Durian traceability model modeling method and device and durian traceability method

A model modeling, durian technology, applied in the field of devices, durian traceability methods, computer equipment and readable storage media, durian traceability model modeling methods, can solve problems such as low efficiency, and achieve the effect of improving efficiency

Pending Publication Date: 2021-05-07
FOOD INSPECTION CENT OF CIQ SHENZHEN
0 Cites 0 Cited by

AI-Extracted Technical Summary

Problems solved by technology

[0002] For the identification of the origin of imported durians, the import and export inspection department gen...
View more

Method used

In the embodiment of the present invention, after generating the traceability model, it is also possible to correctly discriminate the traceability model by inputting the isotope analysis data of durian samples of other known origins and the isotope analysis data of a plurality of corresponding target origins The statistics of the number of times to generate the accuracy of the traceability model. According to whether the accuracy rate of the traceability model passes the minimum accuracy value, determine whether the traceability model needs to continue...
View more

Abstract

The invention provides a durian traceability model modeling method, a durian traceability model modeling device and a durian traceability method. The durian traceability model modeling method comprises the following steps: acquiring isotope analysis data of pulp and kernels of a durian in a target producing area; generating a traceability model of the target producing area by using the isotope analysis data and the corresponding target producing area information through a preset analysis modeling algorithm; and verifying the accuracy rate of the traceability model by using the isotope analysis data of the target producing area and the isotope analysis data of other producing areas to obtain the accuracy rate of the traceability model. According to the durian traceability model modeling method, isotope analysis is performed on the durian in the target producing area, the corresponding traceability model is generated after the isotope analysis data is obtained, and the producing area of the durian sample can be quickly and accurately judged through the traceability model, so that the durian producing area detection efficiency is improved.

Application Domain

Material analysis by electric/magnetic meansResources +2

Technology Topic

Data miningTraceability +3

Image

  • Durian traceability model modeling method and device and durian traceability method
  • Durian traceability model modeling method and device and durian traceability method
  • Durian traceability model modeling method and device and durian traceability method

Examples

  • Experimental program(4)

Example Embodiment

[0045]Example 1
[0046]figure 1 It is a flow chart of a durian traceable model modeling method according to Embodiment 1 of the present invention, including the following steps:
[0047]Step S11: Acquisition of the isotope analysis data of the flesh and fruit core of the target.
[0048]In the embodiment of the present invention, the isotope analysis data comprises at least one panelable content in the pulp, isotope ratio of at least one element in the pulp, isotope content and oysterical content of at least one element in the fruit core. At least one element Isotope ratio. Among them, elemental content includes carbon, nitrogen, and sulfur content, and the isotope ratio comprises nitrogen, carbon, sulfur, hydrogen, oxygen or the like stable isotope ratio. As a preferred solution, isotope analytical data comprises2H isotope ratio,18O Isotope ratio,13C isotope ratio,15N isotope ratio,34S isotope ratio and N element content, S element content and C element content, and in the pulp2H isotope ratio,18O Isotope ratio,13C isotope ratio,15N isotope ratio,34The S isotope ratio and the N-element content, S element content and C element content.
[0049]In the embodiment of the present invention, the above-dependent analysis data is directly analyzed by chemical component analysis on the durian of the target origin. Specific analysis process, for example, when analyzing Malaysia fresh durian samples, first remove the flesh of fresh durian with the fruit core, and then washed after purge, ultra-pure water (resistivity>18.2MΩ · cm, 20 ° C) Washed and dry. The pulp was washed three times with acetone three times to remove sugar in the flesh, and then washed three times with ultrapure water. The rod core and the flesh were placed in a 60 ° C oven to dry to constant weight (such as MM400 ball mill, Lai Chi, Germany Grinding into a powder, saving after 80 sieves in a dry environment. The above-mentioned durian powder, while performing carbon, nitrogen, sulfur stable isotope and content analysis, can be carbon with a stable isotope ratio mass spectrometer with an element analyzer (such as Delta V Advantage, Saima Fei, Germany) Nitrogen isotope and content determination. Hydrogen oxygen-stable isotope ratio analysis can be determined using a stable isotope compared to mass spectrometry (such as Delta V Advantage, Saima Fei Shil, Germany) and element analyzer (HT2000). The above instruments can be connected to the computer device over a network so that the computer device directly reads the isotope analysis data obtained by the instrument analysis.
[0050]Step S12: The traceability model of the target origin is generated by preset analysis of data using the isotope analysis data and the corresponding target property information.
[0051]In the embodiment of the present invention, after obtaining the isotope analysis data, the computer device will utilize the isotope analysis data and the corresponding target property information by pre-set the traceability model of the target production site. Wherein, the preset analysis modeling algorithm includes at least one algorithm in step discriminant analysis, linear discriminating analysis algorithm, and neural network algorithm. For example, after modeling by a linear discrimination algorithm, the corresponding model coefficient matrix can be obtained, that is, the traceability model.
[0052]In the embodiment of the present invention, traceable models include Malaysia traceable models, Thailand traceability models, Cambodia traceability models, and Vietnamese traceability models. The Malaysian trace source model includes:
[0053]YMalaysia = 5.968X1-37.228X2-0.967x3+ 3.898X4+ 9.127x5+ 14.190X6+ 5.096X7-815.054;
[0054]The Thai trace source model includes:
[0055]YThailand = 4.315x1-37.478X2-1.225x3+ 4.410X4+ 11.772X5+ 11.364X6+ 3.659X7-769.801;
[0056]The Cambodian trace source model includes:
[0057]YCambodia = 7.516X1-40.351X2-1.234x3+ 4.699X4+ 11.963X5+ 14.355X6+ 6.921X7-921.633;
[0058]The Vietnamese trace source model includes:
[0059]YVietnam = 5.607X1-37.671X2-1.092x3+ 5.100X4+ 18.562X5+ 13.524X6+ 5.601X7-864.359;
[0060]In X1D (deuterium) isotope ratio of durian fruit, x2Durian13C (carbon thirteen) isotope ratio, x3D isotope ratio of durian flesh, x4Durian flesh18O (oxygen 18) isotope ratio, X5N element content for durian fruit, x6C element content for durian fruit, x7The N element content of durian flesh, y is the output value. Among them, in practical applications, the adjustment of the coefficients of each traceable model can be performed by more sample training, thereby increasing the accuracy of the traceability model discrimination. When the discriminates of unknown area samples, the X of the unknown area1~ X7In the above four calculations, the output value Y is the maximum target origin is determined.
[0061]Step S13: The accuracy verification of the traceable model is obtained by using the isotope analysis data of the target origin, and the isotope analysis data of other originats. The accuracy of the source model is obtained.
[0062]In the embodiment of the present invention, after generating the traceability model, it is also possible to analyze the number of traceability models by entering the isotope analysis data of a plurality of other durian samples, and the isotope analysis data of the plurality of corresponding target origins. Statistics to generate the accuracy of the traceability model. Determine whether the traceability model needs to continue sample training according to the accuracy of the traceability model, and it is determined whether the traceability model needs to continue sample training, thereby increasing the accuracy of the traceability model.
[0063]In the embodiment of the present invention, the isotope analysis of the durian of the target origin is performed. After obtaining the isotope analysis data, the corresponding traceability model is generated, and the origin of the durian can be quickly and accurately discriminated by the traceability model, thereby increasing the efficiency of durian origin.

Example Embodiment

[0064]Example 2
[0065]figure 2 It is a flow chart of a durian traceable model modeling method according to Embodiment 2 of the present invention, which includes the following steps:
[0066]Step S21: Get the isotope analysis data of the peat and fruit core of the target.
[0067]This step is consistent with the above step S11, and details will not be described herein.
[0068]Step S22: The traceability model of the target property is generated by the preset analysis modeling algorithm using the isotopic analysis data and the corresponding target property information.
[0069]This step is consistent with the above step S12, and details will not be described herein.
[0070]Step S23: The accuracy verification of the traceable model is obtained by using the isotopic analysis data of the target origin, and the isotope analysis data of other origins. The accuracy of the source model is obtained.
[0071]This step is consistent with the above step S13, and details will not be described herein.
[0072]Step S24: Using the traceability model of multiple target origins, the isotope analysis data of multiple target origins is cross-validated by a cross-validation of a plurality of target origins, and obtains the corresponding accuracy of the traceability model of multiple target origins.
[0073]In the embodiment of the present invention, after the traceability model of multiple target origins is generated according to the above steps, the isotope analysis data of multiple target origins can be retained, and these isotope analysis data can be used to carry out multiple target origins by leaving one method. Cross-validation, so that the confidence of the corresponding accuracy of the traceability model can be further improved.

Example Embodiment

[0074]Example 3
[0075]image 3 It is a flow chart of a durian traceability method provided in Example 3 of the present invention, which includes the following steps:
[0076]Step S31: Get the isotope analysis data of the flesh and fruit core of the sample durian.
[0077]Step S32: Enter the isotope analysis data into a plurality of traceable models established according to the durian traceability model modeling method to acquire the output value of multiple traceable models.
[0078]Step S33: Determine the origin of the traceability model whose output value is the highest output value, the origin of the sample durian.

PUM

no PUM

Description & Claims & Application Information

We can also present the details of the Description, Claims and Application information to help users get a comprehensive understanding of the technical details of the patent, such as background art, summary of invention, brief description of drawings, description of embodiments, and other original content. On the other hand, users can also determine the specific scope of protection of the technology through the list of claims; as well as understand the changes in the life cycle of the technology with the presentation of the patent timeline. Login to view more.

Similar technology patents

Compositing Windowing System

InactiveUS20100058229A1minimal costimprove efficiency
Owner:QUALCOMM INC

Image reading apparatus

InactiveUS20050238205A1improve efficiency
Owner:FUJIFILM BUSINESS INNOVATION CORP

Method and device for updating data in distributed storage system

ActiveCN103294675AImprove efficiencySolve data consistency
Owner:SHANDA INTERACTIVE ENTERTAINMENT

Classification and recommendation of technical efficacy words

  • Improve efficiency
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products