CN113505661A - Method, device, electronic equipment and storage medium for origin identification - Google Patents
Method, device, electronic equipment and storage medium for origin identification Download PDFInfo
- Publication number
- CN113505661A CN113505661A CN202110693186.3A CN202110693186A CN113505661A CN 113505661 A CN113505661 A CN 113505661A CN 202110693186 A CN202110693186 A CN 202110693186A CN 113505661 A CN113505661 A CN 113505661A
- Authority
- CN
- China
- Prior art keywords
- spectrum
- origin
- hyperspectral data
- identification model
- training
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2431—Multiple classes
Abstract
The invention discloses a method, a device, electronic equipment and a storage medium for identifying a place of origin, comprising the following steps: and acquiring hyperspectral data of the agricultural product to be detected, and inputting the hyperspectral data of the agricultural product to be detected into the trained origin and place identification model to obtain an origin and place identification result of the agricultural product to be detected. The trained origin identification model is obtained by training hyperspectral data corresponding to agricultural products in different origins. According to the invention, the labor identification cost is reduced by establishing the production area identification model, and the identification efficiency and accuracy are improved.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a method and a device for identifying a place of origin, electronic equipment and a storage medium.
Background
The varieties of agricultural products are various, and with the continuous increase of the market demand of the agricultural products, the quality of the agricultural products in the current market is uneven, and the production places are easily confused only by the appearance, so that the phenomena of inferior filling and production place counterfeiting occur sometimes.
In the process of market trading, the error of manually identifying agricultural products is large, the reliability is low, and the traditional method for chemically and biologically identifying agricultural products is complex in operation and long in period, so that the method cannot be popularized.
In view of the above, there is a need for a method for identifying a source of a product, which solves the above problems of the prior art.
Disclosure of Invention
In view of the above problems with the prior art methods, the present invention provides a method, apparatus, electronic device and storage medium for origin authentication.
In a first aspect, the present invention provides a method of identifying a source, comprising:
acquiring hyperspectral data of an agricultural product to be detected;
inputting hyperspectral data of the agricultural product to be detected into a trained origin identification model to obtain an origin identification result of the agricultural product to be detected;
the trained origin identification model is obtained by training hyperspectral data corresponding to agricultural products in different origins.
Further, before the inputting the hyperspectral data of the agricultural product to be tested into the trained origin and place identification model to obtain the origin and place identification result of the agricultural product to be tested, the method further comprises the following steps:
acquiring hyperspectral data of a plurality of groups of agricultural products and a plurality of producing area information corresponding to the hyperspectral data of the plurality of groups of agricultural products; the hyperspectral data of the multiple groups of agricultural products correspond to the production place information one by one;
determining an average spectral value of the region of interest according to the hyperspectral data;
generating a spectrum training set, a spectrum verification set and a spectrum test set according to the average spectrum value of the region of interest;
and training the origin identification model according to the spectrum training set and the origin information corresponding to the spectrum training set, and debugging the origin identification model according to the spectrum verification set and the spectrum test set to obtain the trained origin identification model.
Further, the determining an average spectrum value of the region of interest according to the hyperspectral data includes:
determining relative reflectivity data according to the hyperspectral data;
and determining the average spectral value of the region of interest according to the relative reflectivity data.
Further, before the training of the origin identification model according to the spectrum training set and the origin information corresponding to the spectrum training set, the method further includes:
preprocessing the spectral training set, the spectral verification set and the spectral test set by multivariate scatter correction;
and reducing the dimensions of the spectrum training set, the spectrum verification set and the spectrum test set by partial least square regression.
Further, the generating a spectrum training set, a spectrum verification set and a spectrum test set according to the average spectrum value of the region of interest includes:
and generating the spectrum training set, the spectrum verification set and the spectrum test set according to the ratio of 18:2: 5.
Further, the agricultural product is medlar; the plurality of production place information comprises Ningxia, Qinghai, inner Mongolia, Xinjiang, Gansu, Henan and Hebei.
In a second aspect, the present invention provides an apparatus for identifying a source, comprising:
the acquisition module is used for acquiring hyperspectral data of the agricultural product to be detected;
the processing module is used for inputting the hyperspectral data of the agricultural product to be detected into the trained origin and place identification model to obtain the origin and place identification result of the agricultural product to be detected; the trained origin identification model is obtained by training hyperspectral data corresponding to agricultural products in different origins.
Further, the processing module is further configured to:
before the hyperspectral data of the agricultural products to be tested are input into a trained origin and place identification model and the origin and place identification results of the agricultural products to be tested are obtained, acquiring hyperspectral data of a plurality of groups of agricultural products and a plurality of origin and place information corresponding to the hyperspectral data of the plurality of groups of agricultural products; the hyperspectral data of the multiple groups of agricultural products correspond to the production place information one by one;
determining an average spectral value of the region of interest according to the hyperspectral data;
generating a spectrum training set, a spectrum verification set and a spectrum test set according to the average spectrum value of the region of interest;
and training the origin identification model according to the spectrum training set and the origin information corresponding to the spectrum training set, and debugging the origin identification model according to the spectrum verification set and the spectrum test set to obtain the trained origin identification model.
Further, the processing module is specifically configured to:
determining relative reflectivity data according to the hyperspectral data;
and determining the average spectral value of the region of interest according to the relative reflectivity data.
Further, the processing module is further configured to:
preprocessing the spectrum training set, the spectrum verification set and the spectrum test set by multivariate scattering correction before training the origin identification model according to the spectrum training set and the origin information corresponding to the spectrum training set;
and reducing the dimensions of the spectrum training set, the spectrum verification set and the spectrum test set by partial least square regression.
Further, the processing module is specifically configured to:
and generating the spectrum training set, the spectrum verification set and the spectrum test set according to the ratio of 18:2: 5.
Further, the processing module is specifically configured to:
the agricultural product is medlar; the plurality of production place information comprises Ningxia, Qinghai, inner Mongolia, Xinjiang, Gansu, Henan and Hebei.
In a third aspect, the present invention also provides an electronic device, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the method for identifying a source according to the first aspect.
In a fourth aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of origin authentication according to the first aspect.
According to the technical scheme, the method, the device, the electronic equipment and the storage medium for identifying the producing area reduce the manual identification cost and improve the identification efficiency and accuracy by establishing the producing area identification model.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a system framework for a method of origin identification provided by the present invention;
FIG. 2 is a schematic flow chart of a method for identifying a source of origin according to the present invention;
FIG. 3 is a schematic diagram of a method of identifying a source provided by the present invention;
FIG. 4 is a schematic flow chart of a method for identifying a source of origin provided by the present invention;
FIG. 5 is a spectrum of a method of identifying a source of origin provided by the present invention;
FIG. 6 is a schematic structural diagram of a source identification device provided in the present invention;
fig. 7 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
The following further describes embodiments of the present invention with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and the protection scope of the present invention is not limited thereby.
The method for identifying the origin provided by the embodiment of the invention can be applied to a system architecture as shown in fig. 1, wherein the system architecture comprises a hyperspectral imager 100 and a server 200.
Specifically, the hyperspectral imager 100 is used for acquiring hyperspectral data of the agricultural product to be measured.
In one possible implementation, embodiments of the present invention employ visible light as well as near infrared light.
For example, the lens of the hyperspectral imager is a 400nm to 1000nm lens.
The server 200 is configured to input the hyperspectral data of the agricultural product to be tested to the trained origin and place identification model, and obtain an origin and place identification result of the agricultural product to be tested.
In the embodiment of the invention, the trained origin identification model is obtained by training hyperspectral data corresponding to agricultural products in different origins.
It should be noted that fig. 1 is only an example of a system architecture according to the embodiment of the present invention, and the present invention is not limited to this specifically.
Based on the above illustrated system architecture, fig. 2 is a schematic flow chart corresponding to a method for identifying a source according to an embodiment of the present invention, as shown in fig. 2, the method includes:
In the embodiment of the invention, as shown in fig. 3, a hyperspectral imager is used for spectrum scanning.
In one possible implementation, the embodiments of the present invention perform hyperspectral imaging for visible light and near-infrared light.
The near-infrared light is an electromagnetic wave between visible light and mid-infrared light.
In one possible implementation, embodiments of the invention collect hyperspectral data from 400nm to 1000 nm.
It should be noted that hyperspectral data in a wavelength range of 580nm to 2500nm can also be collected, which is not specifically limited in the embodiment of the present invention.
For example, the conditions for the spectral scan are: the distance between the lens of the hyperspectral imager and the agricultural products is 19 cm; the moving speed of the platform where the conveyor belt is located is 0.45 cm/s; the integration time was 9.7ms with a 400nm to 1000nm lens.
Further, in the embodiment of the present invention, the number of times of spectrum scanning is multiple, for example, 50 times, hyperspectral data of 400nm to 1000nm is collected each time, and then the hyperspectral data is obtained by averaging.
According to the scheme, the hyperspectral data of the agricultural product to be detected are averaged through multiple times of spectrum scanning, and the accuracy of the hyperspectral data of the agricultural product to be detected is improved.
The trained origin identification model is obtained by training hyperspectral data corresponding to agricultural products in different origins.
Taking an agricultural product as the medlar for example, the identification results of the producing areas can be Ningqi No. 5 of Ningxia Zhongning county, Ningqi No. 7 of Qinghai Dulan county, inner Mongolia Bayan Yan Yangte Er Wu Lao Qiangqi No. 5, Xinjiang Jinghe county Jingqi No. 1, Gansu Guozu county Ningqi No. 5, and the like.
According to the scheme, the production place identification model is adopted, so that the accuracy of production place identification is improved, and the cost of manual identification is reduced.
Before step 202, the method of the embodiment of the present invention has a flow as shown in fig. 4, and the specific steps include:
It should be noted that, the hyperspectral data of a plurality of groups of agricultural products correspond to a plurality of production place information one by one.
Specifically, spectral scanning is performed on agricultural products in different producing areas.
In the embodiment of the invention, as shown in fig. 3, a plurality of groups of agricultural products are placed on a background plate, the background plate and a white plate are placed on a conveyor belt, a bromine-tungsten lamp illuminates the plurality of groups of agricultural products, and a lens of a hyperspectral imager performs spectral scanning on the plurality of groups of agricultural products to obtain hyperspectral data and transmits the hyperspectral data to a computer.
Further, in the embodiment of the present invention, the number of times of spectrum scanning is multiple, for example, 50 times, hyperspectral data of 400nm to 1000nm is collected each time, and then the hyperspectral data is obtained by averaging.
According to the scheme, the high spectrum data accuracy is improved through multiple times of spectrum scanning averaging.
In one possible embodiment, the agricultural product is lycium barbarum; the multiple information of origin includes Ningxia, Qinghai, inner Mongolia, Xinjiang, Gansu, Henan and Hebei.
It should be noted that the agricultural products may also be beef, mutton, etc., and the production place information further includes tianjin, beijing, etc., which is not specifically limited in the embodiment of the present invention.
And 402, determining an average spectrum value of the region of interest according to the hyperspectral data.
And 403, generating a spectrum training set, a spectrum verification set and a spectrum test set according to the average spectrum value of the region of interest.
In the embodiment of the invention, the average spectral value of the region of interest is divided into three parts, and the three parts are recorded as a spectral training set, a spectral verification set and a spectral test set.
In one possible implementation, the spectral training set, the spectral validation set, and the spectral test set are generated at an 18:2:5 ratio.
It should be noted that the spectrum training set, the spectrum verification set, and the spectrum test set may also be generated according to a ratio of 16:3:5, which is not specifically limited in the embodiment of the present invention.
Specifically, the data of the spectrum scanning is divided into three parts by a train _ test _ split function in a Sklearn library of Python, and a spectrum training set, a spectrum verification set and a spectrum testing set are randomly grouped according to a proportion.
And step 404, training the origin identification model according to the spectrum training set and the origin information corresponding to the spectrum training set, and debugging the origin identification model according to the spectrum verification set and the spectrum test set to obtain the trained origin identification model.
In one possible embodiment, the source identification model is established based on a SoftMax model.
It should be noted that the SoftMax regression model is a generalization of the logistic regression model to the multi-classification problem.
Specifically, the origin identification model provided by the embodiment of the invention predicts the probability of the origin of the agricultural product to be detected, the origin with the highest predicted probability is used as the origin identification result of the agricultural product to be detected, and if the origin identification result is consistent with the real origin corresponding to the agricultural product to be detected, the prediction is correct.
In the embodiment of the invention, the main spectrum information obtained according to the spectrum training set and the corresponding origin information are modeled through SoftMax, and the origin and the destination of agricultural products are identified through the trained origin identification model through the spectrum verification set and the spectrum test set debugging model.
In step 402, according to the embodiment of the present invention, relative reflectance data is determined according to the hyperspectral data.
In one possible implementation mode, the hyperspectral data of multiple groups of agricultural products are subjected to black and white correction, and relative reflectivity data are obtained.
Further, the specific calculation formula of the black-and-white correction is as follows:
wherein I isnewRepresenting the relative reflectivity, I, of the corrected imagerawRepresenting the energy value, I, of the original imagedarkRepresenting the energy value of the blackboard image, IwhitRepresenting the energy value of the whiteboard image.
According to the scheme, the hyperspectral data of a plurality of groups of agricultural products are denoised through black and white correction, and the accuracy of the production area identification model prediction is improved.
Further, an average spectral value of the region of interest is determined from the relative reflectance data.
In the embodiment of the invention, the region of interest is extracted from the relative reflectivity data, and then the average spectral value of the region of interest is calculated.
For example, in the case where the agricultural product is lycium barbarum, the extracted region of interest is spectral data of a portion of lycium barbarum.
Before step 404, the embodiment of the present invention preprocesses the spectrum training set, the spectrum verification set, and the spectrum test set by multivariate scattering correction.
It should be noted that the multivariate scattering correction can effectively eliminate the spectrum difference caused by the difference of the scattering levels, thereby enhancing the correlation between the spectrum and the data. The baseline shift and shift phenomena of the spectrum data are corrected by the ideal spectrum, so that in practical application, the true ideal spectrum data cannot be acquired, and therefore, the average value of all the spectrum data is often assumed as the ideal spectrum.
Specifically, an average value of the spectrum data is obtained, unitary linear regression is performed on the spectrum of each agricultural product and the average spectrum, the baseline translation amount and the offset amount of each agricultural product are obtained through solving, the baseline translation amount is subtracted, and then the offset amount is divided, so that corrected spectrum data are obtained.
Further, the specific calculation formula is as follows:
it should be noted that the dimension of the original spectrum data is m × n, m represents the number of agricultural products, and n represents the number of wavelength points used for spectrum collection. Formula (II)
Representing the mean spectrum, χ, of the agricultural productiIs a 1 Xn dimensional matrix representing a spectral vector of a single agricultural product, wiAnd biRespectively represent near infrared spectrum χ of each agricultural productiAnd (4) carrying out unary linear regression on the average spectrum A to obtain a relative offset coefficient and a translation amount.
According to the scheme, the spectrum training set, the spectrum verification set and the spectrum test set are preprocessed through the multivariate scattering correction, so that interference data are reduced, and the accuracy of the prediction of the origin identification model is improved.
It should be noted that the preprocessing of the spectrum training set, the spectrum verification set, and the spectrum test set may also be performed by normalization, a convolution smoothing method, a standard normal variable, wavelet transformation, orthogonal signal correction, and the like, which is not specifically limited in this embodiment of the present invention.
And further, reducing the dimensions of the spectrum training set, the spectrum verification set and the spectrum test set by partial least square regression.
It should be noted that the partial least squares regression is to find a linear regression model by projecting the predicted variable and the observed variable to a new space respectively. And (3) performing partial least squares regression dimension reduction by using MATLAB.
Specifically, a component t is extracted from the set of independent variables Xh(h-1, 2, …), each component being independent of the other. Subsequently establishing an extracted component thAnd the dependent variable Y.
According to the scheme, the spectral training set, the spectral verification set and the spectral test set are subjected to dimensionality reduction through partial least squares regression, interference is removed, and meanwhile efficient calculation is achieved.
It should be noted that the dimensions of the spectrum training set, the spectrum verification set, and the spectrum test set may also be reduced through Linear Discriminant Analysis (LDA) and Principal Component Analysis (PCA), which is not specifically limited in this embodiment of the present invention.
According to the scheme, the labor identification cost is reduced by establishing the origin identification model, and the identification efficiency and accuracy are improved.
Further, taking the medlar as an example, fig. 5 is spectrum curves of medlar in different producing areas, in which NX, QH, NM, XJ, GS respectively represent spectrum curves of ningqi No. 5 in ningxia zhongning county, ningqi No. 7 in qinghai dunlan county, inner mongolia Bayan yangtei front ningqi No. 5, xinjiang jing he county fine qi No. 1, and gansu melon county ningqi No. 5.
Based on the same inventive concept, fig. 6 exemplarily illustrates a device for identifying a source, which may be a flow of a method for identifying a source, according to an embodiment of the present invention.
The apparatus, comprising:
the acquisition module 601 is used for acquiring hyperspectral data of an agricultural product to be detected;
the processing module 602 is configured to input the hyperspectral data of the agricultural product to be detected into a trained origin and place identification model, so as to obtain an origin and place identification result of the agricultural product to be detected; the trained origin identification model is obtained by training hyperspectral data corresponding to agricultural products in different origins.
Further, the processing module 602 is further configured to:
before the hyperspectral data of the agricultural products to be tested are input into a trained origin and place identification model and the origin and place identification results of the agricultural products to be tested are obtained, acquiring hyperspectral data of a plurality of groups of agricultural products and a plurality of origin and place information corresponding to the hyperspectral data of the plurality of groups of agricultural products; the hyperspectral data of the multiple groups of agricultural products correspond to the production place information one by one;
determining an average spectral value of the region of interest according to the hyperspectral data;
generating a spectrum training set, a spectrum verification set and a spectrum test set according to the average spectrum value of the region of interest;
and training the origin identification model according to the spectrum training set and the origin information corresponding to the spectrum training set, and debugging the origin identification model according to the spectrum verification set and the spectrum test set to obtain the trained origin identification model.
Further, the processing module 602 is specifically configured to:
determining relative reflectivity data according to the hyperspectral data;
and determining the average spectral value of the region of interest according to the relative reflectivity data.
Further, the processing module 602 is further configured to:
preprocessing the spectrum training set, the spectrum verification set and the spectrum test set by multivariate scattering correction before training the origin identification model according to the spectrum training set and the origin information corresponding to the spectrum training set;
and reducing the dimensions of the spectrum training set, the spectrum verification set and the spectrum test set by partial least square regression.
Further, the processing module 602 is specifically configured to:
and generating the spectrum training set, the spectrum verification set and the spectrum test set according to the ratio of 18:2: 5.
Further, the processing module 602 is specifically configured to:
the agricultural product is medlar; the plurality of production place information comprises Ningxia, Qinghai, inner Mongolia, Xinjiang, Gansu, Henan and Hebei.
Based on the same inventive concept, an embodiment of the present invention provides an electronic device, which specifically includes the following contents, with reference to fig. 7: a processor 701, a memory 702, a communication interface 703 and a communication bus 704;
the processor 701, the memory 702 and the communication interface 703 complete mutual communication through the communication bus 704; the communication interface 703 is used for implementing information transmission between the devices;
the processor 701 is configured to call the computer program in the memory 702, and the processor implements all the steps of the method for identifying the origin when executing the computer program, for example, the processor implements the following steps when executing the computer program: acquiring hyperspectral data of an agricultural product to be detected; inputting hyperspectral data of the agricultural product to be detected into a trained origin identification model to obtain an origin identification result of the agricultural product to be detected; the trained origin identification model is obtained by training hyperspectral data corresponding to agricultural products in different origins.
Based on the same inventive concept, yet another embodiment of the present invention provides a non-transitory computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs all the steps of the above-mentioned method for origin authentication, e.g., the processor performs the following steps when executing the computer program: acquiring hyperspectral data of an agricultural product to be detected; inputting hyperspectral data of the agricultural product to be detected into a trained origin identification model to obtain an origin identification result of the agricultural product to be detected; the trained origin identification model is obtained by training hyperspectral data corresponding to agricultural products in different origins.
In addition, the logic instructions in the memory may be implemented in the form of software functional units and may be stored in a computer readable storage medium when sold or used as a stand-alone product. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a user life pattern prediction apparatus, or a network device) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the embodiment of the present invention. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on such understanding, the above technical solutions may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a user life pattern prediction apparatus, or a network device, etc.) to execute the user life pattern prediction method according to the embodiments or some parts of the embodiments.
In addition, in the present invention, terms such as "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Moreover, in the present invention, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Furthermore, in the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method of identifying a source, comprising:
acquiring hyperspectral data of an agricultural product to be detected;
inputting hyperspectral data of the agricultural product to be detected into a trained origin identification model to obtain an origin identification result of the agricultural product to be detected;
the trained origin identification model is obtained by training hyperspectral data corresponding to agricultural products in different origins.
2. The method for identifying the producing area according to claim 1, wherein before the inputting the hyperspectral data of the agricultural product to be tested into the trained producing area identification model to obtain the producing area identification result of the agricultural product to be tested, the method further comprises:
acquiring hyperspectral data of a plurality of groups of agricultural products and a plurality of producing area information corresponding to the hyperspectral data of the plurality of groups of agricultural products; the hyperspectral data of the multiple groups of agricultural products correspond to the production place information one by one;
determining an average spectral value of the region of interest according to the hyperspectral data;
generating a spectrum training set, a spectrum verification set and a spectrum test set according to the average spectrum value of the region of interest;
and training the origin identification model according to the spectrum training set and the origin information corresponding to the spectrum training set, and debugging the origin identification model according to the spectrum verification set and the spectrum test set to obtain the trained origin identification model.
3. A method of identifying a source according to claim 2 wherein determining a region of interest mean spectral value from the hyperspectral data comprises:
determining relative reflectivity data according to the hyperspectral data;
and determining the average spectral value of the region of interest according to the relative reflectivity data.
4. A method for identifying a source according to claim 2, wherein before the training of the source identification model according to the spectrum training set and the source information corresponding to the spectrum training set, the method further comprises:
preprocessing the spectral training set, the spectral verification set and the spectral test set by multivariate scatter correction;
and reducing the dimensions of the spectrum training set, the spectrum verification set and the spectrum test set by partial least square regression.
5. The method of identifying a source of claim 2, wherein the generating a spectral training set, a spectral validation set, and a spectral test set from the region of interest mean spectral values comprises:
and generating the spectrum training set, the spectrum verification set and the spectrum test set according to the ratio of 18:2: 5.
6. A method of identifying a source according to claim 2 wherein the agricultural product is lycium barbarum; the plurality of production place information comprises Ningxia, Qinghai, inner Mongolia, Xinjiang, Gansu, Henan and Hebei.
7. An apparatus for identifying a source, comprising:
the acquisition module is used for acquiring hyperspectral data of the agricultural product to be detected;
the processing module is used for inputting the hyperspectral data of the agricultural product to be detected into the trained origin and place identification model to obtain the origin and place identification result of the agricultural product to be detected; the trained origin identification model is obtained by training hyperspectral data corresponding to agricultural products in different origins.
8. The apparatus for identifying a source of claim 7, wherein the processing module is further configured to:
before the hyperspectral data of the agricultural products to be tested are input into a trained origin and place identification model and the origin and place identification results of the agricultural products to be tested are obtained, acquiring hyperspectral data of a plurality of groups of agricultural products and a plurality of origin and place information corresponding to the hyperspectral data of the plurality of groups of agricultural products; the hyperspectral data of the multiple groups of agricultural products correspond to the production place information one by one;
determining an average spectral value of the region of interest according to the hyperspectral data;
generating a spectrum training set, a spectrum verification set and a spectrum test set according to the average spectrum value of the region of interest;
and training the origin identification model according to the spectrum training set and the origin information corresponding to the spectrum training set, and debugging the origin identification model according to the spectrum verification set and the spectrum test set to obtain the trained origin identification model.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method according to any of claims 1 to 6 are implemented when the processor executes the program.
10. A non-transitory computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110693186.3A CN113505661A (en) | 2021-06-22 | 2021-06-22 | Method, device, electronic equipment and storage medium for origin identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110693186.3A CN113505661A (en) | 2021-06-22 | 2021-06-22 | Method, device, electronic equipment and storage medium for origin identification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113505661A true CN113505661A (en) | 2021-10-15 |
Family
ID=78010342
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110693186.3A Pending CN113505661A (en) | 2021-06-22 | 2021-06-22 | Method, device, electronic equipment and storage medium for origin identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113505661A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113916825A (en) * | 2021-11-23 | 2022-01-11 | 山东省科学院自动化研究所 | American ginseng production area detection method based on terahertz spectrum technology |
| CN114720420A (en) * | 2022-03-24 | 2022-07-08 | 中国中医科学院中药研究所 | Method and system for identifying production area of Chinese prickly ash based on hyperspectral imaging technology |
| CN116933043A (en) * | 2023-09-15 | 2023-10-24 | 天津现代创新中药科技有限公司 | Identification method of chrysanthemum production place, construction method of model and electronic equipment |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1008952A2 (en) * | 1998-12-11 | 2000-06-14 | Florida Department of Citrus | System and method for identifying the geographic origin of a fresh commodity |
| CN104914156A (en) * | 2015-05-28 | 2015-09-16 | 黑龙江八一农垦大学 | Rice producing area identification method based on mineral matter analysis technology and application |
| CN109765194A (en) * | 2019-01-25 | 2019-05-17 | 中国中医科学院中药研究所 | Fructus lycii Production area recognition method based on high light spectrum image-forming technology |
| CN110514732A (en) * | 2019-09-03 | 2019-11-29 | 北京出入境检验检疫局检验检疫技术中心 | Identify the method for the sea-buckthorn place of production and/or kind based on rapid evaporation ionization mass spectrometry technology |
| WO2019244188A1 (en) * | 2018-06-20 | 2019-12-26 | Prima3 Srl | Innovative detector for "secure authentication" of valuables |
| CN110849828A (en) * | 2019-12-13 | 2020-02-28 | 嘉兴职业技术学院 | Saffron crocus classification method based on hyperspectral image technology |
| US10650279B1 (en) * | 2019-01-31 | 2020-05-12 | StradVision, Inc. | Learning method and learning device for heterogeneous sensor fusion by using merging network which learns non-maximum suppression |
| CN111488851A (en) * | 2020-04-17 | 2020-08-04 | 成都曙光光纤网络有限责任公司 | Traceability detection method, device, equipment and medium for fruit production place |
| CN111881965A (en) * | 2020-07-20 | 2020-11-03 | 北京理工大学 | Hyperspectral pattern classification and identification method, device and equipment for grade of origin of medicinal material |
-
2021
- 2021-06-22 CN CN202110693186.3A patent/CN113505661A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1008952A2 (en) * | 1998-12-11 | 2000-06-14 | Florida Department of Citrus | System and method for identifying the geographic origin of a fresh commodity |
| CN104914156A (en) * | 2015-05-28 | 2015-09-16 | 黑龙江八一农垦大学 | Rice producing area identification method based on mineral matter analysis technology and application |
| WO2019244188A1 (en) * | 2018-06-20 | 2019-12-26 | Prima3 Srl | Innovative detector for "secure authentication" of valuables |
| CN109765194A (en) * | 2019-01-25 | 2019-05-17 | 中国中医科学院中药研究所 | Fructus lycii Production area recognition method based on high light spectrum image-forming technology |
| US10650279B1 (en) * | 2019-01-31 | 2020-05-12 | StradVision, Inc. | Learning method and learning device for heterogeneous sensor fusion by using merging network which learns non-maximum suppression |
| CN110514732A (en) * | 2019-09-03 | 2019-11-29 | 北京出入境检验检疫局检验检疫技术中心 | Identify the method for the sea-buckthorn place of production and/or kind based on rapid evaporation ionization mass spectrometry technology |
| CN110849828A (en) * | 2019-12-13 | 2020-02-28 | 嘉兴职业技术学院 | Saffron crocus classification method based on hyperspectral image technology |
| CN111488851A (en) * | 2020-04-17 | 2020-08-04 | 成都曙光光纤网络有限责任公司 | Traceability detection method, device, equipment and medium for fruit production place |
| CN111881965A (en) * | 2020-07-20 | 2020-11-03 | 北京理工大学 | Hyperspectral pattern classification and identification method, device and equipment for grade of origin of medicinal material |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113916825A (en) * | 2021-11-23 | 2022-01-11 | 山东省科学院自动化研究所 | American ginseng production area detection method based on terahertz spectrum technology |
| CN114720420A (en) * | 2022-03-24 | 2022-07-08 | 中国中医科学院中药研究所 | Method and system for identifying production area of Chinese prickly ash based on hyperspectral imaging technology |
| CN116933043A (en) * | 2023-09-15 | 2023-10-24 | 天津现代创新中药科技有限公司 | Identification method of chrysanthemum production place, construction method of model and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113505661A (en) | Method, device, electronic equipment and storage medium for origin identification | |
| CN108710910B (en) | Target identification method and system based on convolutional neural network | |
| CN110717368A (en) | Qualitative classification method for textiles | |
| Uss et al. | Maximum likelihood estimation of spatially correlated signal-dependent noise in hyperspectral images | |
| Deng et al. | Moisture content prediction in tealeaf with near infrared hyperspectral imaging | |
| Mishra et al. | Automatic de-noising of close-range hyperspectral images with a wavelength-specific shearlet-based image noise reduction method | |
| Chaugule et al. | Identification of paddy varieties based on novel seed angle features | |
| Chen et al. | Evaluating fabric pilling with light-projected image analysis | |
| Madrid-Herrera et al. | Human image complexity analysis using a fuzzy inference system | |
| Swofford et al. | A method for measuring the quality of friction skin impression evidence: Method development and validation | |
| Jenifa et al. | Classification of cotton leaf disease using multi-support vector machine | |
| Yoon et al. | Hyperspectral image processing methods | |
| Taraschi et al. | Computing fractal descriptors of texture images using sliding boxes: An application to the identification of Brazilian plant species | |
| CN110807387A (en) | Object classification method and system based on hyperspectral image characteristics | |
| Kim et al. | ADOM: ADMM-based optimization model for stripe noise removal in remote sensing image | |
| Filipović et al. | Significance of morphological features in rice variety classification using hyperspectral imaging | |
| Quan | Quality evaluation method of agricultural product packaging image based on structural similarity and MTF | |
| Gu et al. | Learning deep vector regression model for no-reference image quality assessment | |
| Yeo et al. | Development of a hybrid autoencoder model for automated edible bird’s nest impurities inspection | |
| Zhang et al. | Nature scene statistics approach based on ICA for no-reference image quality assessment | |
| Farooq et al. | Detection of Late Blight Disease in Tomato Leaf Using Image Processing Techniques | |
| Saez et al. | Global pattern classification in dermoscopic images | |
| CN115841594B (en) | Attention mechanism-based coal gangue hyperspectral variable image domain data identification method | |
| Davis et al. | Automated quantitative description of spiral galaxy arm-segment structure | |
| Yu | Intrinsic decomposition method combining deep convolutional neural network and probability graph model |
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 |