Wheat disease and pest recognition and environment regulation system and method based on improved YOLOv10 model

Abstract

The invention discloses a wheat disease and pest recognition and environment regulation system and method based on an improved YOLOv10 model, and relates to the technical field of disease and pest recognition, the system comprises a data acquisition module, an environment regulation device, a cloud server, a central processing unit and a client device; the data acquisition module acquires a plurality of original images and environmental parameters of the wheat breeding greenhouse; the central processing unit integrates the plurality of original images to obtain an integrated image, and the integrated image is uploaded to the client device through the cloud server; the client device inputs the integrated image into a wheat disease and insect pest recognition model to obtain a wheat disease and insect pest recognition result; the wheat disease and pest identification result comprises the category and position coordinates of wheat diseases and pests; the central processing unit further controls the environment adjusting device to adjust the environment parameters according to the environment parameters and the preset data range. According to the invention, the pest and disease identification efficiency, accuracy and real-time performance are improved, and greenhouse environment regulation and control are realized.

Description

Technical Field

[0001] The present application relates to the technical field of pest and disease identification, and in particular to a wheat pest and disease identification and environmental regulation system and method based on an improved YOLOv10 model. Background Art

[0002] During the wheat breeding process, monitoring and prevention of wheat pests and diseases is a key link in ensuring food security and increasing crop yields. However, traditional pest and disease detection methods mainly rely on manual inspections, which have the following problems: (1) Low identification efficiency: Farmers or agricultural technicians need to observe crop leaves and ears with the naked eye to determine the type and severity of pests and diseases, which is time-consuming and labor-intensive, and cannot achieve large-scale rapid detection. (2) Insufficient accuracy: Manual detection is easily affected by factors such as ambient light, crop growth status, and personnel experience level, resulting in a high number of misjudgments and missed judgments. (3) Lack of real-time performance: Manual detection is usually carried out periodically and cannot achieve 24-hour uninterrupted monitoring, making it difficult to identify pests and diseases early. Once the best prevention and control time is missed, crop yield will be affected. (4) Lack of intelligent prevention and control measures: Traditional greenhouse or farmland environmental control mostly relies on manual operation, making it difficult to achieve a linkage response between pest and disease identification results and greenhouse control equipment. There is a lack of a mechanism to coordinate pest and disease identification with environmental regulation and prevention and control measures (such as spraying of pesticides), resulting in delayed response measures and insufficient accuracy. Summary of the Invention

[0003] The purpose of this application is to provide a wheat disease and pest identification and environmental regulation system and method based on an improved YOLOv10 model to solve the problems of low efficiency, accuracy and real-time performance of traditional disease and pest identification and inability to perform environmental regulation.

[0004] To achieve the above objectives, this application provides the following solutions:

[0005] In a first aspect, the present application provides a wheat pest and disease identification and environmental regulation system based on an improved YOLOv10 model, comprising: a data acquisition module, an environmental regulation device, a cloud server, a central processing unit, and a client device; the central processing unit is connected to the data acquisition module, the environmental regulation device, and the cloud server, respectively, and the client device is connected to the environmental regulation device and the cloud server, respectively;

[0006] The data acquisition module is used to collect multiple original images and environmental parameters of the wheat breeding greenhouse;

[0007] The central processing unit is used to integrate the multiple original images to obtain an integrated image, and upload the integrated image to the client device through the cloud server;

[0008] The client device is used to input the integrated image into the wheat pest and disease recognition model to obtain a wheat pest and disease recognition result; the wheat pest and disease recognition result includes the category and location coordinates of the wheat pest and disease; the wheat pest and disease recognition model is obtained by training the improved YOLOv10 model; the improved YOLOv10 model is obtained by improving YOLOv10 using MobileNetV4;

[0009] The central processing unit is further configured to control the environmental adjustment device to adjust the environmental parameters according to the environmental parameters and a preset data range.

[0010] In one embodiment, the environmental parameters include: temperature, humidity, light intensity and carbon dioxide concentration;

[0011] The data acquisition module includes: a monitoring camera, a temperature and humidity sensor, a light intensity sensor and a carbon dioxide concentration sensor;

[0012] The surveillance camera is used to collect the original image;

[0013] The temperature and humidity sensor is used to collect the temperature and humidity;

[0014] The light sensor is used to collect the light intensity;

[0015] The carbon dioxide concentration sensor is used to collect the carbon dioxide concentration.

[0016] In one embodiment, the wheat pest and disease identification and environmental regulation system based on the improved YOLOv10 model further includes: a data storage module; the data storage module is respectively connected to the data acquisition module, the environmental regulation device, the cloud server and the central processing unit;

[0017] The data storage module is used to store the original image, the environmental parameters and the wheat disease and insect pest identification results.

[0018] In one embodiment, the surveillance camera is a Hikvision panoramic zoom high-definition dome camera, the temperature and humidity sensor is an AHT20 temperature and humidity sensor, the light sensor is a BH1750 light sensor, and the carbon dioxide sensor is an SGP30 carbon dioxide sensor.

[0019] In one embodiment, the environment conditioning equipment includes: a fan, a heater, a water pump, and a lamp.

[0020] According to the environmental parameters and the preset data range, controlling the environmental adjustment device to adjust the environmental parameters includes:

[0021] When the temperature is lower than a preset minimum temperature, controlling the heater to start;

[0022] When the humidity is lower than a preset minimum humidity, controlling the water pump to start;

[0023] When the light intensity is lower than a preset minimum light intensity, controlling the lamp to start;

[0024] When the carbon dioxide concentration is lower than a preset minimum carbon dioxide concentration or higher than a preset maximum carbon dioxide concentration, the fan is controlled to start.

[0025] In one embodiment, the environmental conditioning device further comprises: a solenoid valve;

[0026] The solenoid valve is used to control the mixing ratio of the liquid pesticide under the control of the central processing unit.

[0027] In one embodiment, a MySQL database is deployed in the data storage module; the MySQL database is used to store the original image, the environmental parameters and the wheat disease and pest identification results.

[0028] In one embodiment, the cloud server is an EMQX cloud server.

[0029] In a second aspect, the present application provides a method for wheat disease and pest identification and environmental regulation based on an improved YOLOv10 model, which is implemented based on the wheat disease and pest identification and environmental regulation system based on the improved YOLOv10 model described in any of the above items. The method for wheat disease and pest identification and environmental regulation based on the improved YOLOv10 model includes:

[0030] Collect multiple original images and environmental parameters of the wheat breeding greenhouse;

[0031] Integrate multiple original images to obtain an integrated image;

[0032] The integrated image is input into the wheat pest and disease recognition model to obtain a wheat pest and disease recognition result; the wheat pest and disease recognition result includes the category and location coordinates of the wheat pest and disease; the wheat pest and disease recognition model is obtained by training an improved YOLOv10 model; the improved YOLOv10 model is obtained by improving YOLOv10 using MobileNetV4;

[0033] According to the environmental parameters and the preset data range, the environmental adjustment device is controlled to adjust the environmental parameters.

[0034] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0035] The present application discloses a wheat disease and pest identification and environmental regulation system and method based on an improved YOLOv10 model. The wheat disease and pest identification and environmental regulation system based on the improved YOLOv10 model includes: a data acquisition module, an environmental regulation device, a cloud server, a central processing unit and a client device; the data acquisition module collects multiple original images and environmental parameters of a wheat breeding greenhouse; the central processing unit integrates the multiple original images to obtain an integrated image, and uploads the integrated image to the client device through the cloud server; the client device is used to input the integrated image into the wheat disease and pest identification model to obtain a wheat disease and pest identification result; the wheat disease and pest identification result includes the category and location coordinates of the wheat disease and pest; the wheat disease and pest identification model is obtained by training the improved YOLOv10 model; the improved YOLOv10 model is obtained by improving YOLOv10 using MobileNetV4; the central processing unit also controls the environmental regulation device to adjust the environmental parameters according to the environmental parameters and a preset data range. This application deploys a wheat disease and pest recognition model obtained by training an improved YOLOv10 model obtained by improving YOLOv10 using MobileNetV4 in a cloud server to identify wheat diseases and pests, thereby improving the efficiency, accuracy and real-time performance of disease and pest recognition. In addition, the central processing unit is used to control the environmental adjustment equipment to adjust the environmental parameters according to the environmental parameters and the preset data range, thereby realizing greenhouse environment regulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0037] Figure 1 A schematic diagram of the structure of a wheat pest and disease identification and environmental regulation system based on an improved YOLOv10 model provided in one embodiment of the present application;

[0038] Figure 2 This is a schematic diagram of the wheat pest and disease identification and environmental regulation system architecture based on the improved YOLOv10 model;

[0039] Figure 3 Schematic diagram of the improved YOLOv10 model structure. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0041] The purpose of this application is to provide a wheat disease and pest identification and environmental regulation system and method based on an improved YOLOv10 model, aiming to improve the efficiency, accuracy and real-time performance of disease and pest identification and realize greenhouse environment regulation.

[0042] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0043] In an exemplary embodiment, Figure 1 and Figure 2 As shown, a wheat disease and pest identification and environmental regulation system based on an improved YOLOv10 model is provided, including: a data acquisition module, an environmental regulation device, a cloud server, a central processing unit and a client device; the central processing unit is connected to the data acquisition module, the environmental regulation device and the cloud server respectively, and the client device is connected to the environmental regulation device and the cloud server respectively.

[0044] The data acquisition module is used to collect multiple original images and environmental parameters of the wheat breeding greenhouse.

[0045] The central processing unit is used to integrate multiple original images to obtain an integrated image, and upload the integrated image to the client device through the cloud server.

[0046] The client device is used to input the integrated image into the wheat pest and disease recognition model to obtain the wheat pest and disease recognition result; the wheat pest and disease recognition result includes the category and location coordinates of the wheat pest and disease; the wheat pest and disease recognition model is obtained by training the improved YOLOv10 model; the improved YOLOv10 model is obtained by improving YOLOv10 using the fourth version of the mobile neural network (MobileNetV4).

[0047] The central processing unit is also used to control the environmental adjustment device to adjust the environmental parameters according to the environmental parameters and the preset data range.

[0048] Specifically, such as Figure 3 As shown in Figure 1, the improved YOLOv10 model includes a backbone network, a neck network, and a detection head connected in sequence.

[0049] The backbone network includes: Mobile Neural Network Version 4 Module 1 (Deepthwise Separable Convolution + SE Attention Mechanism), Mobile Neural Network Version 4 Module 2 (Ghost Bottleneck Structure), Mobile Neural Network Version 4 Module 3 (Ghost Bottleneck Structure), Mobile Neural Network Version 4 Module 4, Mobile Neural Network Version 4 Module 5 (Bottleneck Structure with SE Attention Mechanism), Mobile Neural Network Version 4 Downsampling Module 1, Mobile Neural Network Version 4 Module 6, Mobile Neural Network Version 4 Downsampling Module 2, Mobile Neural Network Version 4 Module 7, and SE Attention Module (built-in to Mobile Neural Network Version 4), which are connected in sequence. After the integrated image is input into the backbone network, the backbone network processes the integrated image as follows:

[0050] S11: Use the fourth version of the mobile neural network module 1 to perform deep convolution processing on the integrated image and extract the initial feature map.

[0051] S12: The initial feature map passes through the mobile neural network version 4 module 2, the mobile neural network version 4 module 3, the mobile neural network version 4 module 4, the mobile neural network version 4 module 5, the mobile neural network version 4 downsampling module 1, the mobile neural network version 4 module 6, the mobile neural network version 4 downsampling module 2, and the mobile neural network version 4 module 7 in sequence, and performs layer-by-layer multi-scale feature extraction and downsampling compression to extract a high-order semantic feature map.

[0052] S13: After the high-order semantic feature map is enhanced by the SE attention module, the feature map output by the backbone network is obtained.

[0053] The neck network includes: 2 upsampling modules, 4 feature concatenation modules, 2 Mobile Neural Network Version 4 CiB lightweight modules, Mobile Neural Network Version 4 extension module X, Mobile Neural Network Convolution module, Mobile Neural Network Version 4 downsampling module, and Mobile Neural Network Version 4 module 3. After the feature map output by the backbone network is input into the neck network, the neck network processes the feature map output by the backbone network as follows:

[0054] S21: Upsample the feature map output by the backbone network to restore high-resolution information.

[0055] S22: Combine the upsampled features and the backbone intermediate layer features, and fuse them through the feature splicing module to obtain: the fused features.

[0056] S23: The fused features are processed by the multi-level mobile neural network version 4_CiB lightweight module and the downsampling module to enhance the local and global receptive fields and obtain the feature map output by the neck network.

[0057] Detection heads include: one-to-one detection head, one-to-many detection head and Figure 3 The classification prediction module, regression prediction module, intersection-over-union calculation module, and non-maximum suppression module are not shown. After the feature map output by the neck network is input to the detection head, the detection head processes the feature map output by the neck network in the following steps:

[0058] S31: Use the one-to-one detection head module to perform single target detection on the feature map output by the neck network to obtain the target candidate area.

[0059] S32: Use one-to-many detection head module to implement multi-target detection strategy and improve small target detection accuracy.

[0060] S33: Fusion classification and regression prediction results, combined with intersection-over-union calculation and non-maximum suppression algorithm to optimize and screen the bounding box, and output the wheat pest and disease identification results.

[0061] Client devices are used to:

[0062] Real-time display of wheat disease and pest identification results.

[0063] Real-time display of environmental parameters collected by the data acquisition module and transmitted through the central processing unit and cloud server.

[0064] Remotely adjust environmental conditioning equipment.

[0065] Based on the wheat pest and disease identification results, environmental parameters and the operating status of environmental conditioning equipment, DeepSeek-R1's agricultural intelligent question-answering service is used to provide users with breeding recommendations.

[0066] Specifically, users can remotely and manually adjust environmental conditioning devices through the Flutter cross-platform APP (i.e., client device) to ensure that the breeding environment is in the best condition.

[0067] Users can view wheat pest and disease identification results and environmental parameters in real time through the Flutter app, and support historical data backtracking analysis.

[0068] The DeepSeek-R1 model is deployed on a GPU cloud server using the Ollama tool. Once a user logs in, the system uses Ollama's API to retrieve various greenhouse status information (i.e., wheat breeding greenhouses) based on the user's current login status. This information includes real-time and historical environmental parameters, as well as equipment operating status data. The DeepSeek-R1 model then conducts in-depth summarization and analysis of this data, generating recommendations based on the analysis, such as pest and disease control strategies and analysis of optimal crop growth conditions. This provides strong support for users and helps them make informed decisions.

[0069] When environmental parameters exceed the safe range, the system pushes notifications through the APP to remind users to take necessary measures.

[0070] As an optional implementation, the environmental parameters include: temperature, humidity, light intensity and carbon dioxide concentration.

[0071] The data acquisition module includes: surveillance camera, temperature and humidity sensor, light intensity sensor and carbon dioxide concentration sensor.

[0072] Surveillance cameras are used to collect original images.

[0073] The temperature and humidity sensor is used to collect temperature and humidity.

[0074] The light sensor is used to collect light intensity.

[0075] The carbon dioxide concentration sensor is used to collect carbon dioxide concentration.

[0076] Environmental parameters collected by each sensor are transmitted to the central processing unit (WaffleNano) via interfaces such as I2C and UART. After integration, they are transmitted to the cloud server via the MQTT protocol. The cloud server uses the Django framework and provides a RESTful API.

[0077] As an optional implementation, the wheat pest and disease identification and environmental regulation system based on the improved YOLOv10 model also includes: a client device; the client device is connected to the data acquisition module, the environmental regulation device, the cloud server and the central processing unit respectively.

[0078] As an optional implementation, the wheat pest and disease identification and environmental regulation system based on the improved YOLOv10 model also includes: a data storage module; the data storage module is respectively connected to the data acquisition module, the environmental regulation device, the cloud server and the central processing unit.

[0079] The data storage module is used to store original images, environmental parameters and wheat pest and disease identification results.

[0080] As an optional implementation, the surveillance camera is a Hikvision panoramic zoom high-definition dome camera, the temperature and humidity sensor is an AHT20 temperature and humidity sensor, the light sensor is a BH1750 light sensor, and the carbon dioxide sensor is an SGP30 carbon dioxide sensor.

[0081] As an optional implementation, the environment conditioning equipment includes: a fan, a heater, a water pump and a lamp.

[0082] According to the environmental parameters and the preset data range, the environmental adjustment equipment is controlled to adjust the environmental parameters, including:

[0083] When the temperature is lower than the preset minimum temperature, the heater is controlled to start.

[0084] When the humidity is lower than the preset minimum humidity, the water pump is controlled to start.

[0085] When the light intensity is lower than the preset minimum light intensity, the lamp is controlled to start.

[0086] When the carbon dioxide concentration is lower than the preset minimum carbon dioxide concentration or higher than the preset maximum carbon dioxide concentration, the fan is controlled to start.

[0087] Specifically, the preset data range of temperature is 15°C-25°C, the preset data range of humidity is 50%-70%, the preset data range of light intensity is 30,000 lux-60,000 lux, and the preset data range of carbon dioxide concentration is 350ppm-1000ppm.

[0088] Controlling the start of fans can adjust air circulation and reduce the incidence of pests and diseases; controlling the start of heaters can increase the temperature to adapt to different growing environments; controlling the start of water pumps can adjust soil moisture; controlling the start of lamps can supplement light and optimize crop photosynthesis.

[0089] The environmental adjustment device supports I2C, UART, GPIO and PWM control modes, and the main control board of the central processing unit sends instructions to automatically adjust the environment.

[0090] As an optional embodiment, the environment regulating device further includes: a solenoid valve;

[0091] The solenoid valve is used to control the mixing ratio of the liquid pesticide under the control of the central processing unit.

[0092] Specifically, the central processing unit starts the solenoid valve to control the mixing of pesticides based on the matching of the wheat disease and pest identification results with the preset conditions, realizing the integration of intelligent environmental control and disease intervention.

[0093] As an optional implementation, a MySQL database is deployed in the data storage module; the MySQL database is used to store original images, environmental parameters and wheat disease and pest identification results.

[0094] As an optional implementation, the cloud server is an EMQX cloud server.

[0095] The advantages of the system of this application are:

[0096] 1. High recognition accuracy and fast response speed:

[0097] It uses YOLOv10, combined with the MobileNetV4 lightweight network, and uses deep separable convolution and channel pruning technology to reduce the amount of calculation and improve the inference speed.

[0098] YOLOv10 removes the traditional non-maximum suppression step and adopts a consistent dual-label assignment strategy to reduce inference latency, making pest and disease identification more accurate and efficient.

[0099] In actual tests, the detection accuracy (mAP) of the system of this application can reach 99.04%, which is about 3% higher than the detection accuracy of traditional YOLOv4+ResNet50.

[0100] 2. Low computing resource requirements, suitable for edge computing devices:

[0101] The application of traditional models such as YOLOv4 and YOLOv5 in agricultural environments is limited by high computing requirements and is difficult to deploy on embedded devices or low-power smart terminals.

[0102] This application uses MobileNetV4 as the backbone network of YOLOv10 to reduce the number of model parameters, enabling it to be deployed in Flutter applications and run efficiently on mobile devices without relying on high-performance GPUs or cloud computing.

[0103] This optimization reduces inference time. Compared with YOLOv4, the inference speed is increased by 2.3 times, and pest and disease identification can be completed within 200ms.

[0104] 3. Intelligent environmental control to improve crop growth quality:

[0105] Traditional pest and disease control relies on manual identification and spraying of pesticides, which makes it difficult to adjust environmental parameters such as temperature, humidity, and light in a timely and accurate manner, resulting in delays in pest and disease control or waste of resources.

[0106] The system of this application integrates temperature and humidity sensors, light sensors, and carbon dioxide sensors. Combined with the results of wheat pest and disease identification, it dynamically adjusts environmental conditioning equipment such as fans and lamps. While ensuring the stability of the greenhouse environment, it realizes the coordinated control of liquid pesticides and irrigation water, thereby assisting in the intervention of pests and diseases.

[0107] Automatic control: When the system detects that the measured environmental value exceeds the set threshold, it can automatically trigger the environmental adjustment device.

[0108] Remote control: Users can manually adjust device parameters through the Flutter mobile app, improving flexibility and control accuracy.

[0109] 4. Remote monitoring and data visualization to improve agricultural management efficiency:

[0110] Traditional agricultural pest and disease identification systems are mostly managed on PCs or local area networks, making data sharing and remote equipment control difficult.

[0111] This application's system uses a Flutter cross-platform app to deploy a pest and disease identification model locally, enabling the display of wheat pest and disease identification results without relying on cloud-based calls. It also enables environmental parameter visualization, remote equipment control, and DeepSeek-R1-based agricultural intelligent question-and-answer capabilities, enabling users to access agricultural management information anytime, anywhere.

[0112] The MySQL database stores wheat pest and disease identification results, environmental parameters, equipment status, etc. Users can query historical data, analyze the occurrence patterns of pests and diseases, and optimize agricultural management strategies.

[0113] 5. Data sharing and IoT integration to enhance system stability and scalability:

[0114] Traditional pest and disease detection systems mostly use traditional wireless transmission (such as HTTP), which has the risk of data delay and loss, affecting the real-time control of agricultural equipment.

[0115] The system of this application adopts the MQTT protocol to establish an efficient, low-latency, stable and reliable data transmission mechanism between the central processing unit and the cloud server.

[0116] In an exemplary embodiment, a method for identifying wheat pests and diseases and regulating the environment based on an improved YOLOv10 model is provided. The wheat pest and disease identification and environmental regulation system based on the improved YOLOv10 model is implemented based on any of the above items. The method for identifying wheat pests and diseases and regulating the environment based on the improved YOLOv10 model includes:

[0117] Collect multiple original images and environmental parameters of the wheat breeding greenhouse.

[0118] Multiple original images are integrated to obtain an integrated image.

[0119] The integrated image is input into the wheat disease and pest recognition model to obtain the wheat disease and pest recognition result; the wheat disease and pest recognition result includes the category and location coordinates of the wheat disease and pest; the wheat disease and pest recognition model is obtained by training the improved YOLOv10 model; the improved YOLOv10 model is obtained by improving YOLOv10 using MobileNetV4.

[0120] According to the environmental parameters and the preset data range, the environmental adjustment equipment is controlled to adjust the environmental parameters.

[0121] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0122] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0123] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the system, method, and core concept of this application. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of this application. In summary, the contents of this specification should not be construed as limiting this application.

Claims

1. A wheat pest and disease identification and environmental regulation system based on an improved YOLOv10 model, characterized in that: The wheat pest and disease identification and environmental regulation system based on the improved YOLOv10 model includes: a data acquisition module, an environmental regulation device, a cloud server, a central processing unit, and a client device; the central processing unit is connected to the data acquisition module, the environmental regulation device, and the cloud server respectively, and the client device is connected to the environmental regulation device and the cloud server respectively; The data acquisition module is used to collect multiple original images and environmental parameters of the wheat breeding greenhouse; The central processing unit is used to integrate the multiple original images to obtain an integrated image, and upload the integrated image to the client device through the cloud server; The client device is used to input the integrated image into the wheat pest and disease recognition model to obtain a wheat pest and disease recognition result; the wheat pest and disease recognition result includes the category and location coordinates of the wheat pest and disease; the wheat pest and disease recognition model is obtained by training the improved YOLOv10 model; the improved YOLOv10 model is obtained by improving YOLOv10 using MobileNetV4; The central processing unit is further configured to control the environmental adjustment device to adjust the environmental parameters according to the environmental parameters and a preset data range.

2. The wheat pest and disease identification and environmental regulation system based on the improved YOLOv10 model according to claim 1 is characterized in that: The environmental parameters include: temperature, humidity, light intensity and carbon dioxide concentration; The data acquisition module includes: a monitoring camera, a temperature and humidity sensor, a light intensity sensor and a carbon dioxide concentration sensor; The surveillance camera is used to collect the original image; The temperature and humidity sensor is used to collect the temperature and humidity; The light sensor is used to collect the light intensity; The carbon dioxide concentration sensor is used to collect the carbon dioxide concentration.

3. The wheat pest and disease identification and environmental regulation system based on the improved YOLOv10 model according to claim 1 is characterized in that: The wheat pest and disease identification and environmental regulation system based on the improved YOLOv10 model further includes: a data storage module; the data storage module is respectively connected to the data acquisition module, the environmental regulation device, the cloud server and the central processing unit; The data storage module is used to store the original image, the environmental parameters and the wheat disease and insect pest identification results.

4. The wheat pest and disease identification and environmental regulation system based on the improved YOLOv10 model according to claim 2, characterized in that: The surveillance camera is a Hikvision panoramic zoom high-definition dome camera, the temperature and humidity sensor is an AHT20 temperature and humidity sensor, the light sensor is a BH1750 light sensor, and the carbon dioxide sensor is an SGP30 carbon dioxide sensor.

5. The wheat pest and disease identification and environmental regulation system based on the improved YOLOv10 model according to claim 2 is characterized in that: The environmental conditioning equipment includes: a fan, a heater, a water pump and a lamp.

6. The wheat pest and disease identification and environmental regulation system based on the improved YOLOv10 model according to claim 5, wherein the environmental regulation device is controlled to regulate the environmental parameters according to the environmental parameters and the preset data range, comprising: When the temperature is lower than a preset minimum temperature, controlling the heater to start; When the humidity is lower than a preset minimum humidity, controlling the water pump to start; When the light intensity is lower than a preset minimum light intensity, controlling the lamp to start; When the carbon dioxide concentration is lower than a preset minimum carbon dioxide concentration or higher than a preset maximum carbon dioxide concentration, the fan is controlled to start.

7. The wheat pest and disease identification and environment regulation system based on the improved YOLOv10 model according to claim 5, wherein the environment regulation device further comprises: Solenoid valve; The solenoid valve is used to control the mixing ratio of the liquid pesticide under the control of the central processing unit.

8. The wheat pest and disease identification and environmental regulation system based on the improved YOLOv10 model according to claim 4 is characterized in that: A MySQL database is deployed in the data storage module; the MySQL database is used to store the original image, the environmental parameters and the wheat disease and insect pest identification results.

9. The wheat pest and disease identification and environmental regulation system based on the improved YOLOv10 model according to claim 1, characterized in that: The cloud server is the EMQX cloud server.

10. A wheat pest and disease identification and environmental regulation method based on an improved YOLOv10 model, implemented based on the wheat pest and disease identification and environmental regulation system based on the improved YOLOv10 model according to any one of claims 1 to 9, characterized in that: The wheat pest and disease identification and environmental regulation method based on the improved YOLOv10 model includes: Collect multiple original images and environmental parameters of the wheat breeding greenhouse; Integrate multiple original images to obtain an integrated image; The integrated image is input into the wheat pest and disease recognition model to obtain a wheat pest and disease recognition result; the wheat pest and disease recognition result includes the category and location coordinates of the wheat pest and disease; the wheat pest and disease recognition model is obtained by training an improved YOLOv10 model; the improved YOLOv10 model is obtained by improving YOLOv10 using MobileNetV4; According to the environmental parameters and the preset data range, the environmental adjustment device is controlled to adjust the environmental parameters.

Images