CN113820280A - Agricultural product pesticide residue detection method and system - Google Patents
Agricultural product pesticide residue detection method and system Download PDFInfo
- Publication number
- CN113820280A CN113820280A CN202111086618.0A CN202111086618A CN113820280A CN 113820280 A CN113820280 A CN 113820280A CN 202111086618 A CN202111086618 A CN 202111086618A CN 113820280 A CN113820280 A CN 113820280A
- Authority
- CN
- China
- Prior art keywords
- pesticide residue
- pesticide
- residue detection
- agricultural product
- detection model
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
-
- 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
- G01N21/47—Scattering, i.e. diffuse reflection
- G01N21/4738—Diffuse reflection, e.g. also for testing fluids, fibrous materials
-
- 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/02—Food
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Food Science & Technology (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Medicinal Chemistry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a system for detecting pesticide residues of agricultural products, wherein the method comprises the following steps: obtaining a first production area and a first variety of a first agricultural product; obtaining a first medication information dataset for a first agricultural product of a first variety in a first production area from big data analysis, the first medication information dataset comprising a first medication type dataset and a first medication amount dataset; calling a standard pesticide residue detection model; training a standard pesticide residue detection model by using a first pesticide type data set and a first pesticide dosage data set to obtain a first pesticide residue detection model; obtaining the medication type and the medication dosage of a first agricultural product; and inputting the drug type and the drug dosage of the first agricultural product into the first pesticide residue detection model to obtain a first detection result. The method solves the technical problems of complex pesticide residue detection process, low detection efficiency and single detection method caused by differences of agricultural product varieties, production areas, pesticide types, pesticide use amounts and the like in the prior art.
Description
Technical Field
The invention relates to the field of pesticide detection, in particular to a method and a system for detecting pesticide residues of agricultural products.
Background
The development of agricultural industrialization makes the production of agricultural products depend on exogenous substances such as pesticides, antibiotics and hormones. The amount of agricultural chemicals in agricultural products in China is high, and unreasonable use of agricultural chemicals can lead to the excessive pesticide residue in the agricultural products, which affects the edible safety of consumers, and can cause the consumers to have diseases and abnormal development in serious cases, even directly cause poisoning and death. The overproof pesticide residue also affects the trade of agricultural products, so that the export of agricultural products in China faces a serious challenge. The method is characterized in that the pesticide residue detection is trace or trace analysis and can be realized only by adopting a high-sensitivity detection technology, the chromatography is one of the common methods for pesticide residue analysis, the separation purpose is achieved according to the difference of distribution coefficients of an analysis substance between a stationary phase and a mobile phase, the concentration of the analysis substance is converted into an electric signal which is easy to measure, and then the electric signal is sent to a recorder for recording.
However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the inventors of the present application find that the above-mentioned technology has at least the following technical problems:
in the prior art, the technical problems of complex pesticide residue detection process, low detection efficiency and single detection method exist due to differences of agricultural product varieties, production areas, pesticide types, pesticide use amounts and the like.
Disclosure of Invention
Aiming at the defects in the prior art, the embodiment of the application aims to solve the technical problems of complex pesticide residue detection process, low detection efficiency and single detection method caused by differences of agricultural product varieties, production areas, pesticide types, pesticide use amounts and the like in the prior art by providing the agricultural product pesticide residue detection method and system. The method achieves the technical effects of analyzing the types and the attributes of the residual pesticides of the agricultural products by collecting data information such as the origin attribute information, the medication information and the like of the agricultural products, thereby reducing the complexity of the detection process, improving the detection efficiency and the detection accuracy and enriching the detection means.
In a first aspect, an embodiment of the present application provides an agricultural product pesticide residue detection method, where the method includes: obtaining a first production area and a first variety of a first agricultural product; obtaining a first medication information dataset for the first agricultural product of the first variety in the first production area according to big data analysis, the first medication information dataset comprising a first medication type dataset and a first medication amount dataset; calling a standard pesticide residue detection model; training the standard pesticide residue detection model by using the first pesticide type data set and the first pesticide amount data set to obtain a first pesticide residue detection model; obtaining a medication type and a medication dose for the first agricultural product; and inputting the medication type and the medication dose of the first agricultural product into the first pesticide residue detection model to obtain a first detection result.
In another aspect, the present application further provides a system for detecting pesticide residues on agricultural products, wherein the system includes: a first obtaining unit for obtaining a first production area and a first variety of a first agricultural product; a second obtaining unit for obtaining a first medication information dataset for the first agricultural product of the first variety in the first production area based on big data analysis, the first medication information dataset comprising a first medication type dataset and a first medication amount dataset; the first calling unit is used for calling a standard pesticide residue detection model; a third obtaining unit, configured to train the standard pesticide residue detection model by using the first pesticide type data set and the first pesticide amount data set, so as to obtain a first pesticide residue detection model; a fourth obtaining unit for obtaining a medication type and a medication dose of the first agricultural product; and the fifth obtaining unit is used for inputting the medication type and the medication dosage of the first agricultural product into the first pesticide residue detection model to obtain a first detection result.
In another aspect, the present invention provides an agricultural product pesticide residue detection system, which includes a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method according to the first aspect when executing the program.
One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:
obtaining a first production area and a first variety of a first agricultural product; obtaining a first medication information dataset for the first agricultural product of the first variety in the first production area according to big data analysis, the first medication information dataset comprising a first medication type dataset and a first medication amount dataset; calling a standard pesticide residue detection model; training the standard pesticide residue detection model by using the first pesticide type data set and the first pesticide amount data set to obtain a first pesticide residue detection model; obtaining a medication type and a medication dose for the first agricultural product; and inputting the medication type and the medication dose of the first agricultural product into the first pesticide residue detection model to obtain a first detection result. Based on the method, the agricultural product pesticide residue detection method can be constructed, and the purpose of analyzing the types and the attributes of pesticide residues in agricultural products by collecting data information such as production place attribute information and drug use information of the agricultural products is achieved, so that the complexity of the detection process is reduced, the detection efficiency and the detection accuracy are improved, and the technical effect of detection means is enriched.
The foregoing description is only an overview of the technical solutions of the present application, and the present application can be implemented according to the content of the description in order to make the technical means of the present application more clearly understood, and the following detailed description of the present application is given in order to make the above and other objects, features, and advantages of the present application more clearly understandable.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:
FIG. 1 is a schematic flow chart of a method for detecting pesticide residues of agricultural products according to an embodiment of the present application;
FIG. 2 is a schematic flow chart illustrating the analysis of the pesticide type data and the pesticide dosage data in the method for detecting pesticide residues in agricultural products according to the embodiment of the present application;
FIG. 3 is a schematic flow chart illustrating the updating of the pesticide residue detection model in the pesticide residue detection method for agricultural products according to the embodiment of the present application;
fig. 4 is a schematic flow chart illustrating the optimization of the third pesticide residue detection model in the pesticide residue detection method for agricultural products according to the embodiment of the present application;
FIG. 5 is a schematic flow chart illustrating a data loss analysis performed on a third detection result in the agricultural product pesticide residue detection method according to the embodiment of the present application;
FIG. 6 is a schematic flow chart illustrating the analysis of the collected image information in the agricultural product pesticide residue detection method according to the embodiment of the present application;
FIG. 7 is a schematic flow chart illustrating a diffuse reflection uniformity analysis in a method for detecting pesticide residues of agricultural products according to an embodiment of the present application;
FIG. 8 is a schematic structural diagram of a system for detecting pesticide residues in agricultural products according to an embodiment of the present application;
fig. 9 is a schematic structural diagram of an exemplary electronic device according to an embodiment of the present application.
Description of reference numerals: a first obtaining unit 11, a second obtaining unit 12, a first calling unit 13, a third obtaining unit 14, a fourth obtaining unit 15, a fifth obtaining unit 16, a bus 300, a receiver 301, a processor 302, a transmitter 303, a memory 304, and a bus interface 305.
Detailed Description
The embodiment of the application provides a method and a system for detecting pesticide residues of agricultural products, and solves the technical problems that in the prior art, due to differences of agricultural product varieties, production areas, pesticide types, pesticide usage amount and the like, the pesticide residue detection process is complex, the detection efficiency is low, and the detection method is single. The method achieves the technical effects of analyzing the types and the attributes of the residual pesticides of the agricultural products by collecting data information such as the origin attribute information, the medication information and the like of the agricultural products, thereby reducing the complexity of the detection process, improving the detection efficiency and the detection accuracy and enriching the detection means.
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are merely some embodiments of the present application and not all embodiments of the present application, and it should be understood that the present application is not limited to the example embodiments described herein.
Summary of the application
The development of agricultural industrialization makes the production of agricultural products depend on exogenous substances such as pesticides, antibiotics and hormones. The amount of agricultural chemicals in agricultural products in China is high, and unreasonable use of agricultural chemicals can lead to the excessive pesticide residue in the agricultural products, which affects the edible safety of consumers, and can cause the consumers to have diseases and abnormal development in serious cases, even directly cause poisoning and death. The overproof pesticide residue also affects the trade of agricultural products, so that the export of agricultural products in China faces a serious challenge. The method is characterized in that the pesticide residue detection is trace or trace analysis and can be realized only by adopting a high-sensitivity detection technology, the chromatography is one of the common methods for pesticide residue analysis, the separation purpose is achieved according to the difference of distribution coefficients of an analysis substance between a stationary phase and a mobile phase, the concentration of the analysis substance is converted into an electric signal which is easy to measure, and then the electric signal is sent to a recorder for recording. In the prior art, the technical problems of complex pesticide residue detection process, low detection efficiency and single detection method exist due to differences of agricultural product varieties, production areas, pesticide types, pesticide use amounts and the like.
In view of the above technical problems, the technical solution provided by the present application has the following general idea:
the embodiment of the application provides a method for detecting pesticide residues of agricultural products, wherein the method comprises the following steps: obtaining a first production area and a first variety of a first agricultural product; obtaining a first medication information dataset for the first agricultural product of the first variety in the first production area according to big data analysis, the first medication information dataset comprising a first medication type dataset and a first medication amount dataset; calling a standard pesticide residue detection model; training the standard pesticide residue detection model by using the first pesticide type data set and the first pesticide amount data set to obtain a first pesticide residue detection model; obtaining a medication type and a medication dose for the first agricultural product; and inputting the medication type and the medication dose of the first agricultural product into the first pesticide residue detection model to obtain a first detection result.
For better understanding of the above technical solutions, the following detailed descriptions will be provided in conjunction with the drawings and the detailed description of the embodiments.
Example one
As shown in fig. 1, an embodiment of the present application provides a method for detecting pesticide residues on agricultural products, where the method includes:
step S100: obtaining a first production area and a first variety of a first agricultural product;
step S200: obtaining a first medication information dataset for the first agricultural product of the first variety in the first production area according to big data analysis, the first medication information dataset comprising a first medication type dataset and a first medication amount dataset;
specifically, the first agricultural product refers to any agricultural product needing to detect pesticide residues, a first production area and a first variety of the first agricultural product are obtained, namely specific production area and variety information of the first agricultural product are collected, a first medicine information data set of the first variety in the first production area of the first agricultural product is obtained through a big data analysis result, the first medicine information data set comprises the first medicine type data set and the first medicine amount data set, and the type and the medicine amount of pesticide used by the agricultural product of the variety from the same production area in the past are collected. By constructing the first medicine information data set, the historical medication habits of the first agricultural products in the first production area can be mastered, and real and reliable historical data can be obtained, so that a foundation is laid for the subsequent pesticide residue determination of the same sample.
Step S300: calling a standard pesticide residue detection model;
step S400: training the standard pesticide residue detection model by using the first pesticide type data set and the first pesticide amount data set to obtain a first pesticide residue detection model;
specifically, the first pesticide type data set and the first pesticide amount data set of the first variety of the first agricultural product are input into the standard pesticide residue detection model, the standard pesticide residue detection model is constructed on the basis of a neural network model, the neural network is an operation model formed by connecting a large number of neurons, the output of the network is expressed according to a logic strategy of a network connection mode, the output information is more accurate through model training, and the first pesticide type data set and the first pesticide amount data set are input into the standard pesticide residue detection model to perform comprehensive analysis on pesticide residue, so that the first pesticide residue detection model is obtained.
Further, the training process is essentially a supervised learning process, each group of supervised data includes the first drug type data set and the first drug amount data set information, the standard pesticide residue detection model performs continuous self-correction and adjustment until the obtained output result is consistent with the identification information, the group of supervised learning is finished, and the next group of supervised learning is performed. When the output information of the standard pesticide residue detection model reaches a preset accuracy rate or reaches a convergence state, the supervised learning process is ended, and the technical effect of improving the intelligent degree of data training is achieved.
Step S500: obtaining a medication type and a medication dose for the first agricultural product;
step S600: and inputting the medication type and the medication dose of the first agricultural product into the first pesticide residue detection model to obtain a first detection result.
Specifically, the specific medication type and the medication dose of the first agricultural product are obtained, and the medication type and the medication dose of the first agricultural product are input into the first pesticide residue detection model to obtain a first detection result. The first pesticide residue detection model is used for detecting the first variety of the first agricultural product in the first production area, such as a whitewater red fuji pesticide residue detection model, for red fuji apples in white county, Weinan, Shanxi, by knowing specific pesticide application types and pesticide application doses, the first pesticide residue detection model can generate optimal test conditions and determine an optimal detection scheme, so that the effects of accurate detection and detection accuracy improvement are achieved.
Further, as shown in fig. 2, the embodiment of the present application further includes:
step S610: obtaining a second medication information data set of all agricultural products in an agricultural product set in the first production area according to big data analysis, wherein the second medication information data set comprises a second medication type data set and a second medication amount data set, and the agricultural product set comprises other agricultural products different from the first agricultural product in the first production area;
step S620: adding masks to the second drug type data set and the second drug dosage data set, inputting the masks to the standard pesticide residue detection model, and training to obtain a second pesticide residue detection model;
step S630: obtaining a third pesticide residue detection model according to the first pesticide residue detection model and the second pesticide residue detection model;
step S640: and inputting the medication type and the medication dosage of the first agricultural product into the third pesticide residue detection model to obtain a second detection result.
Specifically, a second medication information data set of all agricultural products in an agricultural product set in the first production area is obtained according to big data analysis, and the second medication information data set comprises a second medication type data set and a second medication amount data set, wherein the agricultural product set comprises other agricultural products which are different from the first agricultural product in the first production area, namely if the first agricultural product is an apple, the second medication information data set comprises medication information data of other agricultural products except the apple. And adding a mask code to the second drug type data set and the second drug dosage data set, inputting the second drug type data set and the second drug dosage data set into the standard pesticide residue detection model, and training to obtain a second pesticide residue detection model, wherein the mask code is a string of binary codes for performing bit AND operation on a target field, and the data security can be ensured.
Further, a third pesticide residue detection model is obtained according to the first pesticide residue detection model and the second pesticide residue detection model, and the pesticide type and pesticide dosage of the first agricultural product are input into the third pesticide residue detection model to obtain a second detection result. Because the agricultural product sample characteristics of the first production area are overlapped more, the user overlap is less, namely the light energy resource, the heat resource and the water resource are very similar when the agricultural product is planted, but the agricultural product varieties are different, the common modeling of the agricultural products of different varieties is realized, the process can be regarded as sample-based distributed model training, in other words, all data are distributed to different machines, each machine downloads the model from the server, then the model is trained by using local data, and then the model is returned to the server for the parameters needing to be updated; the server aggregates the returned parameter update models on each machine and feeds back the latest model to each machine. In the process, the same and complete model, namely the third pesticide residue detection model, is arranged under each machine, the machines are not independent of each other, each machine can also independently predict in the prediction process, and the process can be regarded as sample-based distributed model training. The common modeling of agricultural products of different varieties is realized, and the effect of improving the artificial intelligence model is achieved.
Further, as shown in fig. 3, the step S630 of obtaining a third pesticide residue detection model according to the first pesticide residue detection model and the second pesticide residue detection model includes:
step S631: obtaining a first model parameter according to the first pesticide residue detection model;
step S632: obtaining a second model parameter according to the second pesticide residue detection model;
step S633: and updating the standard pesticide residue detection model according to the first model parameter and the second model parameter to obtain a third pesticide residue detection model.
Specifically, a first model parameter is obtained according to the first pesticide residue detection model, a second model parameter is obtained according to the second pesticide residue detection model, the first model parameter has the variety characteristic of the first agricultural product, the second model parameter has the product characteristic of other agricultural products except the first agricultural product, the standard pesticide residue detection model is further updated according to the first model parameter and the second model parameter, the first model parameter and the second model parameter are used as training data, the standard pesticide residue detection model is trained, a third pesticide residue detection model is obtained, and the trained model enables the output pesticide residue detection result to be more accurate and reliable.
Further, as shown in fig. 4, the embodiment of the present application further includes:
step S650: obtaining a production range for the first agricultural product;
step S660: obtaining all first agricultural product sets in the production range;
step S670: obtaining a third medical information data set of all first agricultural products in the first agricultural product set according to the first agricultural product set, wherein the third medical information data set comprises a third medical type data set and a third medicament amount data set;
step S680: performing incremental learning on the third pesticide residue detection model according to the third pesticide type data set and the third pesticide amount data set to obtain a fourth pesticide residue detection model;
step S690: and obtaining a fourth detection result according to the fourth pesticide residue detection model.
Specifically, the production area of the first agricultural product is large, the production range of the first agricultural product is obtained, all first agricultural product sets are obtained from the production range, all production information of the first agricultural product is covered in the sets, further, according to the first agricultural product sets, third medicine information data sets of all the first agricultural products in the first agricultural product sets are obtained, the third medicine information data sets comprise third medicine type data sets and third medicine quantity data sets, and pesticide use types and dosage data of the same product in different production areas are obtained. And performing incremental learning on the third pesticide residue detection model according to the third pesticide type data set and the third pesticide amount data set to obtain a fourth pesticide residue detection model, and obtaining a fourth detection result according to the fourth pesticide residue detection model.
The third pesticide residue detection model is a corresponding evaluation model obtained by updating the standard pesticide residue detection model and performing machine learning based on the first model parameter and the second model parameter, the first model parameter and the second model parameter need to perform secondary incremental learning, and since screening needs to be combined with old training data of the third pesticide residue detection model to complete a comprehensive incremental learning result, after the incremental learning is performed on the first model parameter and the second model parameter, the basic performance of the third pesticide residue detection model can be kept, corresponding incremental learning is completed, and further corresponding output information, namely a fourth detection result is obtained based on the fourth pesticide residue detection model, wherein the fourth pesticide residue detection model is an updated model after the incremental learning, so that the incremental learning of the delay characteristic is achieved, so as to improve the technical effect of updating performance of the screening result.
Further, as shown in fig. 5, in the step S680, performing incremental learning on the third pesticide residue detection model according to the third pesticide type data set and the third pesticide amount data set to obtain a fourth pesticide residue detection model, the step S680 includes:
step S681: inputting the third pesticide type data set and the third pesticide dosage data set into the third pesticide residue detection model to obtain a third detection result;
step S682: obtaining first loss data by performing data loss analysis on the third detection result;
step S683: and inputting the first loss data into the third pesticide residue detection model for training to obtain the fourth pesticide residue detection model.
Specifically, the third pesticide residue detection model is obtained by performing data training based on the first model parameter and the second model parameter, so that data loss analysis is completed by introducing a loss function to obtain the first loss data, wherein the first loss data represents data knowledge loss data of the third pesticide residue detection model related to the first model parameter and the second model parameter, and incremental learning of the third pesticide residue detection model is completed based on the first loss data, and since the third pesticide residue detection model is obtained by updating and performing machine learning on the standard pesticide residue detection model based on the first model parameter and the second model parameter, the fourth pesticide residue detection model retains the basic function of the third pesticide residue detection model through training of loss data, and the continuous updating performance of the model is maintained, so that the updating performance of the value evaluation is improved, and the technical effect of ensuring the accuracy of the updated value evaluation result is achieved.
Further, as shown in fig. 6, the embodiment of the present application further includes:
step S710: obtaining first image information comprising a pesticide contact site of the first agricultural product;
step S720: performing image segmentation on the first image information to obtain regionalized first image information;
step S730: performing feature extraction on the regionalized first image information according to color features to obtain a color feature traversal result;
step S740: performing feature extraction on the regionalized first image information according to integrity features to obtain an integrity feature traversal result;
step S750: inputting the color feature traversal result and the integrity feature traversal result into an agricultural residue evaluation model to obtain a first agricultural residue evaluation result;
step S760: and obtaining a comprehensive pesticide residue detection result according to the first pesticide residue evaluation result and the first detection result.
Specifically, the first image information is obtained by an image acquisition device, the first image information includes external information characteristics of the first agricultural product, the first image information includes pesticide contact parts of the first agricultural product, such as fruit bases, fruit faces and the like, image segmentation is carried out on the first image information, the first image information is divided into regions according to the characteristics of the first agricultural product, and the segmented image is obtained, namely the regional first image information is obtained. Further, feature extraction is carried out on the regionalized first image information according to color features, and a color feature traversal result is obtained. After the pesticide is sprayed, the appearance of the agricultural product is obviously different from that of a product which is not sprayed with the pesticide, wherein the color is one of the most obvious appearance characteristics of the agricultural product, and after the characteristic extraction is carried out, a color characteristic traversal result is obtained.
Further, feature extraction is carried out on the regional first image information according to integrity features to obtain an integrity feature traversal result, wherein the integrity features refer to whether insect eyes exist on the surface of the agricultural product or not, and whether pesticide is used or not can be judged according to whether the insect eyes exist or not. Inputting the color feature traversal result and the integrity feature traversal result into an agricultural residue evaluation model to obtain a first agricultural residue evaluation result, and obtaining a comprehensive agricultural residue detection result according to the first agricultural residue evaluation result and the first detection result. The first detection result can be supplemented by obtaining the first pesticide residue evaluation result, and whether pesticide is used or not and the pesticide dosage can be roughly evaluated through the appearance of the agricultural product, so that the pesticide residue detection result which is strong in comprehensiveness, scientific and reliable is obtained by combining the first detection result.
Further, as shown in fig. 7, the embodiment of the present application further includes:
step S761: according to the first image information, obtaining first epidermis diffuse reflection uniformity information of the first agricultural product under a first preset lighting condition;
step S762: obtaining standard epidermal diffuse reflection uniformity information of the first agricultural product;
step S763: performing data compensation on the first epidermis diffuse reflection uniformity information according to the standard epidermis diffuse reflection uniformity information to obtain second epidermis diffuse reflection uniformity information;
step S764: and correcting the first pesticide residue evaluation result according to the second epidermis diffuse reflection uniformity information to obtain a second pesticide residue evaluation result.
Specifically, a certain lighting condition is preset, and the first epidermis diffuse reflection uniformity information of the first agricultural product under the first preset lighting condition is obtained, wherein the diffuse reflection is a phenomenon that light projected on a rough surface is reflected towards all directions. The diffuse reflection uniformity information can reflect the roughness of the first skin to obtain standard skin diffuse reflection uniformity information of the first agricultural product, wherein the standard skin diffuse reflection uniformity information refers to the diffuse reflection uniformity of the first agricultural product within the range that the pesticide residue meets the national standard. And performing data compensation on the first epidermis diffuse reflection uniformity information according to the standard epidermis diffuse reflection uniformity information to obtain second epidermis diffuse reflection uniformity information, namely compensating the first epidermis diffuse reflection uniformity information by the standard epidermis diffuse reflection uniformity information, thereby eliminating some interference and reducing system errors. And further, correcting the first pesticide residue evaluation result according to the second epidermis diffuse reflection uniformity information to obtain a second pesticide residue evaluation result. And incorporating the second epidermis diffuse reflection uniformity information into an agricultural residue evaluation method to obtain a second agricultural residue evaluation result, and thus obtaining a comprehensive agricultural residue detection result by combining the first detection result. The auxiliary detection of pesticide residues can be realized by an image acquisition technology, the detection means are enriched, and the comprehensiveness of the detection method is improved.
Compared with the prior art, the invention has the following beneficial effects:
1. obtaining a first production area and a first variety of a first agricultural product; obtaining a first medication information dataset for the first agricultural product of the first variety in the first production area according to big data analysis, the first medication information dataset comprising a first medication type dataset and a first medication amount dataset; calling a standard pesticide residue detection model; training the standard pesticide residue detection model by using the first pesticide type data set and the first pesticide amount data set to obtain a first pesticide residue detection model; obtaining a medication type and a medication dose for the first agricultural product; and inputting the medication type and the medication dose of the first agricultural product into the first pesticide residue detection model to obtain a first detection result. Based on the method, the agricultural product pesticide residue detection method can be constructed, and the purpose of analyzing the types and the attributes of pesticide residues in agricultural products by collecting data information such as production place attribute information and drug use information of the agricultural products is achieved, so that the complexity of the detection process is reduced, the detection efficiency and the detection accuracy are improved, and the technical effect of detection means is enriched.
2. Due to the adoption of the image information acquisition technology, the color characteristic traversal result and the epidermis diffuse reflection uniformity information are deeply analyzed, so that the effects of providing an auxiliary detection means and improving the accuracy of the detection result are achieved.
Example two
Based on the same inventive concept as the agricultural product pesticide residue detection method in the foregoing embodiment, the present invention further provides an agricultural product pesticide residue detection system, as shown in fig. 8, the system includes:
a first obtaining unit 11, wherein the first obtaining unit 11 is used for obtaining a first production area and a first variety of a first agricultural product;
a second obtaining unit 12, the second obtaining unit 12 configured to obtain a first medication information dataset of the first agricultural product of the first variety in the first production area according to big data analysis, the first medication information dataset comprising a first medication type dataset and a first medication amount dataset;
the first calling unit 13 is used for calling a standard pesticide residue detection model;
a third obtaining unit 14, where the third obtaining unit 14 is configured to train the standard pesticide residue detection model by using the first pesticide type data set and the first pesticide amount data set, so as to obtain a first pesticide residue detection model;
a fourth obtaining unit 15, wherein the fourth obtaining unit 15 is used for obtaining the medication type and the medication dosage of the first agricultural product;
a fifth obtaining unit 16, where the fifth obtaining unit 16 is configured to input the medication type and the medication dosage of the first agricultural product into the first pesticide residue detection model, so as to obtain a first detection result.
Further, the system further comprises:
a sixth obtaining unit, configured to obtain, according to big data analysis, a second medication information data set of all agricultural products in a set of agricultural products in the first production area, where the second medication information data set includes a second medication type data set and a second medication amount data set, and the set of agricultural products includes other agricultural products in the first production area that are different from the first agricultural product;
a seventh obtaining unit, configured to add a mask to the second drug type data set and the second drug amount data set, and input the mask to the standard pesticide residue detection model for training to obtain a second pesticide residue detection model;
an eighth obtaining unit, configured to obtain a third pesticide residue detection model according to the first pesticide residue detection model and the second pesticide residue detection model;
a ninth obtaining unit, configured to input the medication type and the medication dose of the first agricultural product into the third pesticide residue detection model, and obtain a second detection result.
Further, the system further comprises:
a tenth obtaining unit, configured to obtain a first model parameter according to the first pesticide residue detection model;
an eleventh obtaining unit, configured to obtain a second model parameter according to the second pesticide residue detection model;
and the twelfth obtaining unit is used for updating the standard pesticide residue detection model according to the first model parameter and the second model parameter to obtain a third pesticide residue detection model.
Further, the system further comprises:
a thirteenth obtaining unit for obtaining all first set of agricultural products in the production range;
a fourteenth obtaining unit, configured to obtain, according to the first set of agricultural products, a third medical information dataset of all first agricultural products in the first set of agricultural products, where the third medical information dataset includes a third medical type dataset and a third medical dose dataset;
a fifteenth obtaining unit, configured to perform incremental learning on the third pesticide residue detection model according to the third pesticide type data set and the third pesticide quantity data set, and obtain a fourth pesticide residue detection model;
and the sixteenth obtaining unit is used for obtaining a fourth detection result according to the fourth pesticide residue detection model.
Further, the system further comprises:
a seventeenth obtaining unit, configured to input the third drug type data set and the third drug amount data set into the third pesticide residue detection model, and obtain a third detection result;
an eighteenth obtaining unit configured to obtain first loss data by performing data loss analysis on the third detection result;
a nineteenth obtaining unit, configured to input the first loss data into the third pesticide residue detection model for training, and obtain the fourth pesticide residue detection model.
Further, the system further comprises:
a twentieth obtaining unit for obtaining first image information including a pesticide contact site of the first agricultural product;
a twenty-first obtaining unit, configured to perform image segmentation on the first image information to obtain localized first image information;
a twenty-second obtaining unit, configured to perform feature extraction on the localized first image information according to color features, and obtain a color feature traversal result;
a twenty-third obtaining unit, configured to perform feature extraction on the localized first image information according to a completeness feature, and obtain a completeness feature traversal result;
a twenty-fourth obtaining unit, configured to input the color feature traversal result and the integrity feature traversal result into an pesticide residue evaluation model, so as to obtain a first pesticide residue evaluation result;
and the twenty-fifth obtaining unit is used for obtaining a comprehensive pesticide residue detection result according to the first pesticide residue evaluation result and the first detection result.
Further, the system further comprises:
a twenty-sixth obtaining unit, configured to obtain, according to the first image information, first epidermis diffuse reflection uniformity information of the first agricultural product under a first predetermined lighting condition;
a twenty-seventh obtaining unit, configured to obtain standard epidermal diffuse reflection uniformity information of the first agricultural product;
a twenty-eighth obtaining unit, configured to perform data compensation on the first epidermis diffuse reflection uniformity information according to the standard epidermis diffuse reflection uniformity information, so as to obtain second epidermis diffuse reflection uniformity information;
and the twenty-ninth obtaining unit is used for correcting the first pesticide residue evaluation result according to the second epidermis diffuse reflection uniformity information to obtain a second pesticide residue evaluation result.
Various changes and specific examples of the agricultural product pesticide residue detection method in the first embodiment of fig. 1 are also applicable to the agricultural product pesticide residue detection system in the present embodiment, and through the foregoing detailed description of the agricultural product pesticide residue detection method, those skilled in the art can clearly know the implementation method of the agricultural product pesticide residue detection system in the present embodiment, so for the brevity of the description, detailed description is not repeated.
EXAMPLE III
The electronic apparatus of the embodiment of the present application is described below with reference to fig. 9.
Fig. 9 illustrates a schematic structural diagram of an electronic device according to an embodiment of the present application.
Based on the inventive concept of the agricultural product pesticide residue detection method in the previous embodiment, the invention also provides an agricultural product pesticide residue detection system, wherein a computer program is stored on the agricultural product pesticide residue detection system, and when the computer program is executed by a processor, the steps of any one of the methods of the agricultural product pesticide residue detection system are realized.
Where in fig. 9 a bus architecture (represented by bus 300), bus 300 may include any number of interconnected buses and bridges, bus 300 linking together various circuits including one or more processors, represented by processor 302, and memory, represented by memory 304. The bus 300 may also link together various other circuits such as peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further herein. A bus interface 305 provides an interface between the bus 300 and the receiver 301 and transmitter 303. The receiver 301 and the transmitter 303 may be the same element, i.e., a transceiver, providing a means for communicating with various other systems over a transmission medium. The processor 302 is responsible for managing the bus 300 and general processing, and the memory 304 may be used for storing data used by the processor 302 in performing operations.
The embodiment of the application provides a method for detecting pesticide residues of agricultural products, wherein the method comprises the following steps: obtaining a first production area and a first variety of a first agricultural product; obtaining a first medication information dataset for the first agricultural product of the first variety in the first production area according to big data analysis, the first medication information dataset comprising a first medication type dataset and a first medication amount dataset; calling a standard pesticide residue detection model; training the standard pesticide residue detection model by using the first pesticide type data set and the first pesticide amount data set to obtain a first pesticide residue detection model; obtaining a medication type and a medication dose for the first agricultural product; and inputting the medication type and the medication dose of the first agricultural product into the first pesticide residue detection model to obtain a first detection result.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a system for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction system which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (9)
1. A method for detecting agricultural product pesticide residues, wherein the method comprises the following steps:
obtaining a first production area and a first variety of a first agricultural product;
obtaining a first medication information dataset for the first agricultural product of the first variety in the first production area according to big data analysis, the first medication information dataset comprising a first medication type dataset and a first medication amount dataset;
calling a standard pesticide residue detection model;
training the standard pesticide residue detection model by using the first pesticide type data set and the first pesticide amount data set to obtain a first pesticide residue detection model;
obtaining a medication type and a medication dose for the first agricultural product;
and inputting the medication type and the medication dose of the first agricultural product into the first pesticide residue detection model to obtain a first detection result.
2. The method of claim 1, wherein the method further comprises:
obtaining a second medication information data set of all agricultural products in an agricultural product set in the first production area according to big data analysis, wherein the second medication information data set comprises a second medication type data set and a second medication amount data set, and the agricultural product set comprises other agricultural products different from the first agricultural product in the first production area;
adding masks to the second drug type data set and the second drug dosage data set, inputting the masks to the standard pesticide residue detection model, and training to obtain a second pesticide residue detection model;
obtaining a third pesticide residue detection model according to the first pesticide residue detection model and the second pesticide residue detection model;
and inputting the medication type and the medication dosage of the first agricultural product into the third pesticide residue detection model to obtain a second detection result.
3. The method of claim 2, wherein the obtaining a third pesticide residue detection model based on the first pesticide residue detection model and the second pesticide residue detection model comprises:
obtaining a first model parameter according to the first pesticide residue detection model;
obtaining a second model parameter according to the second pesticide residue detection model;
and updating the standard pesticide residue detection model according to the first model parameter and the second model parameter to obtain a third pesticide residue detection model.
4. The method of claim 2, wherein the method comprises:
obtaining a production range for the first agricultural product;
obtaining all first agricultural product sets in the production range;
obtaining a third medical information data set of all first agricultural products in the first agricultural product set according to the first agricultural product set, wherein the third medical information data set comprises a third medical type data set and a third medicament amount data set;
performing incremental learning on the third pesticide residue detection model according to the third pesticide type data set and the third pesticide amount data set to obtain a fourth pesticide residue detection model;
and obtaining a fourth detection result according to the fourth pesticide residue monitoring model.
5. The method of claim 4, wherein the incrementally learning the third pesticide residue detection model from the third drug type dataset and the third drug quantity dataset to obtain a fourth pesticide residue detection model comprises:
inputting the third pesticide type data set and the third pesticide dosage data set into the third pesticide residue detection model to obtain a third detection result;
obtaining first loss data by performing data loss analysis on the third detection result;
and inputting the first loss data into the third pesticide residue detection model for training to obtain the fourth pesticide residue detection model.
6. The method of claim 1, wherein the method comprises:
obtaining first image information comprising a pesticide contact site of the first agricultural product;
performing image segmentation on the first image information to obtain regionalized first image information;
performing feature extraction on the regionalized first image information according to color features to obtain a color feature traversal result;
performing feature extraction on the regionalized first image information according to integrity features to obtain an integrity feature traversal result;
inputting the color feature traversal result and the integrity feature traversal result into an agricultural residue evaluation model to obtain a first agricultural residue evaluation result;
and obtaining a comprehensive pesticide residue detection result according to the first pesticide residue evaluation result and the first detection result.
7. The method of claim 6, wherein the method further comprises:
according to the first image information, obtaining first epidermis diffuse reflection uniformity information of the first agricultural product under a first preset lighting condition;
obtaining standard epidermal diffuse reflection uniformity information of the first agricultural product;
performing data compensation on the first epidermis diffuse reflection uniformity information according to the standard epidermis diffuse reflection uniformity information to obtain second epidermis diffuse reflection uniformity information;
and correcting the first pesticide residue evaluation result according to the second epidermis diffuse reflection uniformity information to obtain a second pesticide residue evaluation result.
8. An agricultural product pesticide residue detection system, wherein the system comprises:
a first obtaining unit for obtaining a first production area and a first variety of a first agricultural product;
a second obtaining unit for obtaining a first medication information dataset for the first agricultural product of the first variety in the first production area based on big data analysis, the first medication information dataset comprising a first medication type dataset and a first medication amount dataset;
the first calling unit is used for calling a standard pesticide residue detection model;
a third obtaining unit, configured to train the standard pesticide residue detection model by using the first pesticide type data set and the first pesticide amount data set, so as to obtain a first pesticide residue detection model;
a fourth obtaining unit for obtaining a medication type and a medication dose of the first agricultural product;
and the fifth obtaining unit is used for inputting the medication type and the medication dosage of the first agricultural product into the first pesticide residue detection model to obtain a first detection result.
9. An agricultural product pesticide residue detection system comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111086618.0A CN113820280A (en) | 2021-09-16 | 2021-09-16 | Agricultural product pesticide residue detection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111086618.0A CN113820280A (en) | 2021-09-16 | 2021-09-16 | Agricultural product pesticide residue detection method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113820280A true CN113820280A (en) | 2021-12-21 |
Family
ID=78914740
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111086618.0A Withdrawn CN113820280A (en) | 2021-09-16 | 2021-09-16 | Agricultural product pesticide residue detection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113820280A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114758223A (en) * | 2022-03-08 | 2022-07-15 | 深圳市五谷网络科技有限公司 | Pesticide use monitoring and early warning method and device, terminal equipment and storage medium |
| CN115758888A (en) * | 2022-11-17 | 2023-03-07 | 厦门智康力奇数字科技有限公司 | Agricultural product safety risk assessment method based on fusion of multi-machine learning algorithm |
| CN117688452A (en) * | 2024-02-01 | 2024-03-12 | 山东龙奥生物技术有限公司 | Food pesticide residue detection and early warning method and system based on neural network |
-
2021
- 2021-09-16 CN CN202111086618.0A patent/CN113820280A/en not_active Withdrawn
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114758223A (en) * | 2022-03-08 | 2022-07-15 | 深圳市五谷网络科技有限公司 | Pesticide use monitoring and early warning method and device, terminal equipment and storage medium |
| CN115758888A (en) * | 2022-11-17 | 2023-03-07 | 厦门智康力奇数字科技有限公司 | Agricultural product safety risk assessment method based on fusion of multi-machine learning algorithm |
| CN115758888B (en) * | 2022-11-17 | 2024-04-23 | 厦门智康力奇数字科技有限公司 | Agricultural product security risk assessment method based on multi-machine learning algorithm fusion |
| CN117688452A (en) * | 2024-02-01 | 2024-03-12 | 山东龙奥生物技术有限公司 | Food pesticide residue detection and early warning method and system based on neural network |
| CN117688452B (en) * | 2024-02-01 | 2024-05-07 | 山东龙奥生物技术有限公司 | Food pesticide residue detection and early warning method and system based on neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113820280A (en) | Agricultural product pesticide residue detection method and system | |
| CN113517760B (en) | Battery energy storage station monitoring method and system based on big data and digital twins | |
| EP3438842A1 (en) | Preprocessor and abnormality predictor diagnosis system | |
| Pepe | An interpretation for the ROC curve and inference using GLM procedures | |
| CN101261604A (en) | Software quality evaluation apparatus and software quality evaluation quantitative analysis method | |
| Bellocchi et al. | Validation of biophysical models: issues and methodologies | |
| CN113406558B (en) | Ammeter misalignment detection method and device based on linear regression and electronic equipment | |
| CN102375895B (en) | Method for automatically constructing parameter optimal extraction process of device model | |
| Da Silva et al. | Modeling of light transmission under heterogeneous forest canopy: an appraisal of the effect of the precision level of crown description | |
| CN100470418C (en) | Polypropylene melting index softsensoring instrument based on blind signal analysis and method thereof | |
| CN112633822A (en) | Asset management method based on digital twin technology, storage medium and mobile terminal | |
| CN117291444A (en) | Digital rural business management method and system | |
| CN116664062A (en) | Substation construction management method, system and equipment based on BIM and GIS | |
| CN116739304A (en) | Production error monitoring system and method based on product history data | |
| CN116738261A (en) | Numerical characteristic discretization attribution analysis method and device based on clustering and binning | |
| CN115775064A (en) | Engineering decision calculation result evaluation method and cloud platform | |
| CN114417549B (en) | Digital twin object field data processing system based on super computer | |
| CN109444055A (en) | A kind of EO-1 hyperion calculation method of alkaline land soil salt content | |
| CN115713691A (en) | Pixel-level electric power popularity estimation method and device based on noctilucent remote sensing | |
| CN114331349A (en) | Scientific research project management method and system based on Internet of things technology | |
| Zhang et al. | Fertilizer sphericity measuring device based on equatorial and meridian circles | |
| Chen et al. | Irregular object measurement method based on improved adaptive slicing method | |
| CN116108312B (en) | Land change area calculation method and device based on single-image-spot result data | |
| Li et al. | Key technology of crop precision sowing based on the vision principle | |
| CN115598309B (en) | Method and system for monitoring and early warning of lead content in atmospheric environment |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20211221 |