CN116777484B - Agricultural and sideline product traceability system based on AI technology - Google Patents
Agricultural and sideline product traceability system based on AI technology Download PDFInfo
- Publication number
- CN116777484B CN116777484B CN202311065871.7A CN202311065871A CN116777484B CN 116777484 B CN116777484 B CN 116777484B CN 202311065871 A CN202311065871 A CN 202311065871A CN 116777484 B CN116777484 B CN 116777484B
- Authority
- CN
- China
- Prior art keywords
- planting
- pesticide spraying
- unit
- representing
- pesticide
- 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.)
- Active
Links
- 238000005516 engineering process Methods 0.000 title claims abstract description 21
- 241000220225 Malus Species 0.000 claims abstract description 41
- 239000000447 pesticide residue Substances 0.000 claims abstract description 32
- 235000021016 apples Nutrition 0.000 claims abstract description 29
- 238000003306 harvesting Methods 0.000 claims abstract description 14
- 238000000034 method Methods 0.000 claims abstract description 8
- 238000000605 extraction Methods 0.000 claims abstract description 4
- 239000000575 pesticide Substances 0.000 claims description 66
- 238000005507 spraying Methods 0.000 claims description 66
- 239000002689 soil Substances 0.000 claims description 36
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 claims description 24
- 239000013598 vector Substances 0.000 claims description 24
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 12
- 229910052757 nitrogen Inorganic materials 0.000 claims description 12
- 229910052698 phosphorus Inorganic materials 0.000 claims description 12
- 239000011574 phosphorus Substances 0.000 claims description 12
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 9
- 238000013527 convolutional neural network Methods 0.000 claims description 4
- 235000013399 edible fruits Nutrition 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 230000000306 recurrent effect Effects 0.000 claims description 3
- 239000000047 product Substances 0.000 claims 8
- 239000006227 byproduct Substances 0.000 claims 1
- 238000003062 neural network model Methods 0.000 abstract description 3
- 235000004522 Pentaglottis sempervirens Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
Abstract
The invention relates to the technical field of AI, and discloses an agricultural and sideline product traceability system based on AI technology, which comprises: the planting environment feature extraction module is used for extracting planting environment features according to the planting environment data of apples in different planting areas; the image feature generation module is used for generating image features according to the growing images of apples in different planting areas; a mixed feature generation module for combining the planting environment features and the image features to obtain mixed features; a pesticide residue generating module for generating pesticide residue in apple harvesting periods of different planting areas; according to the method, the planting environment factors of apples and the associated influence factors of different planting areas are comprehensively considered, the pesticide residue in the harvest period of the apples is predicted through the neural network model, and the loss of pesticide residue information in the planting stage in the agricultural and sideline product traceability information is compensated.
Description
Technical Field
The invention relates to the technical field of AI, in particular to an agricultural and sideline product traceability system based on AI technology.
Background
The existing agricultural and sideline product traceability system relates to the problems that links are more, information of each link is lost and incomplete, pesticide spraying amount information of agricultural and sideline product traceability information in a planting stage is easy to ignore, and consumer benefits are damaged.
Disclosure of Invention
The invention provides an agricultural and sideline product traceability system based on an AI technology, which solves the technical problem that agricultural and sideline product traceability information is easy to ignore pesticide spraying amount information in a planting stage in the related technology.
The invention provides an agricultural and sideline product traceability system based on an AI technology, which comprises:
the planting environment acquisition module is used for acquiring planting environment data of apples in different planting areas;
the planting environment data of apples in different planting areas comprise soil moisture content, soil phosphorus content, soil nitrogen content and pesticide spraying information; the pesticide spraying information includes: pesticide spraying time, total pesticide spraying amount and pesticide spraying area; the planting environment feature extraction module is used for extracting planting environment features according to the planting environment data of apples in different planting areas; the image acquisition module is used for acquiring growth images of apples in different planting areas; the image feature generation module is used for generating image features according to the growing images of apples in different planting areas; a mixed feature generation module for combining the planting environment features and the image features to obtain mixed features;
the mixing characteristics are expressed as:wherein->An element representing an nth row and an mth column, and representing a mixed feature vector of an mth planting area of an nth pesticide spraying time node;
the hybrid eigenvector is expressed as:wherein->Respectively representing the soil water content, the soil phosphorus content, the soil nitrogen content and the pesticide spraying unit quantity of an mth planting area of an nth pesticide spraying time node; wherein->A local image feature vector representing an mth planting area of an nth pesticide spraying time node;
a pesticide residue generating module for generating pesticide residue in apple harvesting periods of different planting areas; inputting the mixed characteristics into a pesticide residue prediction model, and outputting the pesticide residue in the apple harvesting period of different planting areas;
the pesticide residue prediction model comprises u rows of prediction units, wherein each row of prediction units comprises a first unit and a second unit; the first unit of the ith column inputs an input feature matrix of the ith pesticide spraying time node; the input feature matrix is expressed as:wherein->A hybrid feature vector representing a 1~d th planting area of the i-th time node;
the second unit of the ith column inputs the first output characteristic output by the ith first unit; the output of the second unit of the u-th column prediction unit is connected to d classifiers, and the set of classification labels of the u-th classifier isThe value range of the pesticide residue is 0.01-0.5 mg/kg when the apple fruits in the u-th planting area are harvested, the average value is discretized,corresponds to a discretized one of the pesticide residue values.
Further, collecting soil water content, soil phosphorus content and soil nitrogen content through a sensor, and recording pesticide spraying information in an artificial mode;
further, the planting environment features are expressed as:wherein->An element representing an nth row and an mth column, and a local planting environment feature vector representing an mth planting area of an nth pesticide spraying time node;
the local planting environment feature vector is expressed as:wherein->Respectively representing the soil water content, the soil phosphorus content, the soil nitrogen content and the pesticide spraying unit quantity of an mth planting area of an nth pesticide spraying time node;
the pesticide spraying unit amount is the pesticide spraying total amount divided by the pesticide spraying area, wherein the pesticide spraying total amount unit is expressed in milliliters, and the pesticide spraying area unit is expressed in square area;
further, acquiring growth images of apples of different planting areas corresponding to pesticide spraying time nodes through a high-definition camera;
further, the image features are expressed as:wherein->An element representing an nth row and an mth column, and a local image feature vector representing an mth planting area of an nth pesticide spraying time node;
further, the local image feature vector is obtained through a convolutional neural network model;
further, each first unit comprises h channels,wherein->A fourth intermediate feature representing the h-th channel of the i-th first unit, +.>、/>、/>First, second and third intermediate features of the h channel of the ith first unit, respectively,/->Representing a normalized exponential function;
;/>;/>the method comprises the steps of carrying out a first treatment on the surface of the Wherein->An input feature matrix representing the input of the ith first cell,/th>、/>、/>First, second and third weight matrices representing the h channel of the i first unit, respectively;
,/>the fourth intermediate feature of the 1~H th channel of the ith first unit is spliced to obtain a fifth intermediate feature +.>,/>Representing the output weight matrix of the i first cell.
Further, the column number of the spliced fifth intermediate feature is equal to the dimension of the mixed feature vector;row and column numbers and->The number of rows and columns is the same;
further, the second unit is a recurrent neural network unit;
further, the distributed account book technology is utilized to write pesticide residue data in the apple harvesting period of different planting areas into the blockchain, so that sharing and traceability of the data are realized, and the integrity and safety of agricultural and sideline product traceability information are ensured.
The invention has the beneficial effects that: according to the method, the planting environment factors of apples and the associated influence factors of different planting areas are comprehensively considered, the pesticide residue in the harvest period of the apples is predicted through the neural network model, and the loss of pesticide residue information in the planting stage in the agricultural and sideline product traceability information is compensated.
Drawings
FIG. 1 is a block diagram of an agricultural and sideline product traceability system based on AI technology of the present invention.
Detailed Description
The subject matter described herein will now be discussed with reference to example embodiments. It is to be understood that these embodiments are merely discussed so that those skilled in the art may better understand and implement the subject matter described herein and that changes may be made in the function and arrangement of the elements discussed without departing from the scope of the disclosure herein. Various examples may omit, replace, or add various procedures or components as desired. In addition, features described with respect to some examples may be combined in other examples as well.
As shown in fig. 1, an agricultural and sideline product traceability system based on AI technology includes:
a planting environment collection module 101 for collecting planting environment data of apples in different planting areas;
the planting environment data of apples in different planting areas comprise soil moisture content, soil phosphorus content, soil nitrogen content and pesticide spraying information;
the pesticide spraying information includes: pesticide spraying time, total pesticide spraying amount and pesticide spraying area;
in one embodiment of the invention, the soil water content, the soil phosphorus content and the soil nitrogen content are collected through the sensor, and pesticide spraying information is recorded in an artificial mode;
a planting environment feature extraction module 102 for extracting planting environment features according to the planting environment data of apples in different planting areas;
the planting environment features are expressed as follows:wherein->An element representing an nth row and an mth column, and a local planting environment feature vector representing an mth planting area of an nth pesticide spraying time node;
one pesticide spraying time node corresponds to the time point of one pesticide spraying.
The local planting environment feature vector is expressed as:wherein->Respectively representing the soil water content, the soil phosphorus content, the soil nitrogen content and the pesticide spraying list of an mth planting area of an nth pesticide spraying time node;
the pesticide spraying unit amount is the pesticide spraying total amount divided by the pesticide spraying area, wherein the pesticide spraying total amount unit is expressed in milliliters, and the pesticide spraying area unit is expressed in square area;
an image acquisition module 103 for acquiring growth images of apples in different planting areas;
in one embodiment of the invention, the growth images of apples of different planting areas corresponding to pesticide spraying time nodes are acquired through a high-definition camera; typically a top view or bird's eye view;
an image feature generation module 104 for generating image features from the growing images of apples in different planting areas;
the image features are expressed as:wherein->An element representing an nth row and an mth column, and a local image feature vector representing an mth planting area of an nth pesticide spraying time node;
the local image feature vector is obtained by inputting a growing image of apples into a convolutional neural network model;
training of the convolutional neural network model is a conventional technical means and is not described in detail herein;
a hybrid feature generation module 105 for combining the planting environment feature and the image feature to obtain a hybrid feature;
the mixing characteristics are expressed as:wherein->Shows the elements of the nth row and the mth column, and represents the nth pesticideSpraying the mixed feature vector of the m-th planting area of the time node;
the hybrid eigenvector is expressed as:wherein->Respectively representing the soil water content, the soil phosphorus content, the soil nitrogen content and the pesticide spraying unit quantity of an mth planting area of an nth pesticide spraying time node; wherein->A local image feature vector representing an mth planting area of an nth pesticide spraying time node;
a pesticide residue amount generation module 106 for generating pesticide residue amounts during apple harvesting of different planting areas; inputting the mixed characteristics into a pesticide residue prediction model, and outputting the pesticide residue in the apple harvesting period of different planting areas;
the pesticide residue prediction model comprises u rows of prediction units, wherein each row of prediction units comprises a first unit and a second unit;
the first unit of the ith column inputs an input feature matrix of the ith pesticide spraying time node;
the input feature matrix is expressed as:wherein->A hybrid feature vector representing a 1~d th planting area of the i-th time node;
each first unit comprises h channels,wherein->A fourth intermediate feature representing the h-th channel of the i-th first unit, +.>、/>、/>First, second and third intermediate features of the h channel of the ith first unit, respectively,/->Representing a normalized exponential function;
;/>;/>the method comprises the steps of carrying out a first treatment on the surface of the Wherein->An input feature matrix representing the input of the ith first cell,/th>、/>、/>First, second and third weight matrices representing the h channel of the i first unit, respectively;
,/>the fourth intermediate feature of the 1~H th channel of the ith first unit is spliced to obtain a fifth intermediateCharacteristics->,/>An output weight matrix representing the i first cell; the column number of the spliced fifth intermediate feature is equal to the dimension number of the mixed feature vector; />Number of rows and columns andthe number of rows and columns is the same.
The second unit of the ith column inputs the first output characteristic output by the ith first unit;
in one embodiment of the invention, the second cells are recurrent neural network cells, the second cells of each column being connected in series.
The output of the second unit of the u-th column prediction unit is connected to d classifiers, and the set of classification labels of the u-th classifier isThe value range of the pesticide residue is 0.01 to 0.5mg/kg when the apple fruits in the ith planting area are harvested, and the average value is discretized, < + >>Corresponding to a discretized pesticide residue value;
further, the distributed account book technology is utilized to write pesticide residue data in the harvest period of apples in different planting areas into the blockchain, so that sharing and traceability of the data are realized, and the integrity and safety of agricultural and sideline product traceability information are ensured;
according to the method, the planting environment factors of apples and the associated influence factors of different planting areas are comprehensively considered, the pesticide residue in the harvest period of the apples is predicted through the neural network model, and the loss of pesticide residue information in the planting stage in the agricultural and sideline product traceability information is compensated.
The embodiment has been described above with reference to the embodiment, but the embodiment is not limited to the above-described specific implementation, which is only illustrative and not restrictive, and many forms can be made by those of ordinary skill in the art, given the benefit of this disclosure, are within the scope of this embodiment.
Claims (9)
1. An agricultural and sideline product traceability system based on an AI technology is characterized by comprising:
the planting environment acquisition module is used for acquiring planting environment data of apples in different planting areas;
the planting environment data of apples in different planting areas comprise soil moisture content, soil phosphorus content, soil nitrogen content and pesticide spraying information; the pesticide spraying information includes: pesticide spraying time, total pesticide spraying amount and pesticide spraying area;
the planting environment feature extraction module is used for extracting planting environment features according to the planting environment data of apples in different planting areas;
the image acquisition module is used for acquiring growth images of apples in different planting areas;
the image feature generation module is used for generating image features according to the growing images of apples in different planting areas;
a mixed feature generation module for combining the planting environment features and the image features to obtain mixed features;
the mixing characteristics are expressed as:wherein->An element representing an nth row and an mth column, and representing a mixed feature vector of an mth planting area of an nth pesticide spraying time node;
the hybrid eigenvector is expressed as:wherein->Respectively representing the soil water content, the soil phosphorus content, the soil nitrogen content and the pesticide spraying unit quantity of an mth planting area of an nth pesticide spraying time node; wherein the method comprises the steps ofA local image feature vector representing an mth planting area of an nth pesticide spraying time node;
a pesticide residue generating module for generating pesticide residue in apple harvesting periods of different planting areas; inputting the mixed characteristics into a pesticide residue prediction model, and outputting the pesticide residue in the apple harvesting period of different planting areas;
the pesticide residue prediction model comprises u rows of prediction units, wherein each row of prediction units comprises a first unit and a second unit; the first unit of the ith column inputs an input feature matrix of the ith pesticide spraying time node; the input feature matrix is expressed as:wherein->A hybrid feature vector representing a 1~d th planting area of the i-th time node;
the second unit of the ith column inputs the first output characteristic output by the ith first unit; the output of the second unit of the u-th column prediction unit is connected to d classifiers, and the set of classification labels of the u-th classifier isThe value range of the pesticide residue in the harvest of apple fruits in the ith planting area is subjected to mean discretization, and the apple fruits are subjected to->Corresponding to a discretized pesticide residue value;
and writing pesticide residue data in the harvest time of apples in different planting areas into the blockchain by using a distributed ledger wall technology.
2. The agricultural and sideline product traceability system based on the AI technology as claimed in claim 1, wherein the sensor is used for collecting soil water content, soil phosphorus content and soil nitrogen content, and pesticide spraying information is recorded in an artificial mode.
3. The agricultural and sideline product traceability system based on the AI technology of claim 1, wherein the planting environment features are expressed as:wherein->An element representing an nth row and an mth column, and a local planting environment feature vector representing an mth planting area of an nth pesticide spraying time node;
the local planting environment feature vector is expressed as:wherein->Respectively representing the soil water content, the soil phosphorus content, the soil nitrogen content and the pesticide spraying unit quantity of an mth planting area of an nth pesticide spraying time node;
the pesticide spraying unit amount is the pesticide spraying total amount divided by the pesticide spraying area, wherein the pesticide spraying total amount unit is expressed in milliliters and the pesticide spraying area unit is expressed in square area.
4. The AI technology-based agricultural and sideline product traceability system according to claim 1, wherein the high-definition camera is used for collecting the growth images of apples in different planting areas corresponding to pesticide spraying time nodes.
5. The agricultural and sideline product traceability system based on the AI technology of claim 1, wherein the image features are expressed as:wherein->Elements representing the nth row and the mth column, and local image feature vectors representing the mth planting area of the nth pesticide spraying time node.
6. The AI-technology-based agricultural and sideline product traceability system of claim 5, wherein the local image feature vectors are obtained by a convolutional neural network model.
7. The agricultural byproduct traceability system based on AI technology as claimed in claim 1, wherein each first unit includes h channels,wherein->A fourth intermediate feature representing the h-th channel of the i-th first unit, +.>、/>、/>First, second and third intermediate features of the h channel of the ith first unit, respectively,/->Representing a normalized exponential function;
;/>;/>the method comprises the steps of carrying out a first treatment on the surface of the Wherein->An input feature matrix representing the input of the ith first cell,/th>、/>、/>First, second and third weight matrices representing the h channel of the i first unit, respectively;
,/>the fourth intermediate feature of the 1~H th channel of the ith first unit is spliced to obtain a fifth intermediate feature +.>,/>Representing the output weight matrix of the i first cell.
8. The AI-technology-based agricultural and sideline product traceability system of claim 7, wherein the number of columns of the spliced fifth intermediate feature is equal to the dimension of the mixed feature vector;row and column numbers and->The number of rows and columns is the same.
9. The AI-technology-based agricultural and sideline product traceability system of claim 7, wherein the second unit is a recurrent neural network unit.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311065871.7A CN116777484B (en) | 2023-08-23 | 2023-08-23 | Agricultural and sideline product traceability system based on AI technology |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311065871.7A CN116777484B (en) | 2023-08-23 | 2023-08-23 | Agricultural and sideline product traceability system based on AI technology |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116777484A CN116777484A (en) | 2023-09-19 |
CN116777484B true CN116777484B (en) | 2023-11-21 |
Family
ID=87989851
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311065871.7A Active CN116777484B (en) | 2023-08-23 | 2023-08-23 | Agricultural and sideline product traceability system based on AI technology |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116777484B (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107506790A (en) * | 2017-08-07 | 2017-12-22 | 西京学院 | Greenhouse winter jujube plant disease prevention model based on agriculture Internet of Things and depth belief network |
CN112116206A (en) * | 2020-08-21 | 2020-12-22 | 淮北市盛世昊明科技服务有限公司 | Intelligent agricultural system based on big data |
CN113298537A (en) * | 2021-04-30 | 2021-08-24 | 华中农业大学 | Rice full-chain quality information intelligent detection system and method based on Internet of things |
CN113297925A (en) * | 2021-04-30 | 2021-08-24 | 华中农业大学 | Intelligent early warning method and system for quality of full chain of fruits and vegetables |
FR3120707A1 (en) * | 2021-03-10 | 2022-09-16 | Saur | METHOD FOR ESTIMATING THE CONTENT OF PESTICIDES OR METABOLITES OF PESTICIDES IN WATER TO BE DRINKABLE |
-
2023
- 2023-08-23 CN CN202311065871.7A patent/CN116777484B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107506790A (en) * | 2017-08-07 | 2017-12-22 | 西京学院 | Greenhouse winter jujube plant disease prevention model based on agriculture Internet of Things and depth belief network |
CN112116206A (en) * | 2020-08-21 | 2020-12-22 | 淮北市盛世昊明科技服务有限公司 | Intelligent agricultural system based on big data |
FR3120707A1 (en) * | 2021-03-10 | 2022-09-16 | Saur | METHOD FOR ESTIMATING THE CONTENT OF PESTICIDES OR METABOLITES OF PESTICIDES IN WATER TO BE DRINKABLE |
CN113298537A (en) * | 2021-04-30 | 2021-08-24 | 华中农业大学 | Rice full-chain quality information intelligent detection system and method based on Internet of things |
CN113297925A (en) * | 2021-04-30 | 2021-08-24 | 华中农业大学 | Intelligent early warning method and system for quality of full chain of fruits and vegetables |
Also Published As
Publication number | Publication date |
---|---|
CN116777484A (en) | 2023-09-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Eberle et al. | Using pennycress, camelina, and canola cash cover crops to provision pollinators | |
Cox et al. | Row spacing, hybrid, and plant density effects on corn silage yield and quality | |
CN111507319A (en) | Crop disease identification method based on deep fusion convolution network model | |
CN113705937B (en) | Farmland yield estimation method combining machine vision and crop model | |
Hacker et al. | Trophic consequences of a positive plant interaction | |
Ofori et al. | Towards deep learning for weed detection: Deep convolutional neural network architectures for plant seedling classification | |
CN113011221A (en) | Crop distribution information acquisition method and device and measurement system | |
CN116777484B (en) | Agricultural and sideline product traceability system based on AI technology | |
Peters et al. | Harnessing AI to transform agriculture and inform agricultural research | |
Tan et al. | Machine learning approaches for rice seedling growth stages detection | |
Ren et al. | Multi-resolution outlier pooling for sorghum classification | |
Wu et al. | Meta‐learning shows great potential in plant disease recognition under few available samples | |
Ahil et al. | Apple and grape leaf disease classification using mlp and cnn | |
Mishra et al. | Prediction and detection of nutrition deficiency using machine learning | |
Geber et al. | Sex allocation in hermaphrodites with partial overlap in male/female resource inputs | |
Wei et al. | Wheat biomass, yield, and straw-grain ratio estimation from multi-temporal UAV-based RGB and multispectral images | |
Sarkar et al. | A UAV and Deep Transfer Learning Based Environmental Monitoring: Application to Native and Invasive Species classification in Southern regions of the USA | |
Figueiredo et al. | Linking high diversification rates of rapidly growing Amazonian plants to geophysical landscape transformations promoted by Andean uplift | |
CN114549962A (en) | Garden plant leaf disease classification method | |
Demetrio et al. | Body size and clonality consequences for sexual reproduction in a perennial herb of Brazilian rupestrian grasslands | |
Ayikpa et al. | CocoaMFDB: A dataset of cocoa pod maturity and families in an uncontrolled environment in Côte d'Ivoire | |
Zhou et al. | Yield estimation of soybean breeding lines using UAV multispectral imagery and convolutional neuron network | |
CN116757867B (en) | Digital village construction method and system based on multi-source data fusion | |
HARAYAMA | Mobilizing science, technology, and innovation to transform Japanese agriculture | |
Bai et al. | An efficient approach to detect and track winter flush growth of litchi tree based on UAV remote sensing and semantic segmentation |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |