component analysis | |
| Di Ruberto et al. | A feature learning framework for histology images classification | |
| Tu et al. | AIseed: An automated image analysis software for high-throughput phenotyping and quality non-destructive testing of individual plant seeds | |
| Popkov et al. | Machine learning for expert‐level image‐based identification of very similar species in the hyperdiverse plant bug family Miridae (Hemiptera: Heteroptera) | |
| Sandberg et al. | Applications of computer vision systems for meat safety assurance in abattoirs: A systematic review | |
| CN113077420B (en) | Fish fiber evaluation method and system based on convolutional neural network | |
| CN117969425A (en) | Method for identifying gadus with big head based on multispectral imaging system | |
| CN103076300B (en) | Method for judging and analyzing traditional Chinese medicine resource fingerprint information by specific mode identification model | |
| Klinkenbuß et al. | Performance of 3D morphological methods in the machine learning assisted classification of closely related fossil bivalve species of the genus Dreissena | |
| Montagner et al. | Features combination for art authentication studies: brushstroke and materials analysis of Amadeo de Souza-Cardoso | |
| Wang et al. | Detection of physical descaling damage in carp based on hyperspectral images and dimension reduction of principal component analysis combined with pixel values | |
| CN105506167A (en) | Method for synchronously detecting animal-origin ingredients in meat and meat products | |
| Hiraoka et al. | Identification of Rhus succedanea L. cultivars using elliptic Fourier descriptors based on fruit shape | |
| CN114878506A (en) | Ivory identification method based on near infrared spectrum | |
| Whelan et al. | On-device prediction for chronic kidney disease | |
| Concepcion II et al. | BivalveNet: A hybrid deep neural network for common cockle (Cerastoderma edule) geographical traceability based on shell image analysis | |
| CN117890311A (en) | Method for identifying piscine based on spectroscopy |
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 |