Plant growth whole process total factor intelligent control method and system

Abstract

The invention discloses a plant growth whole-process total-factor intelligent management and control method and system, and relates to the field of agricultural production, and the method comprises the steps: collecting a plant growth image and cultivation environment data in real time; determining the current growth state of the plant according to the plant growth image and the cultivation environment data; determining a current growth stage according to the current growth state of the plant, and generating an adjustment strategy according to an environment demand corresponding to the current growth stage; and adjusting the environmental factors of plant growth according to the adjustment strategy. According to the method, the environment requirements of the plants in all growth stages can be accurately met, and the growth efficiency of the plants is improved.

Description

Technical Field

[0001] The present application relates to the field of agricultural production, and in particular to a method and system for intelligently controlling all elements of the entire plant growth process. Background Art

[0002] In modern agricultural production, plant growth is influenced by a variety of environmental factors, such as temperature, humidity, light intensity, carbon dioxide concentration, and soil nutrients. These factors play a crucial role in different stages of plant growth, such as germination, vegetative growth, flowering and fruiting. However, due to the uncertainty of the natural environment and the complexity of agricultural production conditions, traditional cultivation methods have difficulty in achieving precise control of these key factors, resulting in low plant growth efficiency, serious waste of resources, and even compromising crop yield and quality.

[0003] In recent years, with the continuous development of the Internet of Things, sensor technology, and artificial intelligence, smart agriculture has gradually become an effective means of addressing this problem. By introducing intelligent equipment and digital management platforms, environmental monitoring and regulation in agricultural production have made some progress. However, most current smart agriculture systems only regulate a single environmental factor and lack comprehensive control over the entire plant growth process. This is especially true given the significant differences in environmental requirements across different growth stages, making it difficult for existing systems to achieve precise adjustments. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for intelligent management and control of all elements of the entire plant growth process, which can ensure that the environmental needs of plants at various growth stages are accurately met and improve the growth efficiency of plants.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In the first aspect, the present application provides a method for intelligently controlling all elements of the entire plant growth process, including:

[0007] Real-time collection of plant growth images and cultivation environment data;

[0008] determining a current growth state of the plant according to the plant growth image and the cultivation environment data;

[0009] Determining a current growth stage according to the current growth state of the plant, and generating a regulation strategy according to environmental requirements corresponding to the current growth stage;

[0010] Environmental factors for plant growth are regulated according to the regulation strategy.

[0011] In one embodiment, the cultivation environment data includes temperature, humidity, CO2 concentration, light intensity and soil moisture.

[0012] In one embodiment, the current growth status of the plant includes appearance characteristics, leaf color, leaf morphology, health index, photosynthesis efficiency, nutrient absorption capacity, and pest and disease characteristics.

[0013] In one embodiment, determining the current growth state of the plant based on the plant growth image and the cultivation environment data specifically includes:

[0014] performing histogram equalization and noise removal processing on the plant growth image in sequence to obtain a denoised image;

[0015] Performing image segmentation on the denoised image to extract the plant area and obtain a plant area image;

[0016] performing noise removal and normalization processing on the cultivation environment data in sequence to obtain normalized environment data;

[0017] fusing the plant area image with the normalized environmental data to obtain fused data;

[0018] A deep learning algorithm is used to extract features from the fused data to obtain the current growth status of the plant.

[0019] In one embodiment, the deep learning algorithm is a pre-trained improved yolov10; the improved yolov10 is to replace the C2f module of yolov10 with a C2f_iRMB module; wherein the C2f_iRMB module is to replace the Bottleneck module in the C2f module with an iRMB module.

[0020] In one embodiment, the regulation strategy includes light, temperature, humidity, CO2 concentration, nutrient solution ratio and irrigation amount.

[0021] In one embodiment, a regulation strategy is generated according to the environmental requirements corresponding to the current growth stage, specifically including: according to the current growth stage, based on a pre-established database, a weighted decision algorithm and a multi-objective optimization algorithm are used to generate a regulation strategy for each plant; the database stores the environmental requirements of plants at different growth stages, the growth patterns of various types of plants, and the impact of various environmental factors on plant growth in historical data.

[0022] In a second aspect, the present application provides an intelligent management and control system for the entire plant growth process, including:

[0023] A data acquisition device for collecting plant growth images and cultivation environment data in real time;

[0024] A central data processing platform is configured to determine the current growth state of the plant based on the plant growth image and the cultivation environment data; determine the current growth stage based on the current growth state of the plant, and generate an adjustment strategy based on the environmental requirements corresponding to the current growth stage;

[0025] Environmental control equipment is used to adjust environmental factors for plant growth according to the adjustment strategy.

[0026] In one embodiment, the data acquisition device includes a visual sensor and an environmental sensor.

[0027] In one embodiment, the cultivation environment data includes temperature, humidity, CO2 concentration, light intensity and soil moisture; the visual sensor includes a camera, a depth camera and an infrared sensor; the environmental sensor includes a temperature sensor, a humidity sensor, a CO2 concentration sensor, a light intensity sensor and a soil moisture sensor.

[0028] According to the specific embodiments provided in this application, this application has the following technical effects:

[0029] The present application provides a method and system for intelligent management and control of all elements of the entire plant growth process. By combining multi-dimensional data of plant growth images and cultivation environment data, it can more accurately evaluate the growth status of plants, further determine the current growth stage of plants, and generate corresponding adjustment strategies. It can dynamically adjust environmental factors according to the growth stage of plants and changes in the external environment, avoiding the problems of fixed environmental conditions and untimely manual adjustment in traditional agriculture, thereby improving the efficiency and quality of plant growth. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, 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.

[0031] Figure 1 This is an overall flow chart of a method for intelligently controlling all elements of the entire plant growth process provided by one embodiment of the present application;

[0032] Figure 2 A detailed flow chart of a method for intelligently controlling all elements of the entire plant growth process provided in one embodiment of the present application;

[0033] Figure 3 This is a schematic diagram of the structure of the improved yolov10 in one embodiment of the present application;

[0034] Figure 4This is a schematic diagram of the structure of the C2f_iRMB module in one embodiment of the present application;

[0035] Figure 5 This is a schematic diagram of the structure of the Bottleneck module in the C2f_iRMB module in one embodiment of the present application;

[0036] Figure 6 This is a block diagram of a system for intelligent management and control of all elements of the entire plant growth process provided by one embodiment of the present application. DETAILED DESCRIPTION

[0037] 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.

[0038] This application uses digital and intelligent technologies to comprehensively monitor and precisely regulate key growth factors of plants throughout the entire process from seed germination to maturity and harvest, covering all stages of plant growth, including germination, vegetative growth, flowering, and fruiting. Targeted at the specific needs of each stage, through real-time data collection, intelligent algorithm analysis, and automated control systems, dynamic adjustments to multiple growth factors such as temperature, humidity, light, carbon dioxide concentration, soil nutrition, irrigation, and fertilization are achieved to ensure that the environmental needs of plants at each growth stage are accurately met, thereby improving plant growth efficiency, yield, and quality, reducing resource waste, and enhancing the overall benefits of agricultural production.

[0039] 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.

[0040] In an exemplary embodiment, Figure 1 As shown, a method for intelligent control of all elements of the entire plant growth process is provided, including the following steps 101 to 104. Step 101 is performed by visual sensors and environmental sensors, steps 102 and 103 are performed by a central data processing platform, and step 104 is performed by an environmental control device.

[0041] Step 101: real-time collection of plant growth images and cultivation environment data, including temperature, humidity, CO2 concentration, light intensity, and soil moisture.

[0042] In order to accurately monitor the growth status of plants and their environmental factors, such as Figure 2As shown, this application first deploys visual sensors and environmental sensors (such as temperature sensors, humidity sensors, CO2 concentration sensors, light intensity sensors, soil moisture sensors, etc.) in the plant cultivation environment.

[0043] Visual sensors primarily include high-definition cameras, depth cameras, and infrared sensors. These devices, through their physical layout and coordinated sensing technology, ensure comprehensive coverage of all key growth zones within the cultivation area. HD cameras capture the overall appearance of plants, capturing key growth indicators such as morphological characteristics, leaf color changes, and stem growth. Depth cameras acquire three-dimensional spatial information about plants, helping to analyze their structure and their spatial relationship to the surrounding environment. Infrared sensors operate in low-light environments, capturing image information in these conditions to ensure 24 / 7 monitoring of plant growth.

[0044] The above-mentioned visual sensors are arranged in different locations in the plant cultivation environment to ensure that detailed images of plants are obtained from multiple angles and under different lighting conditions. Especially during the critical growth stage of the plant, the number of visual sensors and the acquisition frequency can be increased in a timely manner to ensure the accuracy and timeliness of the data.

[0045] Depending on actual needs, vision sensors can also capture plant growth images in timed capture mode or through triggers when they sense changes in plant growth status, ensuring that critical image information is captured at different growth stages. The frequency of image capture is typically set based on the plant's growth cycle and specific monitoring requirements, and may include capturing images hourly, daily, or every time a plant's status changes.

[0046] Environmental sensors primarily include temperature sensors, humidity sensors, CO2 concentration sensors, light intensity sensors, and soil moisture sensors. Temperature sensors accurately measure changes in air temperature and humidity within a plant cultivation environment, providing an optimal growth environment for plants. CO2 concentration sensors monitor carbon dioxide concentrations to ensure efficient photosynthesis and avoid growth problems caused by environmental factors. Light intensity sensors record the light intensity in the plant's area in real time, helping to adjust light source configuration to avoid insufficient or excessive light. Soil moisture sensors monitor the moisture level of plant roots to prevent negative impacts on plant growth caused by excessively dry or wet soil. Environmental sensors are strategically positioned based on the plant's growth needs and the specific conditions of the cultivation environment to ensure comprehensive collection of environmental data.

[0047] All of the aforementioned sensors are connected to a central data processing platform via IoT communication protocols, transmitting collected data in real time. Wireless network technology enables efficient and stable data transmission. To ensure data transmission stability and image quality, plant growth images are typically compressed, for example using JPEG format, to reduce network bandwidth pressure while maintaining image clarity. During transmission, edge computing devices perform preprocessing to remove unnecessary background noise and ensure data stability. Plant growth images and cultivation environment data are also encrypted to ensure information security and integrity.

[0048] This application utilizes appropriate transmission methods based on the needs of different sensor types, ensuring data is delivered to the central data processing platform without delay or error, providing timely and accurate raw data support for subsequent data processing and intelligent control. Through a series of hardware deployments, comprehensive data on plant growth status and environmental conditions can be obtained in real time during each cultivation cycle, laying the foundation for subsequent analysis and control.

[0049] Step 102: Determine the current growth status of the plant based on the plant growth image and the cultivation environment data. The current growth status of the plant includes appearance characteristics, leaf color, leaf morphology, health index, photosynthesis efficiency, nutrient absorption capacity, and pest and disease characteristics.

[0050] In a specific application example, step 102 includes the following steps 21 to 25.

[0051] Step 21 : performing histogram equalization and noise removal processing on the plant growth image in sequence to obtain a denoised image.

[0052] Histogram equalization is used to enhance the contrast of plant growth images, improve image brightness and contrast, and make plant details more prominent. Gaussian filtering or median filtering is used to remove noise caused by light, weather, and other factors during image capture, improving the clarity of plant growth images.

[0053] Step 22: performing image segmentation on the denoised image to extract the plant area to obtain a plant area image.

[0054] This application uses image segmentation techniques to separate plant areas from the background, ensuring that subsequent analysis focuses on the plant itself. This can be achieved using thresholding, edge detection (such as the Sobel operator), or deep learning-based semantic segmentation algorithms (such as U-Net). Image segmentation extracts the outlines and morphological information of key plant parts, such as leaves and stems.

[0055] Step 23 , performing noise removal and normalization processing on the cultivation environment data in sequence to obtain normalized environment data to ensure the stability and accuracy of the data.

[0056] To ensure data accuracy and stability, the cultivation environment data will first undergo preliminary noise removal and smoothing using a low-pass filter to eliminate instantaneous fluctuations in environmental sensor readings, effectively reducing noise fluctuations caused by sensor drift or external environmental interference, and obtaining more stable environmental data. For example, a simple moving average filter algorithm is used to smooth temperature and humidity, using the following formula:

[0057]

[0058] Among them, S filtered (t) is the smoothed signal value at time t, S(tk) is the original signal value at time tk, and K is the size of the smoothing window.

[0059] For data such as CO2 concentration and light intensity, data normalization can also be performed so that different environmental data can be compared and analyzed on a unified scale, which helps to more accurately assess the impact of environmental changes on plant growth. For example, for the processing of CO2 concentration, it can be normalized using the following formula:

[0060]

[0061] Among them, C current Indicates the current CO2 concentration, C min is the minimum CO2 concentration in historical data, C max is the maximum CO2 concentration in historical data, C normalized Indicates the normalized CO2 concentration. Normalization processing.

[0062] Normalized environmental data is transmitted to a central data processing platform for further modeling and intelligent decision-making. Machine learning algorithms are further employed to predict the potential impact of future environmental changes on plant growth based on the relationship between historical environmental data and plant growth feedback. Environmental control devices, such as thermostats, humidity regulators, and CO2 generators, are automatically adjusted to ensure ideal plant growth conditions.

[0063] Step 24: Fusing the plant area image with the normalized environmental data to obtain fused data.

[0064] To comprehensively assess plant growth status, plant region images are effectively integrated with normalized environmental data to identify plant health, photosynthesis efficiency, and pest and disease characteristics. Furthermore, normalized environmental data and changes in plant appearance are used to predict plant growth stages (such as germination, vegetative growth, and flowering), and corresponding regulation strategies are generated based on the needs of different growth stages.

[0065] The integration of visual and environmental data is crucial for monitoring and analyzing plant growth throughout its lifecycle. Visual data primarily comes from high-definition cameras and depth cameras, reflecting changes in plant appearance (such as leaf color, morphology, and the presence of pests and diseases). Environmental data, on the other hand, is collected in real time through temperature and humidity sensors, CO2 concentration sensors, and light intensity sensors, capturing key factors in the plant's growth environment (such as temperature, humidity, CO2 concentration, and light intensity). Data fusion algorithms effectively integrate these sensor data to generate a comprehensive analysis model.

[0066] Common data fusion algorithms include Kalman filtering and deep learning fusion models. Through continuous learning and adjustment, deep learning fusion models can discover the complex relationships between plant region images and normalized environmental data, thereby inferring the plant's growth status. First, leaf features such as color, morphology, and texture are extracted from the plant region images. Temperature, humidity, and light characteristics are then extracted using the normalized environmental data. Finally, a deep learning fusion model (such as a CNN or LSTM) is trained to learn the nonlinear relationships between the plant region images and the normalized environmental data. The training process uses supervised learning from labeled datasets, enabling the deep learning fusion model to accurately predict plant health and the required environment based on the plant region images and normalized environmental data.

[0067] In an optional embodiment, the following algorithm is used to effectively combine the plant area image and the normalized environmental data:

[0068] (1) Weighted average method: We can assign weights to different data sources based on the credibility of different sensors, thereby achieving more accurate fusion. For example, if the image quality of a visual sensor is high, we can give it a higher weight: Among them, D fused is the fused data, w n is the weight of the nth data source, D n The data collected from the nth data source, where N is the number of data sources.

[0069] (2) Deep learning fusion model: The plant region image and normalized environmental data are input into a convolutional neural network or recurrent neural network model. The model automatically learns the optimal fusion method through training. The multi-layer neural network model comprehensively considers the plant appearance, environmental factors, and the interaction between them to infer the plant's growth status.

[0070] Step 25: Use a deep learning algorithm to extract features from the fused data to obtain the current growth status of the plant.

[0071] The evaluation of plant growth status can be measured through multiple dimensions, including leaf color, leaf morphology, pest and disease characteristics, etc. Combined with environmental data (such as temperature, humidity, light, CO2 concentration, etc.), it can be inferred whether the plant is in the optimal growth environment.

[0072] For example, changes in leaf color can be quantified using RGB values or the HSV model. If the green component (i.e., chromaticity) of the leaves decreases significantly, it may indicate a decrease in the plant's photosynthesis efficiency, which in turn indicates a lack of water, light, or nutrients.

[0073] In an optional embodiment, the deep learning algorithm is a convolutional neural network (CNN), which can automatically learn and identify features such as color changes, morphological changes, and damage caused by pests and diseases on plant leaves. In the specific implementation process, deep network architectures such as ResNet, VGGNet or InceptionNet can be used to analyze the health index, photosynthesis efficiency, nutrient absorption capacity, and pest and disease characteristics in plant area images through neural network training and reasoning. The training of neural networks usually requires a large amount of labeled data sets, and a graphics processing unit (GPU) is used to accelerate the training process of the neural network to improve the accuracy and robustness of the deep learning algorithm.

[0074] For example, the health index can be calculated by the following formula:

[0075]

[0076] Among them, H is the health index, (x i ,y i ) is the coordinate of the i-th pixel, I current (x i ,y i ) is the brightness or color value of the i-th pixel in the plant area image, I healthy (x i ,y i ) is the color value of the pixel at the same position under ideal health conditions, and N is the total number of pixels in the plant area image.

[0077] Photosynthesis efficiency is an important indicator of plant health. This application estimates plant photosynthesis efficiency by analyzing data such as leaf color, leaf area, and light intensity. This efficiency can be used to further infer a plant's nutrient absorption capacity. Specifically, a preliminary estimate can be made using the following formula based on a photosynthesis model:

[0078]

[0079] Where PE is photosynthesis efficiency, I leaf is the light absorption intensity of the leaves, L is the light intensity, and A is the leaf area.

[0080] In a preferred embodiment, the deep learning algorithm is a pre-trained improved yolov10. Figure 3 As shown in the figure, the improvement of yolov10 is to replace the C2f module of yolov10 with the C2f_iRMB module. Among them, the C2f_iRMB module replaces the Bottleneck module in the C2f module with the iRMB module.

[0081] Yolov10 is improved through the C2f module and iRMB module, combining the lightweight CNN architecture with the attention-based model structure to create an efficient mobile network. While maintaining the high precision and lightweight of the model, it achieves effective utilization of computing resources and high accuracy.

[0082] like Figure 4 As shown in the figure, the C2f_iRMB module first passes the input data through the first CBS module, and then uses the chunk method to split the output into two parts along the channel dimension; then one of the split data is sent to j Bottleneck modules in sequence through a loop for processing, and the data is updated every time it passes through a Bottleneck module, and the data flow between each Bottleneck module is sequentially connected. After processing, the data output by each Bottleneck module is spliced with the other split data in the channel dimension, and then sent to the second CBS module for the final feature integration transformation, and the result processed by the C2f_iRMB module is output.

[0083] Among them, such as Figure 5 As shown in the figure, the input data in the Bottleneck module passes through two convolutional layers in sequence for feature extraction and transformation, and then is sent to the iRMB module for further processing to finally obtain the features processed by the iRMB module.

[0084] The improved YOLOv10 architecture proposed in this application is particularly suitable for dense prediction tasks on mobile devices because it can provide efficient performance in environments with limited computing power. C2f_iRMB allows capturing and utilizing long-distance dependencies while keeping the model lightweight, which is critical for tasks such as image classification, object detection, and semantic segmentation. This design enables the improved YOLOv10 to run efficiently on resource-constrained devices while maintaining or improving prediction accuracy.

[0085] Step 103: Determine the current growth stage based on the current growth state of the plant and generate a regulation strategy based on the environmental requirements corresponding to the current growth stage. The regulation strategy includes light, temperature, humidity, CO2 concentration, nutrient solution ratio, and irrigation amount.

[0086] By combining plant growth cycles with environmental parameters, it's possible to infer a plant's growth stage. Deep learning algorithms, such as long short-term memory (LSTM), combined with normalized environmental data and temporal features of plant region images, can determine a plant's growth stage (such as germination, vegetative growth, and flowering) in real time.

[0087] For example, by analyzing changes in leaf morphology and light intensity, it is possible to predict whether a plant has entered the vegetative growth phase or the flowering phase. Incorporating normalized environmental data (such as CO2 concentration and humidity) can further optimize the accuracy of growth phase recognition.

[0088] By analyzing plant region images and combining deep learning techniques, it is possible to identify leaf pest and disease characteristics, such as spots, discoloration, and wilting. By combining this with environmental data (such as humidity, temperature, and CO2 concentration), it is possible to assess whether plants are affected by pests and diseases and propose appropriate prevention and control measures.

[0089] In a specific application example, step 103 generates a regulation strategy based on the environmental requirements corresponding to the current growth stage, specifically including: generating a regulation strategy for each plant based on the current growth stage, using a weighted decision algorithm and a multi-objective optimization algorithm (such as a genetic algorithm or a particle swarm optimization algorithm) based on a pre-established database. For example: increasing the light intensity to 1000 lux for 6 hours; starting the irrigation equipment with an irrigation volume of 200 ml / hour for 30 minutes; starting the CO2 generator to increase the CO2 concentration to 800 ppm, etc. The database stores the environmental requirements of plants at different growth stages, the growth patterns of various plants, and the impact of various environmental factors on plant growth in historical data.

[0090] This application not only takes into account the current needs of plants, but also integrates factors such as plant variety, cultivation stage, and environmental conditions to generate customized environmental adjustment strategies for each plant. Based on the fused data, a personalized growth status report is generated for each plant. The growth status report includes information on multiple dimensions such as the plant's health score, photosynthesis efficiency, nutrient absorption, pest and disease problems, and growth stage. The growth status report can help growers understand the growth of plants in real time, adjust the cultivation environment in a timely manner, optimize crop growth, and improve crop yield and quality.

[0091] In a specific application example, a weighted decision-making algorithm is used to assign different adjustment weights based on the needs of the plants. For example, if the plant is short of water and the temperature is too high, the irrigation equipment will be activated first and the temperature will be adjusted; if the plant is in a state of low photosynthesis efficiency, the light intensity will be increased; the water demand and irrigation duration can also be automatically optimized according to the plant's growth stage to avoid excessive or insufficient irrigation. The formula of the weighted decision-making algorithm is as follows:

[0092] S strategy =w1·ΔL+w2·ΔT+w3·ΔH+w4·ΔCO2;

[0093] Among them, S strategy represents the adjustment strategy, ΔL is the light adjustment amount, ΔT is the temperature adjustment amount, ΔH is the humidity adjustment amount, ΔCO2 is the CO2 concentration adjustment amount, w1 is the light adjustment weight coefficient, w2 is the temperature adjustment weight coefficient, w3 is the humidity adjustment weight coefficient, and w4 is the CO2 concentration adjustment weight coefficient.

[0094] Multi-objective optimization algorithms can also be used to further refine regulation strategies based on the characteristics of different plant varieties and cultivation stages. For example, when performing irrigation operations, the water volume and irrigation duration can be automatically optimized based on the plant's water needs and actual soil moisture conditions, combined with the plant's growth stage (such as vegetative growth or flowering), effectively avoiding over-irrigation or under-irrigation.

[0095] Step 104: regulating environmental factors for plant growth according to the regulation strategy.

[0096] After the adjustment strategy is generated, it is sent to environmental control devices such as LED lights, humidifiers, dehumidifiers, thermostats, CO2 generators, irrigation equipment, etc. These devices automatically perform corresponding operations according to the adjustment strategy to ensure that plants grow in ideal environmental conditions.

[0097] For example, the brightness of the LED fill light is adjusted when light intensity needs to be increased; the humidifier or dehumidifier is activated when humidity does not meet requirements; and the CO2 generator is activated when CO2 concentration is too low. These operations dynamically adjust the working status of the equipment according to the actual needs of the plants to ensure healthy growth.

[0098] Through intelligent regulation strategy generation and execution mechanisms, this application can provide precise environmental regulation for each plant at different growth stages, maximize the plant's photosynthesis efficiency, optimize its nutrient absorption, reduce the occurrence of diseases and pests, and thus improve the growth quality and yield of crops.

[0099] In another exemplary embodiment, the regulation strategy is continuously optimized based on the actual plant growth and changes in environmental conditions. After each regulation operation, real-time feedback is provided on plant and environmental changes, and subsequent operations are adjusted based on this feedback. For example, when irrigation equipment is activated, soil moisture sensors monitor soil moisture changes. If the soil moisture sensor indicates that the soil moisture has reached a set optimum level, irrigation is stopped, thereby preventing over-irrigation from negatively impacting the plants.

[0100] In addition, the growth stage of a plant is a dynamically changing process. Over time, a plant may enter the flowering stage or the mature stage from the vegetative growth stage. The demand for environmental conditions in each growth stage will change. The present application can adjust the regulation strategy in real time according to the characteristics of the different growth stages of the plant. For example, after entering the flowering stage, the plant may have an increased demand for light and a reduced demand for water, and the light intensity and irrigation frequency will be automatically optimized to ensure that the plant obtains the best growth environment. Therefore, the present application can not only adjust the operation according to the instant feedback, but also automatically adjust the strategy at different stages of the plant growth cycle to avoid over-regulation and ensure that the needs of the plant at different stages are met.

[0101] In order to further improve the accuracy of plant growth status analysis and the precision of regulation strategies, this application continuously learns and optimizes historical data through built-in machine learning algorithms. By collecting a large amount of sensor data (such as environmental parameters, plant growth data, etc.) and visual data (such as changes in plant appearance) for training, and through continuously updated machine learning models, it automatically identifies the growth patterns and demand changes of plants. As data accumulates, it can more accurately predict the needs of plants at different growth stages, thereby generating more personalized regulation strategies for each plant.

[0102] For example, by analyzing historical data, it can be discovered that certain plant varieties grow better under specific environmental conditions, or that certain combinations of environmental factors have a significant impact on plant health. By analyzing this data, machine learning models can optimize plant growth assessment models, improve the accuracy of health predictions, and dynamically adjust the effectiveness of strategy execution. As the algorithm runs over time, the learning process continues to improve, enabling it to better identify potential plant health issues and make adaptive adjustments.

[0103] In addition, with the changes in different cultivation environments and the introduction of different plant varieties, this application will further improve its adaptability to plant needs. It will not only rely on preset rules and algorithms, but will be able to dynamically adjust learning strategies based on new data and plant growth conditions, thereby continuously optimizing the effects of regulatory decisions and realizing an adaptive and intelligent plant growth environment management method.

[0104] The beneficial effects of this application are as follows:

[0105] (1) This application combines visual sensors (high-definition cameras, depth cameras, infrared sensors) with environmental sensors (temperature and humidity, CO2 concentration, light intensity, etc.) to obtain multi-dimensional plant growth data. This application uses deep learning algorithms for data fusion analysis, enabling more accurate assessment of plant growth and health. Compared to single environmental monitoring methods, this application can provide more comprehensive plant growth information, significantly improving the accuracy and intelligence of environmental control.

[0106] (2) This application uses real-time collected environmental and plant data, combined with artificial intelligence algorithms, to automatically generate personalized regulation strategies to ensure that plants are in the optimal growth environment. It can dynamically adjust multiple factors such as light, temperature, humidity, and CO2 concentration according to the plant's growth stage and external environmental changes, avoiding the problems of fixed environmental conditions and untimely manual adjustments in traditional agriculture, thereby improving the efficiency and quality of plant growth.

[0107] (3) The present application has a real-time dynamic feedback mechanism, which can provide real-time feedback of plant and environmental change data through sensors after each adjustment, and continuously optimize the adjustment strategy based on these data. It can quickly respond to changes in plant needs and environmental changes, avoid over-adjustment or environmental imbalance, improve the adaptability and flexibility of the method, and surpass the common lag control problem in existing technologies.

[0108] In summary, this application significantly improves the accuracy of data collection, analysis, and regulation throughout the entire plant growth process by combining advanced visual sensors with environmental sensors and utilizing deep learning and intelligent control technologies. Compared to existing technologies, this application has significant advantages in multi-dimensional data fusion, intelligent regulation strategy generation and execution, and real-time feedback and adaptive optimization, and can provide plants with more accurate growth environment regulation. This full-factor intelligent management and control method not only improves the efficiency and quality of plant growth, but also provides a more efficient and reliable solution for intelligent management in modern agriculture, smart greenhouses, horticultural cultivation, and other fields, and has broad application prospects.

[0109] In an exemplary embodiment, Figure 6 As shown, a plant growth whole process and all-factor intelligent management and control system is provided, comprising: a data acquisition device 601, a central data processing platform 602 and an environmental control device 603. The central data processing platform 602 is connected to the data acquisition device 601 and the environmental control device 603 respectively.

[0110] Data acquisition device 601 is used to collect plant growth images and cultivation environment data in real time. The data acquisition device includes a visual sensor and an environmental sensor. The cultivation environment data includes temperature, humidity, CO2 concentration, light intensity, and soil moisture. The visual sensor includes a camera, a depth camera, and an infrared sensor. The environmental sensors include a temperature sensor, a humidity sensor, a CO2 concentration sensor, a light intensity sensor, and a soil moisture sensor.

[0111] The central data processing platform 602 is used to determine the current growth state of the plant based on the plant growth image and the cultivation environment data; determine the current growth stage based on the current growth state of the plant, and generate an adjustment strategy based on the environmental requirements corresponding to the current growth stage.

[0112] The environmental control device 603 is used to adjust the environmental factors for plant growth according to the adjustment strategy.

[0113] 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.

[0114] In this application, all actions to obtain signals, information or data are carried out in compliance with the relevant data protection laws and policies of the country where they are located and with the authorization given by the owner of the corresponding device.

[0115] The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may include, but are not limited to, general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic units, data processing logic units based on quantum computing, and the like.

[0116] 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.

[0117] 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 method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for intelligent control of all elements of plant growth throughout the entire process, characterized in that: include: Real-time collection of plant growth images and cultivation environment data; determining a current growth state of the plant according to the plant growth image and the cultivation environment data; Determining a current growth stage according to the current growth state of the plant, and generating a regulation strategy according to environmental requirements corresponding to the current growth stage; Environmental factors for plant growth are regulated according to the regulation strategy.

2. The method for intelligent control of all elements of the whole process of plant growth according to claim 1, characterized in that: The cultivation environment data includes temperature, humidity, CO2 concentration, light intensity and soil moisture.

3. The method for intelligent control of all elements of the whole process of plant growth according to claim 1, characterized in that: The current growth status of the plant includes appearance characteristics, leaf color, leaf morphology, health index, photosynthesis efficiency, nutrient absorption capacity and pest and disease characteristics.

4. The method for intelligent control of all elements of the whole process of plant growth according to claim 1, characterized in that: Determining the current growth state of the plant according to the plant growth image and the cultivation environment data, specifically including: performing histogram equalization and noise removal processing on the plant growth image in sequence to obtain a denoised image; Performing image segmentation on the denoised image to extract the plant area and obtain a plant area image; performing noise removal and normalization processing on the cultivation environment data in sequence to obtain normalized environment data; fusing the plant area image with the normalized environmental data to obtain fused data; A deep learning algorithm is used to extract features from the fused data to obtain the current growth status of the plant.

5. The method for intelligent control of all elements of the whole process of plant growth according to claim 4 is characterized in that: The deep learning algorithm is a pre-trained improved yolov10; the improved yolov10 is to replace the C2f module of yolov10 with the C2f_iRMB module; wherein, the C2f_iRMB module is to replace the Bottleneck module in the C2f module with the iRMB module.

6. The method for intelligent control of all elements of the entire plant growth process according to claim 1, characterized in that: The regulation strategy includes light, temperature, humidity, CO2 concentration, nutrient solution ratio and irrigation amount.

7. The method for intelligent control of all elements of the whole process of plant growth according to claim 1, characterized in that: Generate adjustment strategies based on the environmental requirements of the current growth stage, including: Based on the current growth stage and a pre-established database, a weighted decision-making algorithm and a multi-objective optimization algorithm are used to generate a regulation strategy for each plant. The database stores the environmental requirements of plants at different growth stages, the growth patterns of various plants, and the impact of various environmental factors on plant growth in historical data.

8. An intelligent management and control system for the entire plant growth process, characterized in that: include: A data acquisition device for collecting plant growth images and cultivation environment data in real time; A central data processing platform is configured to determine the current growth state of the plant based on the plant growth image and the cultivation environment data; determine the current growth stage based on the current growth state of the plant, and generate an adjustment strategy based on the environmental requirements corresponding to the current growth stage; Environmental control equipment is used to adjust environmental factors for plant growth according to the adjustment strategy.

9. The plant growth whole process and all-factor intelligent management and control system according to claim 8, characterized in that: The data acquisition device includes a visual sensor and an environmental sensor.

10. The plant growth whole process and all-factor intelligent management and control system according to claim 9, characterized in that: The cultivation environment data includes temperature, humidity, CO2 concentration, light intensity and soil moisture; the visual sensor includes a camera, a depth camera and an infrared sensor; the environmental sensor includes a temperature sensor, a humidity sensor, a CO2 concentration sensor, a light intensity sensor and a soil moisture sensor.

Images