Intelligent agricultural unmanned aerial vehicle inspection and data analysis system

Abstract

The invention discloses an intelligent agricultural unmanned aerial vehicle inspection and data analysis system, which belongs to the field of unmanned aerial vehicle inspection and data analysis systems, and can simultaneously pop up corresponding drug types and dosage information for farmers according to the identified diseased plant types and pest types in a response measure link after pest and disease damage identification. According to the method and the system, a corresponding pesticide sales merchant platform link is established, so that a farmer can perform prevention and control operation in time, a mode of automatically matching a sales merchant link and popping up the sales merchant link to a hand-held terminal by utilizing a block chain technology is utilized, transaction security and traceability are ensured, an intelligent contract is introduced to realize a purchase automation process, transaction efficiency and coherence in a response measure link are improved, and the method and the system are convenient for farmers to perform prevention and control operation. The requirements of modern agriculture on efficient and accurate pest control are met.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle inspection and data analysis systems, and more specifically, to a smart agricultural unmanned aerial vehicle inspection and data analysis system. Background Art

[0002] In the agricultural field, pests and diseases have always been an important factor affecting crop yield and quality. With the development of science and technology, drone inspection technology has gradually been applied to the agricultural field, providing a new means for pest and disease monitoring.

[0003] Drones have many advantages in agricultural inspections. They can quickly cover large areas of farmland, and the various devices they carry can obtain rich data information. For example, through image acquisition devices such as high-definition cameras and infrared spectrometer cameras, high-definition picture data information and high-definition infrared spectrometer image data information of farmland can be collected. Combined with the corresponding sensor modules, the flight status of the drone itself can be monitored, thereby ensuring the stability and accuracy of data collection. These data are of great significance for analyzing the pest and disease situation of farmland.

[0004] However, the current drone inspection technology still has some shortcomings in its application. For example, in terms of data processing, although a large amount of image data can be obtained, the processing and analysis technology of image data needs to be further optimized to improve the accuracy of pest and disease identification; more importantly, in the response measures after pest and disease identification, the existing drone inspection system cannot directly pop up the corresponding drug type and dosage information for farmers based on the identified diseased plant type and pest type, as well as the corresponding pesticide sales merchant platform link, so that farmers can carry out prevention and control operations in a timely manner. This incoherence makes the entire pest and disease control process inefficient, unable to give full play to the advantages of drone inspection technology in agricultural production, and difficult to meet the needs of modern agriculture for efficient and accurate pest and disease control. Therefore, it is particularly necessary to develop a smart agricultural drone inspection and data analysis system that can solve the above problems. Summary of the invention

[0005] 1. Technical issues to be solved In view of the problems existing in the prior art, the purpose of the present invention is to provide a smart agricultural drone inspection and data analysis system. In the system, first, in terms of image recognition, the diseased plant recognition unit uses DeepLabV3+ segmentation model technology to remove the background, grayscale co-occurrence matrix extraction technology, HSV color space analysis disease spot technology, and spectral index analysis NDVI, PRI detection technology to detect texture, color, and spectral anomalies, and uses the pre-trained model ResNet50 or EfficientNet-B0 technology and MMF multimodal fusion framework technology to match and determine the type of diseased plants, so that the matching and identification of diseased plants is more accurate. Then, the pest recognition unit uses image pyramid technology to enhance the resolution, HOG directional gradient histogram technology and combines SVM or deep learning model to extract appearance features, and simultaneously uses the dynamically changing hot spots and infrared spectral reflectivity technology in the infrared heat map to extract the differences in the appearance of pests. Finally, the pre-trained MobileNet or Inception-V3 technology is used to classify the types of pests. The combination of multiple technical means greatly improves the accuracy of identifying small pests and reduces the probability of misjudgment. The intelligent analysis module uses convolutional neural network (CNN) technology and image synthesis algorithm technology to match the database of the agricultural supply platform to determine the required drug information and dosage, introduces knowledge graph technology to build a relationship graph, and provides accurate recommendations and suggestions. According to the drug information and dosage required for diseased plant types and pest types, blockchain technology is used to automatically match sales merchant links and pop up on the handheld terminal to ensure transaction security and traceability. Smart contracts are introduced to realize automated procurement processes, improve transaction efficiency, and meet the needs of modern agriculture for efficient and accurate pest and disease control.

[0006] 2. Technical solution

[0007] To solve the above problems, the present invention adopts the following technical solutions.

[0008] A smart agricultural drone inspection and data analysis system includes a drone and a handheld terminal device, wherein the drone is equipped with a sensor module, an image acquisition module, a central processing module and a communication module 1, and the handheld terminal device is equipped with a communication module 2, a flight control module, a data processing module and an image analysis module, and the handheld terminal device is also equipped with a report generation module and an intelligent analysis module, and the communication module 1 is remotely connected to the communication module 2 by data communication, and the communication module 2 is also remotely connected to an agricultural supply platform by data communication; The agricultural material supply platform stores a database of diseased plant types and pest species in advance through the relational database MySQL, and the database also records and stores drug information and drug usage for different diseased plant types, and drug information and drug usage for different pest species. The agricultural material supply platform is also connected to multiple pesticide sales merchant platforms through a standardized data communication interface. The sensor module is data-connected with the motor sensor and attitude sensor mounted on the drone, and is used to monitor the motor speed and attitude of the drone in real time. The image acquisition module is data-connected with the high-definition camera and infrared spectrum camera mounted on the drone, and is used to collect high-definition picture data information and high-definition infrared spectrum image data information collected by the drone in real time. The central processing module processes the data information collected by the sensor module and the image acquisition module and uploads them to the data processing module on the handheld terminal device through the communication module one and the communication module two. The flight control module is data-communicated with the central processing module through the communication module two and the communication module one, so as to conveniently control the flight attitude of the drone.

[0009] Furthermore, the flight control module is internally provided with a navigation path planning unit and a flight parameter setting unit. The navigation path planning unit plans and locates the flight route of the UAV based on the geographic information system GIS and the global positioning system GPS, and the flight parameter setting unit actively controls the flight parameters of the UAV based on the display interface displayed on the handheld terminal device.

[0010] Furthermore, the data processing module first uses the distributed computing technology Hadoop to denoise the image data and improve the contrast of the high-definition image data information and high-definition infrared spectrum image data information received by the communication module 2, and then performs spatial alignment to ensure that the corresponding pixels in the two images represent the same geographic image location. Then, the weighted average method of the image fusion algorithm is used to perform weighted summation on each pixel of the high-definition image and the infrared spectrum image to obtain a multimodal fused image.

[0011] Furthermore, a diseased plant identification unit and a pest identification unit are provided inside the image analysis module; The diseased plant identification unit first uses the DeepLabV3+ segmentation model technology based on deep learning to segment the multimodal fused image and remove the background, and then uses the GLCM grayscale co-occurrence matrix extraction technology, HSV color space analysis disease spot technology and spectral index analysis NDVI, PRI detection technology to detect the texture, color and spectral anomalies of the multimodal fused image, and then uses the pre-trained model ResNet50 or EfficientNet-B0 technology and MMF multimodal fusion framework technology to match the texture, color and spectral anomalies of the detected image, and matches the texture, color and spectral anomalies of a certain type of diseased plant that has been pre-trained and stored, so as to determine the corresponding diseased plant type; The pest recognition unit first uses image pyramid technology to enhance the resolution of pests, then uses HOG directional gradient histogram technology combined with SVM or deep learning model to extract pest appearance features, and simultaneously uses dynamically changing hot spots in infrared heat maps and infrared spectral reflectivity technology to extract differences in pest appearance. Finally, the pre-trained MobileNet or Inception-V3 technology is used to classify the types of pests, thereby determining the types of pests on the diseased plants.

[0012] Furthermore, the report generation module is based on the diseased plant type results and pest species results analyzed by the image analysis module, and uses natural language processing technology NLP to convert data such as diseased plant type, pest species, location information, etc. into natural language descriptions, so as to generate report content that is easier for farmers to understand. At the same time, data visualization technology (such as bar charts and heat maps) is used to display the area information and density information of diseased plants and pests in an intuitive graphical manner in the report, so that farmers can understand the distribution of farmland diseases and pests more quickly. Finally, the generated report is displayed on the display interface of the handheld terminal device.

[0013] Furthermore, the intelligent analysis module matches the diseased plant types, pest species, location information, area information and density information generated in the report with the relational database within the agricultural supplies supply platform based on convolutional neural network (CNN) technology and image synthesis algorithm technology, thereby determining the drug information and usage required for the corresponding diseased plant types and pest species. At the same time, the intelligent analysis module introduces knowledge graph technology to construct the drug information, diseased plant types, pest species and the relationships between them in the agricultural supplies supply platform into a knowledge graph, and through intelligent reasoning and query functions, more accurate drug recommendations and usage suggestions are provided to farmers.

[0014] Furthermore, the agricultural supplies supply platform uses blockchain technology to automatically match the corresponding sales merchant link according to the diseased plant type, pest type, required drug information and usage, and pops it up on the display interface of the handheld terminal device, so that farmers can place orders to buy drugs for timely prevention and control with one click. Blockchain technology is used to ensure the security and traceability of transactions. Farmers' purchase records, drug information and other data are stored on the blockchain and cannot be tampered with, ensuring the fairness and transparency of transactions. At the same time, smart contract technology is introduced to realize the automated process of pesticide procurement. For example, when farmers place an order, the smart contract automatically executes order processing, payment settlement and other operations to improve transaction efficiency.

[0015] 3. Beneficial effects

[0016] Compared with the prior art, the advantages of the present invention are: (1) In this scheme, the system first uses DeepLabV3+ segmentation model technology to remove the background in image recognition. It also uses grayscale co-occurrence matrix extraction technology, HSV color space analysis disease spot technology, and spectral index analysis NDVI and PRI detection technology to detect texture, color, and spectral anomalies. It uses the pre-trained model ResNet50 or EfficientNet-B0 technology and MMF multimodal fusion framework technology to match and determine the type of diseased plants, making the matching and identification of diseased plants more accurate. Then, the pest recognition unit uses image pyramid technology to enhance the resolution, HOG directional gradient histogram technology combined with SVM or deep learning model to extract shape features, and uses the dynamically changing hot spots in the infrared heat map and infrared spectral reflectance technology to extract the differences in the appearance of pests. Finally, it uses the pre-trained MobileNet or Inception-V3 technology to classify the types of pests. The combination of multiple technical means greatly improves the accuracy of identifying small pests and reduces the probability of misjudgment. (2) In this solution, the intelligent analysis module uses convolutional neural network (CNN) technology and image synthesis algorithm technology to match the agricultural supply platform database to determine the required drug information and dosage, introduces knowledge graph technology to build a relationship graph, and provides accurate recommendations and suggestions based on the drug information and dosage required by the diseased plant type and pest type; (3) In this solution, in the response phase after pest and disease identification, the system can directly pop up the corresponding drug type and dosage information, as well as the corresponding pesticide sales merchant platform link for farmers based on the identified diseased plant type and pest species, so that farmers can take timely prevention and control operations. The system also uses blockchain technology to automatically match the sales merchant link and pop it up on the handheld terminal to ensure transaction security and traceability. It introduces smart contracts to realize the automated procurement process, improve transaction efficiency and the consistency of the response phase, and meet the needs of modern agriculture for efficient and accurate pest and disease control. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a schematic diagram of the UAV of the present invention in operation; Figure 2 It is a schematic diagram of the overall system architecture structure of the present invention.

[0018] Description of the numbers in the figure: 1. UAV; 2. Sensor module; 3. Image acquisition module; 4. Central processing module; 5. Communication module one; 6. Handheld terminal device; 7. Communication module two; 8. Flight control module; 801. Navigation path planning unit; 802. Flight parameter setting unit; 9. Data processing module; 10. Image analysis module; 1001. Diseased plant identification unit; 1002. Pest identification unit; 11. Report generation module; 12. Intelligent analysis module; 13. Agricultural materials supply platform. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the specification of the present invention; it is obvious that the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments, and all other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making creative work are within the scope of protection of the present invention.

[0020] Example 1

[0021] See also Figure 1-Figure 2 ,The working principle of the smart agricultural drone inspection and data analysis system is: First, the flight route of the drone 1 is planned on the handheld terminal device 6 through the navigation path planning unit 801 in the flight control module 8 based on the geographic information system GIS and the global positioning system GPS. For example, for a rectangular farmland, according to the farmland boundary coordinates and the preset inspection interval, the drone is set to take off from a corner of the farmland and fly parallel to the long side in sequence to cover the flight path of the entire farmland area. Then, the flight parameter setting unit 802 is used to set the flight parameters such as the flight altitude of 10 meters (to ensure that the farmland crops can be clearly photographed without affecting the growth of the crops), the flight speed of 5 meters / second, and the shooting interval of 2 seconds according to the performance of the drone 1 and the actual inspection requirements.

[0022] Example 2

[0023] In view of the above embodiment 1, further description is given: The drone 1 is started. During the flight, the sensor module 2 monitors the motor speed and attitude of the drone 1 in real time. For example, when the drone 1 encounters breeze interference and its attitude is slightly tilted, the motor sensor and attitude sensor quickly transmit data to the sensor module 2. The sensor module (2) processes the data and transmits it to the central processing module 4. The flight parameter setting unit 802 then resets the parameters to control the flight attitude of the drone 1. During this period, the image acquisition module 3 collects high-definition picture data information and high-definition infrared spectrum image data information of the farmland in real time through the high-definition camera and the infrared spectrum camera. For example, when flying over a certain area, the high-definition camera captures images of suspected disease spots on the leaves of the crops, and the infrared spectrum camera captures the infrared spectrum reflection characteristic images of the crops in the area.

[0024] Example 3

[0025] In view of the above-mentioned Embodiment 1 and Embodiment 2, further description is given: The central processing module 4 performs preliminary processing on the data information collected by the image acquisition module 3, and then uploads it to the data processing module 9 on the handheld terminal device 6 through the communication module 1 5 and the communication module 2 7. The data processing module 9 first uses the distributed computing technology Hadoop to denoise and improve the contrast of the received high-definition image data information and high-definition infrared spectrum image data information. For example, it removes the image noise caused by uneven light and enhances the contrast between the diseased spots and the normal leaf parts. Then it performs spatial registration to ensure that the corresponding pixels in the two images represent the same geographic image location. Then, it uses the weighted average method of the image fusion algorithm to perform weighted summation on each pixel of the high-definition image and the infrared spectrum image to obtain a multimodal fused image.

[0026] Example 4

[0027] In view of the above-mentioned Embodiment 1, Embodiment 2 and Embodiment 3, further description is given: The diseased plant identification unit 1001 in the image analysis module 10 starts working. First, the DeepLabV3 + segmentation model technology based on deep learning is used to segment the multimodal fused image, remove the background, and highlight the diseased plant part. Then, the GLCM grayscale co-occurrence matrix extraction technology, HSV color space analysis of diseased spots technology, and spectral index analysis NDVI and PRI detection technology are used to detect the texture, color, and spectral anomalies of the image. For example, the roughness, directionality and other characteristics of the texture of the diseased spot area are analyzed by the GLCM grayscale co-occurrence matrix, the difference between the diseased spot color and the normal leaf color is analyzed by the HSV color space, and the abnormality of the spectral reflectance of the diseased plant area is detected by spectral index analysis. Finally, the pre-trained model ResNet50 or EfficientNet - B0 technology and MMF multimodal fusion framework technology are used to match the detected image features. Assume that the texture of a crop leaf is rough, the color is yellow, and the spectral index is abnormal, and it is successfully matched with the characteristics of a certain disease type stored in the pre-training, so that the crop is determined to be the corresponding diseased plant type.

[0028] The pest identification unit 1002 also processes the image. First, the resolution of the pest image is enhanced by using the image pyramid technology to make the pest details clearer. Then, the HOG directional gradient histogram technology is used in combination with SVM or deep learning models to extract the pest appearance features, such as the shape and outline of the pest. At the same time, the dynamic hot spots in the infrared heat map and the infrared spectral reflectivity technology are used to extract the differences in the pest appearance. For example, the activity location and aggregation degree of the pest are determined by the hot spots, and the spectral reflectivity differences are used to distinguish different types of pests. Finally, the pre-trained MobileNet or Inception-V3 technology is used to classify the types of pests. For example, an insect with a specific shape in the image, a specific hot spot distribution in the infrared heat map, and a spectral reflectivity that meets the characteristics of a certain type of pest is identified as a certain type of pest.

[0029] Example 5

[0030] In view of the above-mentioned Embodiment 1, Embodiment 2, Embodiment 3 and Embodiment 4, further description is given: The report generation module 11 is based on the results of the diseased plant type and the pest species analyzed by the image analysis module 10, and uses the natural language processing technology NLP to convert the data such as the diseased plant type, pest species, and location information into natural language descriptions. For example, a description such as "5 crops infected with powdery mildew were found in the northeast corner of the farmland, and 10 aphids were found to be active in the area" is generated. At the same time, data visualization technology (such as bar graphs and heat maps) is used to display the area information and density information of diseased plants and pests in the report in an intuitive graphical manner. For example, a heat map is used to display the distribution density of diseased plants in the farmland. The darker the color, the higher the density of diseased plants, so that farmers can more quickly understand the distribution of farmland pests and diseases. The final generated report is displayed on the display interface of the handheld terminal device 6.

[0031] Example 6

[0032] In view of the above-mentioned Embodiment 1, Embodiment 2, Embodiment 3, Embodiment 4 and Embodiment 5, further description is given: The intelligent analysis module 12 matches the diseased plant type, pest species, location information, area information and density information in the report with the relational database inside the agricultural supply platform 13 based on the convolutional neural network CNN technology and image synthesis algorithm technology. For example, it is determined that crops infected with powdery mildew need to be controlled with carbendazim, and the required drug usage is calculated to be 200 grams per mu based on the diseased plant area and density. At the same time, the intelligent analysis module 12 introduces knowledge graph technology to construct the drug information, diseased plant type, pest species and the relationship between them in the agricultural supply platform 13 into a knowledge graph. Through intelligent reasoning and query functions, it can provide farmers with more accurate drug recommendations and usage suggestions, such as recommending carbendazim brands, manufacturers and precautions when using it; The agricultural material supply platform 13 automatically matches the corresponding sales merchant link based on the diseased plant type, the drug information and the usage required by the pest type using blockchain technology and pops up on the display interface of the handheld terminal device 6. For example, if a nearby pesticide sales merchant with good reputation and reasonable price is matched, the merchant's carbendazim product link will pop up on the handheld terminal device 6, which is convenient for farmers to place orders to buy drugs for timely prevention and control. Blockchain technology is used to ensure the security and traceability of transactions. Farmers' purchase records, drug information and other data are stored on the blockchain and cannot be tampered with, ensuring the fairness and transparency of transactions. At the same time, smart contract technology is introduced to realize the automated process of pesticide procurement. When farmers place an order, the smart contract automatically executes order processing, payment settlement and other operations to improve transaction efficiency.

[0033] The above is only a preferred specific implementation of the present invention; however, the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solution and its improved conception within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A smart agricultural drone inspection and data analysis system, comprising a drone (1) and a handheld terminal device (6), characterized in that: The drone (1) is equipped with a sensor module (2), an image acquisition module (3), a central processing module (4) and a communication module 1 (5); the handheld terminal device (6) is equipped with a communication module 2 (7), a flight control module (8), a data processing module (9) and an image analysis module (10); the handheld terminal device (6) is also equipped with a report generation module (11) and an intelligent analysis module (12); the communication module 1 (5) and the communication module 2 (7) are remotely connected by data communication, and the communication module 2 (7) is also remotely connected by data communication to an agricultural supply platform (13); The agricultural material supply platform (13) stores a database of diseased plant types and pest species in advance via a relational database MySQL, and the database also records and stores drug information and drug usage amounts for different diseased plant types, and drug information and drug usage amounts for different pest species. The agricultural material supply platform (13) is also connected to a plurality of pesticide sales merchant platforms via a standardized data communication interface. The sensor module (2) is data-connected with the motor sensor and the attitude sensor mounted on the drone (1) for real-time monitoring of the motor speed and attitude of the drone (1). The image acquisition module (3) is data-connected with the high-definition camera and the infrared spectrum camera mounted on the drone (1) for real-time acquisition of high-definition picture data information and high-definition infrared spectrum image data information collected by the drone (1). The central processing module (4) processes the data information collected by the sensor module (2) and the image acquisition module (3) and uploads the processed data to the data processing module (9) on the handheld terminal device (6) through the communication module 1 (5) and the communication module 2 (7). The flight control module (8) is data-connected with the central processing module (4) through the communication module 2 (7) and the communication module 1 (5) for convenient control of the flight attitude of the drone (1).

2. The smart agricultural drone inspection and data analysis system according to claim 1 is characterized by: The flight control module (8) is internally provided with a navigation path planning unit (801) and a flight parameter setting unit (802); the navigation path planning unit (801) plans and locates the flight route of the UAV (1) based on a geographic information system (GIS) and a global positioning system (GPS); and the flight parameter setting unit (802) actively controls the flight parameters of the UAV (1) based on a display interface displayed on the handheld terminal device (6).

3. The smart agricultural drone inspection and data analysis system according to claim 1 is characterized by: The data processing module (9) first uses the distributed computing technology Hadoop to perform denoising and contrast enhancement on the high-definition image data information and high-definition infrared spectrum image data information received by the communication module 2 (7), and then performs spatial registration to ensure that the corresponding pixels in the two images represent the same geographical image position. Then, using the weighted average method of the image fusion algorithm, the high-definition image and the infrared spectrum image are weightedly summed at each pixel to obtain a multi-modal fused image.

4. The smart agricultural drone inspection and data analysis system according to claim 1 is characterized by: The image analysis module (10) is internally provided with a diseased plant identification unit (1001) and a pest identification unit (1002); The diseased plant identification unit (1001) first uses the DeepLabV3+ segmentation model technology based on deep learning to segment the multimodal fused image and remove the background, and then uses the GLCM grayscale co-occurrence matrix extraction technology, the HSV color space analysis disease spot technology and the spectral index analysis NDVI and PRI detection technology to detect the texture, color and spectral anomalies of the multimodal fused image, and then uses the pre-trained model ResNet50 or EfficientNet-B0 technology and the MMF multimodal fusion framework technology to match the texture, color and spectral anomalies of the detected image, and matches the texture, color and spectral anomalies of a certain type of diseased plant that has been pre-trained and stored, so as to determine the corresponding diseased plant type; The pest identification unit (1002) first uses image pyramid technology to enhance the resolution of pests, then uses HOG directional gradient histogram technology combined with SVM or deep learning model to extract pest appearance features, and simultaneously uses dynamically changing hot spots in infrared heat maps and infrared spectral reflectivity technology to extract differences in pest appearance, and finally uses pre-trained MobileNet or Inception-V3 technology to classify the types of pests, thereby determining the types of pests on the diseased plants.

5. The smart agricultural drone inspection and data analysis system according to claim 4 is characterized by: The report generation module (11) is based on the diseased plant type results and pest species results analyzed by the image analysis module (10), and uses natural language processing technology (NLP) to convert data such as diseased plant type, pest species, location information, etc. into natural language descriptions, thereby generating report content that is easier for farmers to understand. At the same time, data visualization technology, such as bar graphs and heat maps, is used to display the area information and density information of diseased plants and pests in the report in an intuitive graphical manner, so that farmers can more quickly understand the distribution of farmland diseases and pests. The finally generated report is displayed on the display interface of the handheld terminal device (6).

6. The smart agricultural drone inspection and data analysis system according to claim 1 is characterized by: The intelligent analysis module (12) matches the diseased plant type, pest species, location information, area information and density information generated in the report with the relational database within the agricultural supply platform (13) based on convolutional neural network (CNN) technology and image synthesis algorithm technology, thereby determining the drug information and usage required for the corresponding diseased plant type and pest species. At the same time, the intelligent analysis module (12) introduces knowledge graph technology to construct the drug information, diseased plant type, pest species and the relationship between them in the agricultural supply platform (13) into a knowledge graph, and through intelligent reasoning and query functions, more accurately provide farmers with drug recommendations and usage suggestions.

7. The smart agricultural drone inspection and data analysis system according to claim 6 is characterized by: The agricultural material supply platform (13) uses blockchain technology to automatically match the corresponding sales merchant link according to the diseased plant type, the pest type, the drug information required, and the usage amount, and pops up the link on the display interface of the handheld terminal device (6), so that farmers can place orders to buy drugs for timely prevention and control with one click. Blockchain technology is used to ensure the security and traceability of transactions. Farmers' purchase records, drug information and other data are stored on the blockchain and cannot be tampered with, ensuring the fairness and transparency of transactions. At the same time, smart contract technology is introduced to realize the automated process of pesticide procurement. For example, when farmers place an order, the smart contract automatically executes order processing, payment settlement and other operations to improve transaction efficiency.

Images