Intelligent accurate strawberry growth management system and method based on Internet of Things

Abstract

The invention relates to the technical field of intelligent agriculture, and particularly discloses an intelligent precise strawberry growth management system and method based on the Internet of Things, and the system comprises a sensing layer which collects the actual growth environment and growth vigor data of strawberry seedlings in real time; the growth environment analysis layer compares actual and target growth environment data to obtain a deviation as a control analysis result; the growth vigor analysis layer gives out a growth vigor analysis result in combination with the growth vigor data and the current target growth stage; the growth environment optimization layer calculates the environment data supplement amount according to the two results, optimizes the target growth environment control data of the adjacent next control period, and obtains the optimal growth environment control data; the control layer regulates and controls execution equipment in real time according to the optimal growth environment control data, obtains the newest growth management result of the strawberry seedlings, and realizes accurate control and optimization of the growth environment of the strawberry seedlings; the accuracy and scientificity of strawberry growth management are improved, and the yield and quality of strawberries can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of smart agriculture, and particularly to an intelligent and precise strawberry growth management system and method based on the Internet of Things. Background Art

[0002] In recent years, the application of Internet of Things technology in the agricultural field has been gradually deepened, and strawberry planting management has gradually developed towards intelligence and dataization. In the prior art, a strawberry growth monitoring system based on the Internet of Things usually obtains strawberry growth environment data and plant phenotype information in real time by deploying environmental sensors (such as temperature and humidity, light intensity, soil pH value, CO2 concentration, etc.) and image acquisition devices. Some systems combine wireless communication technologies (such as LoRa, NB-IoT) to transmit data to a cloud platform for storage and analysis. In terms of environmental regulation, existing technologies achieve automatic control by setting fixed thresholds or simple logic rules (such as threshold-triggered irrigation and ventilation). Some studies have attempted to introduce machine learning models (such as decision trees, random forests) to perform correlation analysis on historical environmental data and yield to generate static regulation strategies. In terms of growth trend monitoring, existing technologies mostly use image recognition or spectral analysis methods to evaluate indicators such as strawberry leaf color, plant height, and fruit maturity, and compare them with a preset growth stage model. In addition, some systems integrate an expert knowledge base to provide farmers with suggestions on fertilization and pest control.

[0003] However, the prior art generally focuses on single-dimensional data collection or control execution, lacking the ability of collaborative analysis and dynamic optimization of multi-source heterogeneous data. Existing systems mostly independently process environmental data and growth trend data, without establishing a dynamic coupling relationship between the two, resulting in the disconnection between the environmental regulation strategy and the actual physiological needs of the plants. For example, irrigation is triggered only based on the soil moisture threshold, ignoring the differences in plant transpiration efficiency and water requirements during the fruit swelling period. Relying on preset fixed thresholds or historical data models, it is impossible to dynamically adapt to the optimal environmental parameter changes in different growth stages of strawberries. Especially when climate change is frequent or the facility environment fluctuates, it is easy to cause resource waste (such as excessive light supplementation) or growth inhibition (such as failure to intervene in time when the temperature suddenly rises). Existing growth trend analysis models are mostly based on general growth cycle divisions, resulting in deviation in growth trend diagnosis. For example, the same leaf area index may correspond to opposite regulation requirements during the flower bud differentiation period and the fruit maturity period. The execution device control only compensates for the current environmental deviation, without rolling optimization of the target parameters for the next control cycle by combining future growth trend prediction, and it is easy to fall into a vicious cycle of "lag response - frequent adjustment". The perception, analysis, and control layers mostly adopt an isolated architecture, and the data flow is unidirectional, without forming a "monitoring - decision - execution - feedback" closed loop, making it difficult to achieve the collaborative optimization of environmental parameters and growth trend goals.

[0004] Therefore, the present invention proposes an intelligent and precise strawberry growth management system and method based on the Internet of Things. Summary of the Invention

[0005] The present invention provides an Internet of Things-based intelligent and precise strawberry growth management system and method, including: The sensing layer collects data in real time, providing an accurate and timely information basis for subsequent analysis and control. The growth environment analysis layer determines the deviation data of the growth environment, helping to accurately identify problems and take targeted measures. The growth trend analysis layer can analyze based on the growth trend data and the current target growth stage to comprehensively evaluate the growth status of strawberry seedlings. The growth environment analysis layer and the growth trend analysis layer are linked to combine environmental deviations (such as insufficient CO2 concentration) with growth stages (such as a higher carbon assimilation rate required during the flowering period) to generate a differential supplement amount calculation logic. The growth environment optimization layer analyzes and optimizes the target growth environment control data through the supplement amount to achieve refined management. Based on the current analysis results, the growth environment optimization layer rolls and updates the target environment parameters for the next cycle (such as the temperature and humidity curve by time period), rather than static thresholds. The control layer controls the execution device in real time according to the optimal growth environment control data, ensuring the timely and effective implementation of management measures. And by providing real-time feedback on the latest growth management results, an adaptive iterative optimization chain is formed, ultimately achieving a high-precision match between environmental resource input and strawberry physiological requirements. Through multi-source data fusion in the sensing layer, dynamic coupling of the growth environment and growth trend analysis layers, rolling prediction and regulation in the optimization layer, and closed-loop execution in the control layer, the defects of the existing technology are broken through one by one. The precision and scientificity of strawberry growth management are improved, which helps to increase the yield and quality of strawberries.

[0006] The present invention provides an Internet of Things-based intelligent and precise strawberry growth management system, including:

[0007] A sensing layer for collecting in real time the actual growth environment data and growth trend data of strawberry seedlings;

[0008] A growth environment analysis layer for determining the deviation data between the actual growth environment data and the target growth environment data of strawberry seedlings as the growth environment control analysis result of strawberry seedlings;

[0009] A growth trend analysis layer for analyzing the growth trend analysis result of strawberry seedlings based on the growth trend data of strawberry seedlings and the current target growth stage of strawberry seedlings;

[0010] A growth environment optimization layer for analyzing the supplement amount of the strawberry seedling growth environment data based on the growth environment control analysis result and the growth trend analysis result of strawberry seedlings, and optimizing the target growth environment control data of strawberry seedlings in the next adjacent control cycle based on the supplement amount of the strawberry seedling growth environment data to obtain the optimal growth environment control data of strawberry seedlings in the next adjacent control cycle;

[0011] A control layer for controlling the execution device in real time based on the optimal growth environment control data of strawberry seedlings in the next adjacent control cycle to obtain the latest growth management result of strawberry seedlings.

[0012] Preferably, the perception layer includes:

[0013] A growth environment perception module for real-time collecting actual growth environment data of strawberry seedlings based on multiple sensors distributed at multiple positions in the greenhouse;

[0014] A growth trend perception module for fitting the appearance images of all strawberry seedlings in the greenhouse as the growth trend data of strawberry seedlings based on the monitoring images of multiple perspectives of strawberry seedlings collected in real time.

[0015] Preferably, the growth environment perception module includes:

[0016] A temperature perception sub-module for real-time monitoring the environmental temperature at corresponding positions in the greenhouse based on temperature sensors distributed at different heights and positions in the greenhouse;

[0017] A humidity perception sub-module for real-time monitoring the environmental humidity at corresponding positions in the greenhouse based on humidity sensors distributed at different heights and positions in the greenhouse;

[0018] A carbon dioxide concentration perception sub-module for real-time monitoring the carbon dioxide concentration at corresponding positions in the greenhouse based on carbon dioxide concentration sensors distributed at different heights and positions in the greenhouse;

[0019] Among them, the actual growth environment data of strawberry seedlings includes: environmental temperature, environmental humidity and carbon dioxide concentration at multiple positions in the greenhouse.

[0020] Preferably, the growth trend perception module includes:

[0021] A strawberry seedling image extraction sub-module for extracting the strawberry seedling image area in the monitoring image of each perspective of strawberry seedlings;

[0022] A feature matching sub-module for generating the feature vectors of each pixel point in the strawberry seedling image area of each perspective, and taking the quotient of the dot product of the feature vectors of any two pixel points in the strawberry seedling image areas of every two perspectives and the product of the norms of the feature vectors of the corresponding two pixel points as the feature matching degree between any two pixel points in the strawberry seedling image areas of every two perspectives;

[0023] A pixel matching sub-module for obtaining all matching pixel point combinations in the strawberry seedling image areas of every two perspectives based on the pixel point combinations in the strawberry seedling image areas of every two perspectives where the feature matching degree exceeds the matching degree threshold;

[0024] A perspective fusion sub-module for performing perspective fusion on the strawberry seedling image areas of all perspectives based on all matching pixel point combinations in the strawberry seedling image areas of every two perspectives to obtain the appearance images of all strawberry seedlings in the greenhouse as the growth trend data of strawberry seedlings.

[0025] Preferably, the perspective fusion sub-module includes:

[0026] An image matching unit for summarizing the strawberry seedling image regions of the remaining perspectives with the largest total number of matching pixel points between the strawberry seedling image regions of a single perspective, to obtain at least one image region group;

[0027] A distortion analysis unit for generating distortion data of the strawberry seedling image regions of the corresponding two perspectives based on the relative position vectors between pairwise pixel points belonging to the same strawberry seedling image region among all matching pixel point combinations in the strawberry seedling image regions of every two perspectives in each image region group;

[0028] A distortion correction unit for uniformly correcting the deformation of all strawberry seedling image regions in the corresponding image region group based on the distortion data of the strawberry seedling image regions of every two perspectives in each image region group, to obtain a first corrected image group for each image region group;

[0029] A filtering and fusion unit for filtering out interference pixels and fusing pixels of all the first corrected images in all the first corrected image groups, to obtain the appearance images of all strawberry seedlings in the greenhouse as the growth data of the strawberry seedlings.

[0030] Preferably, the filtering and fusion unit includes:

[0031] An interference filtering sub-unit for filtering out interference pixels of all the first corrected images in each first corrected image group, to obtain a second corrected image group for each first corrected image group;

[0032] A pixel fusion sub-unit for fusing all the matching pixel point combinations in each second corrected image group based on the method of weighted pixel averaging, to obtain multiple strawberry seedling perspective fusion images;

[0033] A cyclic fusion sub-unit for continuously performing perspective fusion on all the strawberry seedling perspective fusion images until the appearance images of all strawberry seedlings in the greenhouse are obtained as the growth data of the strawberry seedlings.

[0034] Preferably, the growth analysis layer includes:

[0035] An actual growth stage analysis module for determining the actual growth stage of the strawberry seedlings based on the growth data of the strawberry seedlings;

[0036] A target growth stage acquisition module for determining the current target growth stage of the strawberry seedlings based on the planting plan of the strawberries;

[0037] A growth analysis module for analyzing the growth analysis result of the strawberry seedlings based on the actual growth stage of the strawberry seedlings and the current target growth stage of the strawberry seedlings.

[0038] Preferably, the actual growth stage analysis module includes:

[0039] The similarity calculation sub-module is used to generate a multi-dimensional feature vector of the strawberry seedling based on the growth data of the strawberry seedling, and take the dot product of the multi-dimensional feature vector of the strawberry seedling and the standard multi-dimensional feature vector of each growth stage, divided by the product of the modulus of the multi-dimensional feature vector of the strawberry seedling and the modulus of the standard multi-dimensional feature vector of the corresponding growth stage, as the similarity between the multi-dimensional feature vector of the strawberry seedling and the standard multi-dimensional feature vector of each growth stage;

[0040] The rough growth stage division sub-module is used to take the growth stage with the maximum similarity as the actual rough growth stage of the strawberry seedling;

[0041] The stage refinement progress determination sub-module is used to take the quotient of the similarity between the multi-dimensional feature vector of the strawberry seedling and the standard multi-dimensional feature vector of the actual growth stage, divided by the sum of the similarity between the multi-dimensional feature vector of the strawberry seedling and the standard multi-dimensional feature vector of the actual growth stage and the similarity between the multi-dimensional feature vector of the strawberry seedling and the standard multi-dimensional feature vector of the adjacent next growth stage of the actual growth stage, as the stage refinement growth progress ratio of the actual rough growth stage of the strawberry seedling, and take the stage refinement growth progress ratio of the actual rough growth stage of the strawberry seedling as the actual growth stage of the strawberry seedling.

[0042] Preferably, the growth environment optimization layer includes:

[0043] The primary supplement amount analysis module is used to determine the primary supplement amount of the strawberry seedling growth environment data based on the growth environment control analysis result of the strawberry seedling;

[0044] The secondary supplement amount analysis module is used to analyze the secondary supplement amount of the strawberry seedling growth environment data based on the growth analysis result of the strawberry seedling;

[0045] The environment control optimization module is used to optimize the target growth environment control data of the strawberry seedling in the adjacent next control period based on the primary supplement amount and the secondary supplement amount of the strawberry seedling growth environment data, and obtain the optimal growth environment control data of the strawberry seedling in the adjacent next control period;

[0046] Among them, the supplement amount of the strawberry seedling growth environment data includes the primary supplement amount and the secondary supplement amount of the strawberry seedling growth environment data.

[0047] The present invention provides an intelligent precision growth management method for strawberries based on the Internet of Things, which is applied to any of the above intelligent precision growth management systems for strawberries based on the Internet of Things, and includes:

[0048] S1: Real-time collect the actual growth environment data and growth data of the strawberry seedling;

[0049] S2: Determine the deviation data between the actual growth environment data and the target growth environment data of the strawberry seedlings as the growth environment control analysis result of the strawberry seedlings;

[0050] S3: Analyze the growth trend analysis result of the strawberry seedlings based on the growth trend data of the strawberry seedlings and the current target growth stage of the strawberry seedlings;

[0051] S4: Analyze the supplement amount of the strawberry seedling growth environment data based on the growth environment control analysis result and the growth trend analysis result of the strawberry seedlings, and optimize the target growth environment control data of the strawberry seedlings in the next adjacent control period based on the supplement amount of the strawberry seedling growth environment data to obtain the optimal growth environment control data of the strawberry seedlings in the next adjacent control period;

[0052] S5: Based on the optimal growth environment control data of the strawberry seedlings in the next adjacent control period, the execution device is controlled in real time to obtain the latest growth management result of the strawberry seedlings.

[0053] The beneficial effects of the present invention compared with the prior art are as follows: The perception layer collects data in real time, providing an accurate and timely information basis for subsequent analysis and control. The growth environment analysis layer determines the deviation data of the growth environment, which helps to accurately discover problems and take targeted measures. The growth trend analysis layer can analyze based on the growth trend data and the current target growth stage to comprehensively evaluate the growth status of the strawberry seedlings. The growth environment analysis layer and the growth trend analysis layer are linked to combine environmental deviations (such as insufficient CO2 concentration) with growth stages (such as the flowering period requiring a higher carbon assimilation rate) to generate a differential supplement amount calculation logic. The growth environment optimization layer optimizes the target growth environment control data by analyzing the supplement amount, realizing refined management. The growth environment optimization layer updates the target environment parameters (such as the temperature and humidity curve in segments) of the next cycle based on the current analysis result, rather than static thresholds. The control layer controls the execution device in real time according to the optimal growth environment control data to ensure the timely and effective implementation of management measures. And by real-time feedback of the latest growth management result, an adaptive iterative optimization chain is formed, ultimately achieving a high-precision match between environmental resource input and the physiological needs of strawberries. Through the multi-source data fusion of the perception layer, the dynamic coupling of the growth environment and the growth trend analysis layer, the rolling prediction and regulation of the optimization layer, and the closed-loop execution of the control layer, the defects of the prior art are broken through one by one. The accuracy and scientificity of strawberry growth management are improved, which helps to improve the yield and quality of strawberries.

[0054] Other features and advantages of the present invention will be described in the following specification, and, in part, will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the structures specifically pointed out in this application document.

[0055] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Description of the Drawings

[0056] The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention, but do not constitute a limitation to the present invention. In the accompanying drawings:

[0057] Figure 1 is a schematic diagram of an intelligent precise growth management system for strawberries based on the Internet of Things in an embodiment of the present invention;

[0058] Figure 2 is a schematic diagram of the connection of various sensors of the growth environment perception module in an embodiment of the present invention;

[0059] Figure 3 is a flowchart of an intelligent precise growth management method for strawberries based on the Internet of Things in an embodiment of the present invention. Detailed Embodiments

[0060] The following describes the preferred embodiments of the present invention with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0061] Embodiment 1:

[0062] The present invention provides an intelligent precise growth management system for strawberries based on the Internet of Things. Referring to Figure 1 , including:

[0063] A perception layer for real-time collection of actual growth environment data and growth trend data of strawberry seedlings;

[0064] A growth environment analysis layer for determining the deviation data between the actual growth environment data and the target growth environment data of strawberry seedlings as the growth environment control analysis result of strawberry seedlings;

[0065] A growth trend analysis layer for analyzing the growth trend analysis result of strawberry seedlings based on the growth trend data of strawberry seedlings and the current target growth stage of strawberry seedlings;

[0066] A growth environment optimization layer for analyzing the supplement amount of the growth environment data of strawberry seedlings based on the growth environment control analysis result and the growth trend analysis result of strawberry seedlings, and optimizing the target growth environment control data of strawberry seedlings in the next adjacent control period based on the supplement amount of the growth environment data of strawberry seedlings to obtain the optimal growth environment control data of strawberry seedlings in the next adjacent control period;

[0067] A control layer for controlling the execution device in real time based on the optimal growth environment control data of strawberry seedlings in the next adjacent control period to obtain the latest growth management result of strawberry seedlings.

[0068] In this embodiment, the target growth environment data are the ideal environmental parameter values preset for strawberry seedlings at a specific growth stage, such as ideal temperature, humidity, carbon dioxide concentration, etc.

[0069] In this embodiment, the deviation data between the actual growth environment data and the target growth environment data of the strawberry seedlings is obtained by comparing the various data of the actual environment where the strawberry seedlings are located (such as actual temperature, humidity, etc.) with the preset ideal environment data, and calculating the difference value between the two.

[0070] In this embodiment, the current target growth stage of the strawberry seedlings refers to the growth and development stage that the current strawberry seedlings should be in according to the planting plan and growth law of strawberries, such as the seedling stage, flowering stage, fruiting stage, etc.

[0071] In this embodiment, the supplementary amount of the growth environment data of the strawberry seedlings analyzed based on the growth environment control analysis result and the growth trend analysis result is to comprehensively consider the deviation of the growth environment and the growth trend of the strawberry seedlings, and calculate the amount value of the environmental parameters that need to be supplemented or adjusted to improve the growth environment of the strawberry seedlings.

[0072] In this embodiment, the target growth environment control data of the strawberry seedlings in the adjacent next control cycle is the target value of the environmental parameter control that is expected to be achieved for the strawberry seedlings in the upcoming next control cycle.

[0073] In this embodiment, the target growth environment control data of the strawberry seedlings in the adjacent next control cycle is optimized based on the supplementary amount of the growth environment data of the strawberry seedlings to obtain the optimal growth environment control data of the strawberry seedlings in the adjacent next control cycle. That is, according to the calculated supplementary amount of the environmental data, the target environmental data of the next control cycle preset is adjusted and improved, so as to obtain the environmental control data most suitable for the growth of the strawberry seedlings. For example, the target environment data adjustment and improvement model trained in advance can be used for optimization, that is, the supplementary amount of the growth environment data of the strawberry seedlings is input into the target environment data adjustment and improvement model trained in advance for the target growth environment control data of the strawberry seedlings in the adjacent next control cycle, and the optimal growth environment control data of the strawberry seedlings in the adjacent next control cycle is obtained. In this way, a closed-loop control of the environmental data is formed.

[0074] In this embodiment, the latest growth management result of the strawberry seedlings is the latest situation about the growth status of the strawberry seedlings obtained after managing the strawberry seedlings according to the optimal growth environment control data, including information on growth status, fruit yield and quality, etc.

[0075] The beneficial effects of the above technologies are as follows: The sensing layer collects data in real time, providing an accurate and timely information basis for subsequent analysis and control. The growth environment analysis layer determines the deviation data of the growth environment, which helps to accurately identify problems and take targeted measures. The growth trend analysis layer can analyze based on the growth trend data and the current target growth stage to comprehensively evaluate the growth status of strawberry seedlings. The growth environment analysis layer and the growth trend analysis layer are linked to combine the environmental deviation (such as insufficient CO2 concentration) with the growth stage (such as a higher carbon assimilation rate required during the flowering period) to generate a differential calculation logic for the supplementary amount. The growth environment optimization layer analyzes the supplementary amount to optimize the target growth environment control data, achieving refined management. Based on the current analysis results, the growth environment optimization layer rolls up and updates the target environment parameters for the next cycle (such as the temperature and humidity curves at different times), rather than static thresholds. The control layer controls the execution device in real time according to the optimal growth environment control data, ensuring the timely and effective implementation of management measures. And by providing real-time feedback on the latest growth management results, an adaptive iterative optimization chain is formed, ultimately achieving a high-precision match between the input of environmental resources and the physiological needs of strawberries. Through the multi-source data fusion of the sensing layer, the dynamic coupling of the growth environment and growth trend analysis layers, the rolling prediction and regulation of the optimization layer, and the closed-loop execution of the control layer, the defects of the existing technologies are broken through one by one. The accuracy and scientific nature of strawberry growth management are improved, which helps to increase the yield and quality of strawberries.

[0076] Embodiment 2:

[0077] Based on Embodiment 1, the sensing layer includes:

[0078] The growth environment sensing module is used to collect the actual growth environment data of strawberry seedlings in real time based on multiple sensors distributed at multiple locations in the greenhouse;

[0079] The growth trend sensing module is used to fit the appearance images of all strawberry seedlings in the greenhouse as the growth trend data of strawberry seedlings based on the monitoring images of multiple perspectives of strawberry seedlings collected in real time.

[0080] In this embodiment, the appearance images of all strawberry seedlings in the greenhouse are images that can comprehensively display the overall shape and growth status of all strawberry seedlings in the greenhouse after a series of processes such as extraction, matching, fusion, correction, and filtering of the monitoring images of multiple perspectives of strawberry seedlings.

[0081] The beneficial effects of the above technical solutions are as follows: The growth environment perception module collects data at multiple positions in the greenhouse through a variety of sensors, making the collected environmental data more comprehensive and accurate, and capable of reflecting the environmental differences in different regions of the greenhouse. The growth trend perception module fits the appearance image of strawberry seedlings based on monitoring images from multiple perspectives as growth trend data. This method can more intuitively and comprehensively understand the growth trend of strawberry seedlings. The comprehensive and accurate growth environment data and intuitive growth trend data provide a reliable basis for subsequent analysis and management. It helps to more accurately discover the problems and needs in strawberry growth, thereby formulating more effective management strategies. It improves the accuracy and reliability of data collection in the perception layer and provides strong support for the intelligent and precise growth management of strawberries.

[0082] Example 3:

[0083] Based on Example 2, the growth environment perception module, referring to Figure 2 , includes:

[0084] The temperature perception sub-module is used to real-time monitor the environmental temperature at the corresponding position in the greenhouse based on temperature sensors distributed at different heights and positions in the greenhouse;

[0085] The humidity perception sub-module is used to real-time monitor the environmental humidity at the corresponding position in the greenhouse based on humidity sensors distributed at different heights and positions in the greenhouse;

[0086] The carbon dioxide concentration perception sub-module is used to real-time monitor the carbon dioxide concentration at the corresponding position in the greenhouse based on carbon dioxide concentration sensors distributed at different heights and positions in the greenhouse;

[0087] Among them, the actual growth environment data of strawberry seedlings includes: the environmental temperature, environmental humidity, and carbon dioxide concentration at multiple positions in the greenhouse.

[0088] The beneficial effects of the above technical solutions are as follows: The temperature perception sub-module can accurately monitor the temperature changes at various places in the greenhouse by setting temperature sensors at different heights and positions, and comprehensively understand the temperature distribution. The humidity perception sub-module also sets humidity sensors at different heights and positions to accurately obtain the humidity differences in the greenhouse, providing a fine basis for humidity control. The carbon dioxide concentration perception sub-module monitors the carbon dioxide concentration at multiple positions, which helps to precisely regulate the gas environment in the greenhouse to meet the growth needs of strawberries. The comprehensive collection of various environmental data provides rich and accurate basic information for subsequent growth environment analysis and optimization. It improves the accuracy and comprehensiveness of the perception of the growth environment in the greenhouse and is conducive to realizing more precise strawberry growth management.

[0089] Example 4:

[0090] Based on Example 2, the growth trend perception module includes:

[0091] The strawberry seedling image extraction sub-module is used to extract the strawberry seedling image area from the monitoring images of each perspective of the strawberry seedlings;

[0092] The feature matching sub-module is used to generate the feature vectors of each pixel point in the strawberry seedling image area of each perspective, and take the quotient of the dot product of the feature vectors of any two pixel points in the strawberry seedling image areas of every two perspectives and the product of the norms of the feature vectors of the corresponding two pixel points as the feature matching degree between any two pixel points in the strawberry seedling image areas of every two perspectives;

[0093] The pixel matching sub-module is used to obtain all the matching pixel point combinations in the strawberry seedling image areas of every two perspectives based on the pixel point combinations in the strawberry seedling image areas of every two perspectives whose feature matching degree exceeds the matching degree threshold;

[0094] The perspective fusion sub-module is used to perform perspective fusion on the strawberry seedling image areas of all perspectives based on all the matching pixel point combinations in the strawberry seedling image areas of every two perspectives, and obtain the appearance images of all the strawberry seedlings in the greenhouse as the growth data of the strawberry seedlings.

[0095] In this embodiment, the strawberry seedling image area is the image range of the specific part containing the strawberry seedlings that is recognized and extracted from the monitoring images of each perspective

[0096] In this embodiment, generating the feature vectors of each pixel point in the strawberry seedling image area of each perspective means generating a set of numerical values that can describe the characteristics of each pixel point in the strawberry seedling image area under each perspective through a specific algorithm or method. The feature vectors of each pixel point can be generated in the following way: for a certain pixel point, consider its color value (such as RGB value), brightness value, position coordinates in the image, etc. Through a specific mathematical formula or algorithm, these information are converted into a set of numerical values. For example, after converting the RGB value into a grayscale value and performing a certain operation combination with the position coordinates, a vector containing multiple numerical values is obtained, which is the feature vector of this pixel point. Another example is to use the convolutional neural network algorithm in deep learning, input the information of the pixel point and the pixel points within a certain range around it into the network, and the network calculates and outputs a set of numerical values as the feature vector of this pixel point to describe its characteristics in the image.

[0097] In this embodiment, obtaining all the matching pixel point combinations in the strawberry seedling image areas of every two perspectives based on the pixel point combinations in the strawberry seedling image areas of every two perspectives whose feature matching degree exceeds the matching degree threshold means comparing the feature matching degrees between the pixel points in the strawberry seedling image areas of every two perspectives, and only selecting the pixel points whose feature matching degree exceeds the pre-set matching degree threshold for combination, so as to obtain the combinations of all the qualified matching pixel points in the image areas of every two perspectives.

[0098] In this embodiment, the matching degree threshold is a preset numerical standard used to determine whether the feature matching degree between pixel points in the strawberry seedling image regions of two perspectives is high enough. Only the pixel point combinations exceeding this threshold will be considered valid matching combinations.

[0099] The beneficial effects of the above technical solutions are as follows: The strawberry seedling image extraction sub-module can accurately extract the strawberry seedling image region from the monitoring image, providing an accurate object for subsequent analysis. The feature matching sub-module determines the feature matching degree between pixel points by calculating the dot product of feature vectors, etc., and the method is scientific and effective. The pixel matching sub-module filters out the matching pixel point combinations according to the feature matching degree, improving the accuracy and reliability of the data. The perspective fusion sub-module fuses the strawberry seedling image regions of multiple perspectives to obtain a comprehensive and accurate strawberry seedling appearance image as the growth trend data. It improves the accuracy and reliability of obtaining the strawberry seedling growth trend data, providing strong support for subsequent growth trend analysis and growth management.

[0100] Embodiment 5:

[0101] Based on Embodiment 4, the perspective fusion sub-module includes:

[0102] An image matching unit for summarizing the strawberry seedling image regions of the remaining perspectives with the largest total number of matching pixel points between the strawberry seedling image regions of a single perspective to obtain at least one image region group;

[0103] A distortion analysis unit for generating distortion data corresponding to the strawberry seedling image regions of two perspectives based on the relative position vectors between pairwise pixel points belonging to the same strawberry seedling image region among all the matching pixel point combinations in the strawberry seedling image regions of every two perspectives in each image region group;

[0104] A distortion correction unit for uniformly correcting the deformation of all the strawberry seedling image regions in the corresponding image region group based on the distortion data of the strawberry seedling image regions of every two perspectives in each image region group to obtain the first corrected image group of each image region group;

[0105] A filtering and fusion unit for filtering out interfering pixels and fusing pixels for all the first corrected images in all the first corrected image groups to obtain the appearance image of all the strawberry seedlings in the greenhouse as the growth trend data of the strawberry seedlings.

[0106] In this embodiment, the matching pixel points refer to the pixel points whose feature matching degree exceeds the matching degree threshold when comparing the strawberry seedling image regions of two perspectives.

[0107] In this embodiment, the strawberry seedling image regions from other perspectives are the strawberry seedling image regions corresponding to perspectives other than the currently processed perspective.

[0108] In this embodiment, the relative position vector between two pixel points is a vector that describes the relative position relationship between any two pixel points in the same strawberry seedling image region. Suppose there are two pixel points A and B in a strawberry seedling image region, the coordinates of A are (x1, y1), and the coordinates of B are (x2, y2). Then the relative position vector between these two pixel points can be expressed as (x2 - x1, y2 - y1). For example, if the coordinates of point A are (3, 5) and the coordinates of point B are (7, 9), then the relative position vector between them is (7 - 3, 9 - 5), that is, (4, 4). This vector (4, 4) describes the relative position relationship between pixel points A and B.

[0109] In this embodiment, the distortion data for the strawberry seedling image regions corresponding to every two perspectives in each image region group is generated based on the relative position vectors between pairwise pixel points that belong to the same strawberry seedling image region among all the matching pixel point combinations in the strawberry seedling image regions of every two perspectives in each image region group. That is, by analyzing the relative position vectors between each pair of pixel points from two perspectives and belonging to the same strawberry seedling image region within the same image region group, relevant data that can reflect the deformation of the strawberry seedling image regions from these two perspectives is calculated.

[0110] In this embodiment, the deformation of all the strawberry seedling image regions in the corresponding image region group is uniformly corrected based on the distortion data of the strawberry seedling image regions corresponding to every two perspectives in each image region group, and the first corrected image group for each image region group is obtained. That is, according to the distortion data of the strawberry seedling image regions corresponding to two perspectives in each image region group, all the strawberry seedling image regions in this region group are corrected, thereby obtaining the first corrected image group for this image region group. For example, correction processing is performed using a pre-trained distortion correction model, that is, the distortion data of the strawberry seedling image regions corresponding to every two perspectives in each image region group and all the strawberry seedling image regions in the corresponding image region group are input into the pre-trained distortion correction model to obtain the first corrected image group for each image region group.

[0111] In this embodiment, the first corrected image group is the set of images obtained after the above correction processing.

[0112] The beneficial effects of the above technical solutions are as follows: The image matching unit provides a reasonable grouping for subsequent processing by summarizing the perspective image area with the largest total number of matching pixel points. The distortion analysis unit generates distortion data based on the relative position vectors of the matching pixel points, which helps to accurately evaluate the distortion of the image. The distortion correction unit uniformly corrects the deformation of the image area, improving the accuracy and consistency of the image. The filtering and fusion unit filters out interference pixels and fuses pixels to obtain a clearer and more accurate appearance image of strawberry seedlings as growth data. The accuracy and effect of perspective fusion are improved, providing high-quality data support for accurately evaluating the growth of strawberry seedlings.

[0113] Embodiment 6:

[0114] Based on Embodiment 5, the filtering and fusion unit includes:

[0115] An interference filtering subunit, configured to filter out interference pixels from all the first corrected images in each first corrected image group to obtain a second corrected image group for each first corrected image group;

[0116] A pixel fusion subunit, configured to fuse all the matching pixel point combinations in each second corrected image group based on pixel weighted averaging to obtain multiple perspective fusion images of strawberry seedlings;

[0117] A cyclic fusion subunit, configured to continue performing perspective fusion on all the perspective fusion images of strawberry seedlings until an appearance image of all the strawberry seedlings in the greenhouse is obtained as the growth data of the strawberry seedlings.

[0118] In this embodiment, filtering out interference pixels from all the first corrected images in each first corrected image group means removing those pixels in the corrected images of each first corrected image group that do not meet the requirements and affect the accuracy of the image.

[0119] In this embodiment, the second corrected image group is the first corrected image group after interference pixel filtering.

[0120] In this embodiment, fusing all the matching pixel point combinations in each second corrected image group based on pixel weighted averaging to obtain multiple perspective fusion images of strawberry seedlings means calculating the average value of the values of the matching pixel points in each second corrected image group according to a certain weight ratio, so as to combine these pixel points together to form multiple strawberry seedling images after perspective fusion from different perspectives.

[0121] In this embodiment, the perspective fusion is continued for all the perspective fusion images of strawberry seedlings until an appearance image of all the strawberry seedlings in the greenhouse is obtained as the growth data of the strawberry seedlings. That is, the multiple fusion images obtained previously are fused again (the principle and method of fusion are the same as the process of obtaining multiple perspective fusion images of strawberry seedlings by fusing strawberry seedling image regions disclosed in the embodiment), until an image that can completely show the overall appearance and growth of all the strawberry seedlings in the greenhouse is obtained, and it is used as the data for judging the growth of strawberry seedlings.

[0122] The beneficial effects of the above technical solutions are as follows: The interference filtering subunit can remove interference pixels, improving the clarity and accuracy of the image. The pixel fusion subunit performs fusion by means of pixel weighted averaging, making the fusion result smoother and more natural. The cyclic fusion subunit further improves the integrity and accuracy of the appearance image of strawberry seedlings through multiple fusions. It can obtain higher-quality and more comprehensive growth data of strawberry seedlings, providing a more reliable basis for subsequent analysis and management. It improves the effects of image filtering and fusion, contributing to more accurately grasping the growth trend of strawberry seedlings.

[0123] Embodiment 7:

[0124] Based on Embodiment 1, the growth trend analysis layer includes:

[0125] An actual growth stage analysis module, used to judge the actual growth stage of strawberry seedlings based on the growth data of strawberry seedlings;

[0126] A target growth stage acquisition module, used to determine the current target growth stage of strawberry seedlings based on the planting plan of strawberries;

[0127] A growth trend analysis module, used to analyze the growth trend analysis result of strawberry seedlings based on the actual growth stage of strawberry seedlings and the current target growth stage of strawberry seedlings.

[0128] In this embodiment, judging the actual growth stage of strawberry seedlings based on the growth data of strawberry seedlings is to clarify the current actual growth period of strawberry seedlings according to the relevant data on the growth status of strawberry seedlings, such as plant height, number of leaves, flower and fruit conditions, etc.

[0129] In this embodiment, determining the current target growth stage of strawberry seedlings based on the planting plan of strawberries is to determine the ideal growth period that strawberry seedlings should reach at the current time point according to the pre-made strawberry planting plan.

[0130] In this embodiment, analyzing the growth trend analysis result of strawberry seedlings based on the actual growth stage of strawberry seedlings and the current target growth stage of strawberry seedlings is to compare and comprehensively consider the actual growth stage of strawberry seedlings and the target growth stage that should be reached in the plan, so as to draw an analysis conclusion on the growth trend of strawberry seedlings.

[0131] The beneficial effects of the above technical solutions are as follows: The actual growth stage analysis module can accurately determine the actual growth stage of strawberry seedlings based on the growth data, providing a basis for subsequent analysis. The target growth stage acquisition module determines the current target growth stage according to the planting plan, making the analysis have clear criteria and directions. The growth trend analysis module comprehensively analyzes the actual growth stage and the target growth stage, and can more comprehensively and objectively evaluate the growth status of strawberry seedlings. It helps to timely detect the deviation between the growth of strawberry seedlings and the expectation, so as to take targeted measures for adjustment and optimization. It improves the accuracy and scientificity of the growth trend analysis of strawberry seedlings, providing strong support for realizing precise growth management.

[0132] Example 8:

[0133] Based on Example 7, the actual growth stage analysis module includes:

[0134] The similarity calculation sub-module is used to generate a multi-dimensional feature vector of strawberry seedlings based on the growth data of strawberry seedlings, and regard the dot product of the multi-dimensional feature vector of strawberry seedlings and the standard multi-dimensional feature vector of each growth stage, divided by the product of the modulus of the multi-dimensional feature vector of strawberry seedlings and the modulus of the standard multi-dimensional feature vector of the corresponding growth stage, as the similarity between the multi-dimensional feature vector of strawberry seedlings and the standard multi-dimensional feature vector of each growth stage;

[0135] The rough growth stage division sub-module is used to regard the growth stage with the largest similarity as the actual rough growth stage of strawberry seedlings;

[0136] The stage refinement progress determination sub-module is used to regard the quotient of the similarity between the multi-dimensional feature vector of strawberry seedlings and the standard multi-dimensional feature vector of the actual growth stage, divided by the sum of the similarity between the multi-dimensional feature vector of strawberry seedlings and the standard multi-dimensional feature vector of the actual growth stage and the similarity between the multi-dimensional feature vector of strawberry seedlings and the standard multi-dimensional feature vector of the adjacent next growth stage of the actual growth stage, as the stage refinement growth progress ratio of the actual rough growth stage of strawberry seedlings, and regard the stage refinement growth progress ratio of the actual rough growth stage of strawberry seedlings as the actual growth stage of strawberry seedlings.

[0137] In this embodiment, generating a multi-dimensional feature vector of strawberry seedlings based on the growth data of strawberry seedlings means constructing a vector that can describe the growth characteristics of strawberry seedlings from multiple aspects according to the data related to the growth trend of strawberry seedlings by using specific algorithms and methods. Suppose the following growth data of strawberry seedlings are obtained: the plant height is 20 cm, the number of leaves is 10, the thickness of the stem is 5 mm, and the number of flowers is 3. We can set an algorithm, for example: multiply the plant height value by 1, the number of leaves by 2, the thickness of the stem by 3, and the number of flowers by 4, and then combine these calculation results. According to the above algorithm, the obtained values are 20, 20, 15, and 12 respectively. Combining them into a vector (20, 20, 15, 12), this is the multi-dimensional feature vector of strawberry seedlings generated based on these growth data, which describes the growth characteristics of strawberry seedlings from multiple aspects such as the overall size of the plant, leaf growth, stem development, and flowering status.

[0138] In this embodiment, the actual rough growth stage is the approximate growth period of strawberry seedlings obtained through preliminary judgment without more refined division.

[0139] In this embodiment, the stage refinement growth progress ratio is a proportional value used to more accurately represent the specific growth degree of strawberry seedlings within the actual rough growth stage.

[0140] The beneficial effects of the above technical solutions are as follows: The similarity calculation sub-module provides a quantitative basis for the judgment of the growth stage by calculating the similarity of the multi-dimensional feature vectors. The rough growth stage division sub-module can initially determine the actual rough growth stage of strawberry seedlings and narrow the analysis scope. The stage refinement progress determination sub-module further determines the stage refinement growth progress ratio, thereby more accurately determining the actual growth stage. It can more accurately analyze the actual growth stage of strawberry seedlings, provide more accurate information for subsequent growth trend analysis and management strategy formulation. It improves the accuracy and reliability of the actual growth stage analysis and helps to achieve more refined strawberry planting management.

[0141] Embodiment 9:

[0142] On the basis of Embodiment 1, the growth environment optimization layer includes:

[0143] The primary supplement amount analysis module is used to determine the primary supplement amount of the growth environment data of strawberry seedlings based on the analysis result of the growth environment control of strawberry seedlings;

[0144] The secondary supplement amount analysis module is used to analyze the secondary supplement amount of the growth environment data of strawberry seedlings based on the analysis result of the growth trend of strawberry seedlings;

[0145] An environmental control optimization module is used to optimize the target growth environment control data of strawberry seedlings in the next adjacent control period based on the primary supplement amount and secondary supplement amount of strawberry seedling growth environment data, so as to obtain the optimal growth environment control data of strawberry seedlings in the next adjacent control period;

[0146] Among them, the supplement amount of strawberry seedling growth environment data includes the primary supplement amount and secondary supplement amount of strawberry seedling growth environment data.

[0147] In this embodiment, the primary supplement amount of strawberry seedling growth environment data determined based on the growth environment control analysis result of strawberry seedlings is the amount of environmental data that needs to be supplemented initially calculated according to the deviation analysis result between the actual situation and the target situation of the strawberry seedling growth environment. Assume that the ideal target environment for strawberry seedling growth is a temperature of 25°C, a humidity of 60%, and a carbon dioxide concentration of 800 ppm. The current growth environment data actually monitored by various sensors is: temperature 20°C, humidity 50%, and carbon dioxide concentration 600 ppm. First, calculate the temperature deviation: target temperature 25°C - actual temperature 20°C = 5°C. Assume that according to experience and relevant algorithms, the primary supplement amount of temperature is initially determined to be 120% of the deviation, that is, 5°C × 120% = 6°C. For humidity, the deviation is 60% - 50% = 10%. Similarly, assume that the primary supplement amount is 120% of the deviation, that is, 10% × 120% = 12%. For carbon dioxide concentration, the deviation is 800 ppm - 600 ppm = 200 ppm. Assume that the primary supplement amount is 120% of the deviation, that is, 200 ppm × 120% = 240 ppm. In this way, the primary supplement amounts of strawberry seedling growth environment data are initially determined, which are 6°C for temperature supplement, 12% for humidity supplement, and 240 ppm for carbon dioxide concentration supplement.

[0148] In this embodiment, the primary supplement amount of strawberry seedling growth environment data refers to the preliminary numerical value of the growth environment data supplement obtained through the above method.

[0149] In this embodiment, the secondary supplement amount of the strawberry seedling growth environment data analyzed based on the growth trend analysis result of the strawberry seedlings is the supplement amount of the growth environment data further calculated according to the analysis result of the growth trend of the strawberry seedlings. Suppose we have analyzed the growth trend of the strawberry seedlings as follows: the leaves of the strawberry seedlings are light green, the growth rate of the plants is slow, and the number of flowers is small. Through professional botanical analysis, it is judged that it may be due to insufficient light intensity and soil fertility. First, for the light intensity, the originally set target value is 8 hours of sufficient light per day, but the actual monitored average daily light is only 6 hours. According to research, for every 1 hour of light shortage, 1.2 hours of supplementary light may need to be added in the next growth stage to improve the growth trend. Then the secondary supplement amount of the light intensity is (8 - 6)×1.2 = 2.4 hours. Second, for the soil fertility, it is found through soil testing that the contents of main nutrients such as nitrogen, phosphorus, and potassium are lower than the ideal values. Suppose the ideal content of nitrogen element is 150 ppm, and the actual detection is 100 ppm. For every 10 ppm of nitrogen element shortage, 15 ppm needs to be supplemented in the follow-up to promote growth. Then the secondary supplement amount of the nitrogen element is (150 - 100)÷10×15 = 75 ppm. Using the same method, the secondary supplement amounts of other nutrients such as phosphorus and potassium and other environmental factors that may be involved (such as ventilation conditions, etc.) can be calculated. Combining these calculation results, the secondary supplement amount of the growth environment data based on the growth trend analysis of the strawberry seedlings is obtained, such as supplementing 1 hour of light intensity and 25 ppm of nitrogen element, etc., to more accurately adjust the growth environment of the strawberry seedlings and promote their better growth.

[0150] In this embodiment, the secondary supplement amount of the strawberry seedling growth environment data refers to the additional supplementary value of the growth environment data obtained based on the growth trend analysis.

[0151] In this embodiment, the target growth environment control data of strawberry seedlings in the next adjacent control period is optimized based on the primary supplement amount and secondary supplement amount of the strawberry seedling growth environment data, and the optimal growth environment control data of strawberry seedlings in the next adjacent control period is obtained by comprehensively considering the previously calculated primary supplement amount and secondary supplement amount, adjusting and improving the target growth environment control data of strawberry seedlings in the next control period, so as to obtain the environment control data most conducive to the growth of strawberry seedlings. For example, the primary supplement amount of the temperature in the current growth environment of strawberry seedlings is +3°C, and the primary supplement amount of the humidity is +10%; the secondary supplement amount of the temperature is +2°C, and the secondary supplement amount of the humidity is +5%. In the next control period, the originally set temperature target value is 22°C, and the humidity target value is 50%. Considering the primary and secondary supplement amounts comprehensively, the optimized value of the temperature is 22°C + 3°C + 2°C = 27°C, and the optimized value of the humidity is 50% + 10% + 5% = 65%. In this way, the optimal growth environment control data of strawberry seedlings in the next control period is obtained, that is, the temperature is 27°C and the humidity is 65%, which is more conducive to the growth of strawberry seedlings.

[0152] The beneficial effects of the above technical solutions are as follows: The primary supplement amount analysis module determines the primary supplement amount according to the growth environment control analysis result, providing data support for preliminary optimization. The secondary supplement amount analysis module determines the secondary supplement amount based on the growth trend analysis result, further supplementing the optimization basis from the perspective of growth trend. The environment control optimization module comprehensively optimizes the primary supplement amount and the secondary supplement amount, making the optimization result more comprehensive and accurate. It can optimize the growth environment control data of strawberry seedlings more scientifically and reasonably to meet the growth needs of strawberry seedlings. It improves the accuracy and effectiveness of growth environment optimization and helps to create more suitable growth conditions for strawberry seedlings.

[0153] Embodiment 10:

[0154] The present invention provides an intelligent precise growth management method for strawberries based on the Internet of Things, which is applied to any one of the intelligent precise growth management systems for strawberries based on the Internet of Things in Embodiments 1 to 9, with reference to Figure 3 , including:

[0155] S1: Real-time collect the actual growth environment data and growth trend data of strawberry seedlings;

[0156] S2: Determine the deviation data between the actual growth environment data and the target growth environment data of strawberry seedlings as the growth environment control analysis result of strawberry seedlings;

[0157] S3: Analyze the growth trend analysis result of strawberry seedlings based on the growth trend data of strawberry seedlings and the current target growth stage of strawberry seedlings;

[0158] S4: Analyze the supplement amount of the strawberry seedling growth environment data based on the growth environment control analysis result and the growth trend analysis result of the strawberry seedlings, and optimize the target growth environment control data of the strawberry seedlings in the next adjacent control period based on the supplement amount of the strawberry seedling growth environment data to obtain the optimal growth environment control data of the strawberry seedlings in the next adjacent control period;

[0159] S5: Based on the optimal growth environment control data of the strawberry seedlings in the next adjacent control period, the execution device is controlled in real time to obtain the latest growth management result of the strawberry seedlings.

[0160] The beneficial effects of the above technology are as follows: In step S1, data is collected in real time, providing an accurate and timely information basis for subsequent analysis and control. In step S2, the deviation data of the growth environment is determined, which helps to accurately discover problems and take targeted measures. In step S3, analysis can be carried out according to the growth trend data and the current target growth stage to comprehensively evaluate the growth status of the strawberry seedlings. Steps S2 and S3 are linked to combine the environmental deviation (such as insufficient CO2 concentration) with the growth stage (such as a higher carbon assimilation rate required during the flowering period) to generate a differential supplement amount calculation logic. In step S4, the target growth environment control data is optimized by analyzing the supplement amount, realizing refined management. And based on the current analysis result, the target environment parameters in the next cycle (such as the segmented temperature and humidity curve) are updated in a rolling manner, rather than static thresholds. In step S5, the execution device is controlled in real time according to the optimal growth environment control data to ensure the timely and effective implementation of the management measures. And by providing real-time feedback on the latest growth management result, an adaptive iterative optimization chain is formed, ultimately achieving a high-precision match between the input of environmental resources and the physiological needs of strawberries. Through the multi-source data fusion in step S1, the dynamic coupling of steps S2 and S3, the rolling prediction and regulation in step S4, and the closed-loop execution in step S5, the defects of the existing technology are broken through one by one. The accuracy and scientificity of strawberry growth management are improved, which helps to increase the yield and quality of strawberries.

[0161] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these changes and modifications.

Claims

1. The strawberry intelligent precise growth management system based on the Internet of Things is characterized in that Including: The perception layer is used to collect the actual growth environment data and growth trend data of strawberry seedlings in real time; The growth environment analysis layer is used to determine the deviation data between the actual growth environment data and the target growth environment data of strawberry seedlings as the growth environment control analysis result of strawberry seedlings; The growth trend analysis layer is used to analyze the growth trend analysis result of strawberry seedlings based on the growth trend data of strawberry seedlings and the current target growth stage of strawberry seedlings; The growth environment optimization layer is used to analyze the supplement amount of strawberry seedling growth environment data based on the growth environment control analysis result and growth trend analysis result of strawberry seedlings, and optimize the target growth environment control data of strawberry seedlings in the next adjacent control period based on the supplement amount of strawberry seedling growth environment data to obtain the optimal growth environment control data of strawberry seedlings in the next adjacent control period; The control layer is used to control the execution device in real time based on the optimal growth environment control data of strawberry seedlings in the next adjacent control period to obtain the latest growth management result of strawberry seedlings.

2. The strawberry intelligent precise growth management system based on the Internet of Things according to claim 1, wherein The perception layer includes: The growth environment perception module is used to collect the actual growth environment data of strawberry seedlings in real time based on a variety of sensors distributed at multiple locations in the greenhouse; The growth trend perception module is used to fit the appearance images of all strawberry seedlings in the greenhouse as the growth trend data of strawberry seedlings based on the monitoring images of multiple perspectives of strawberry seedlings collected in real time.

3. The strawberry intelligent precise growth management system based on the Internet of Things according to claim 2, characterized in that, The growth environment perception module includes: The temperature perception sub-module is used to monitor the environmental temperature at the corresponding location in the greenhouse in real time based on temperature sensors distributed at different heights and locations in the greenhouse; The humidity perception sub-module is used to monitor the environmental humidity at the corresponding location in the greenhouse in real time based on humidity sensors distributed at different heights and locations in the greenhouse; The carbon dioxide concentration perception sub-module is used to monitor the carbon dioxide concentration at the corresponding location in the greenhouse in real time based on carbon dioxide concentration sensors distributed at different heights and locations in the greenhouse; Among them, the actual growth environment data of strawberry seedlings includes: environmental temperature, environmental humidity and carbon dioxide concentration at multiple locations in the greenhouse.

4. The strawberry intelligent precise growth management system based on the Internet of Things according to claim 2, characterized in that, The growth trend perception module includes: The strawberry seedling image extraction sub-module is used to extract the strawberry seedling image area in the monitoring image of each perspective of strawberry seedlings; The feature matching sub-module is used to generate the feature vector of each pixel point in the strawberry seedling image area of each perspective, and regard the quotient of the dot product of the feature vectors of any two pixel points in the strawberry seedling image areas of every two perspectives and the product of the norms of the feature vectors of the corresponding two pixel points as the feature matching degree between any two pixel points in the strawberry seedling image areas of every two perspectives; The pixel matching sub-module is used to obtain all matching pixel point combinations in the strawberry seedling image areas of every two perspectives based on the pixel point combinations in the strawberry seedling image areas of every two perspectives where the feature matching degree exceeds the matching degree threshold; The perspective fusion sub-module is used to perform perspective fusion on the strawberry seedling image areas of all perspectives based on all matching pixel point combinations in the strawberry seedling image areas of every two perspectives to obtain the appearance images of all strawberry seedlings in the greenhouse as the growth trend data of strawberry seedlings.

5. The strawberry intelligent precise growth management system based on the Internet of Things according to claim 4, characterized in that, The perspective fusion sub-module includes: An image matching unit, configured to summarize the strawberry seedling image regions of the remaining viewpoints with the largest total number of matching pixel points between the strawberry seedling image regions of a single viewpoint, to obtain at least one image region group; A distortion analysis unit, configured to generate distortion data for the strawberry seedling image regions of the corresponding two viewpoints based on the relative position vectors between pairwise pixel points belonging to the same strawberry seedling image region among all matching pixel point combinations in the strawberry seedling image regions of every two viewpoints in each image region group; A distortion correction unit, configured to perform deformation unified correction on all strawberry seedling image regions in the corresponding image region group based on the distortion data of the strawberry seedling image regions of every two viewpoints in each image region group, to obtain a first corrected image group for each image region group; A filtering and fusion unit, configured to filter out interference pixels and fuse pixels for all the first corrected images in all the first corrected image groups, to obtain the appearance images of all the strawberry seedlings in the greenhouse as the growth data of the strawberry seedlings.

6. The strawberry intelligent precise growth management system based on the Internet of Things according to claim 5, characterized in that, The filtering and fusion unit includes: An interference filtering sub-unit, configured to filter out interference pixels for all the first corrected images in each first corrected image group, to obtain a second corrected image group for each first corrected image group; A pixel fusion sub-unit, configured to fuse all the matching pixel point combinations in each second corrected image group based on the method of weighted pixel averaging, to obtain multiple strawberry seedling viewpoint fusion images; A cyclic fusion sub-unit, configured to continue to perform viewpoint fusion on all the strawberry seedling viewpoint fusion images until the appearance images of all the strawberry seedlings in the greenhouse are obtained as the growth data of the strawberry seedlings.

7. The strawberry intelligent precision growth management system based on the Internet of Things according to claim 1, characterized in that, The growth analysis layer includes: An actual growth stage analysis module, configured to determine the actual growth stage of the strawberry seedlings based on the growth data of the strawberry seedlings; A target growth stage acquisition module, configured to determine the current target growth stage of the strawberry seedlings based on the planting plan of the strawberries; A growth analysis module, configured to analyze the growth analysis result of the strawberry seedlings based on the actual growth stage of the strawberry seedlings and the current target growth stage of the strawberry seedlings.

8. The strawberry intelligent precise growth management system based on the Internet of Things according to claim 7, characterized in that, The actual growth stage analysis module includes: A similarity calculation sub-module, configured to generate a multi-dimensional feature vector of the strawberry seedlings based on the growth data of the strawberry seedlings, and regard the quotient of the dot product of the multi-dimensional feature vector of the strawberry seedlings and the standard multi-dimensional feature vector of each growth stage, and the product of the norm of the multi-dimensional feature vector of the strawberry seedlings and the norm of the standard multi-dimensional feature vector of the corresponding growth stage, as the similarity between the multi-dimensional feature vector of the strawberry seedlings and the standard multi-dimensional feature vector of each growth stage; A rough growth stage division sub-module, configured to regard the growth stage with the largest similarity as the actual rough growth stage of the strawberry seedlings; A stage refinement progress determination sub-module, which is used to take the similarity between the multi-dimensional feature vector of the strawberry seedling and the standard multi-dimensional feature vector of the actual growth stage, and the quotient of the sum of the similarity between the multi-dimensional feature vector of the strawberry seedling and the standard multi-dimensional feature vector of the actual growth stage and the similarity between the multi-dimensional feature vector of the strawberry seedling and the standard multi-dimensional feature vector of the adjacent next growth stage of the actual growth stage as the stage refinement growth progress ratio of the actual rough growth stage of the strawberry seedling, and take the stage refinement growth progress ratio of the actual rough growth stage of the strawberry seedling as the actual growth stage of the strawberry seedling.

9. The strawberry intelligent precise growth management system based on the Internet of Things according to claim 1, characterized in that, The growth environment optimization layer includes: A primary supplement amount analysis module, which is used to determine the primary supplement amount of the strawberry seedling growth environment data based on the growth environment control analysis result of the strawberry seedling; A secondary supplement amount analysis module, which is used to analyze the secondary supplement amount of the strawberry seedling growth environment data based on the growth trend analysis result of the strawberry seedling; An environment control optimization module, which is used to optimize the target growth environment control data of the strawberry seedling in the adjacent next control period based on the primary supplement amount and the secondary supplement amount of the strawberry seedling growth environment data, and obtain the best growth environment control data of the strawberry seedling in the adjacent next control period; Among them, the supplement amount of the strawberry seedling growth environment data includes the primary supplement amount and the secondary supplement amount of the strawberry seedling growth environment data.

10. The intelligent and precise growth management method for strawberries based on the Internet of Things is characterized in that Applied to the Internet of Things-based intelligent precision growth management system for strawberries described in any one of claims 1 to 9, it includes: S1: Real-time collect the actual growth environment data and growth trend data of the strawberry seedling; S2: Determine the deviation data between the actual growth environment data of the strawberry seedling and the target growth environment data as the growth environment control analysis result of the strawberry seedling; S3: Analyze the growth trend analysis result of the strawberry seedling based on the growth trend data of the strawberry seedling and the current target growth stage of the strawberry seedling; S4: Analyze the supplement amount of the strawberry seedling growth environment data based on the growth environment control analysis result and the growth trend analysis result of the strawberry seedling, and optimize the target growth environment control data of the strawberry seedling in the adjacent next control period based on the supplement amount of the strawberry seedling growth environment data, and obtain the best growth environment control data of the strawberry seedling in the adjacent next control period; S5: Based on the best growth environment control data of the strawberry seedling in the adjacent next control period, real-time control the execution device to obtain the latest growth management result of the strawberry seedling.

Images