Dynamic regulation and control method, system and equipment for plant growth environment and storage medium

Abstract

The invention relates to the technical field of plant cultivation, and discloses a dynamic regulation and control method, system and device for a plant growth environment and a storage medium. Inputting the image data into a growth recognition model to obtain a growth stage and a growth state of the plant; according to the growth stage, obtaining a soil temperature and humidity optimal value of the plant growth environment by adopting a response surface method, and inputting the environment data into the temperature and humidity prediction model to obtain a soil temperature and humidity prediction value; and comparing the predicted soil temperature and humidity value with the optimal soil temperature and humidity value, and regulating and controlling the soil temperature and humidity of the plant growth environment according to a comparison result. According to the method, through refined environment management, temperature and humidity, illumination and pest control required by plant growth are dynamically regulated and controlled, global management and accurate regulation and control of the cultivation process are achieved, the intensification and intelligence level of the whole cultivation method is improved, and cost reduction and efficiency improvement of plant cultivation are achieved.

Description

Technical Field

[0001] The present invention relates to the field of plant cultivation technology, and in particular to a method, system, equipment and storage medium for dynamically regulating a plant growth environment. Background Art

[0002] Lucky Red is a plant with high ornamental value. Its unique red leaves symbolize auspiciousness and prosperity, so it is deeply loved by consumers and has broad prospects in the Chinese horticultural market. However, Lucky Red has strict requirements on the growth environment and is very sensitive to small environmental changes. Slight fluctuations in light, soil, temperature and humidity can have a significant impact on its growth. Therefore, adopting a scientific, data-driven cultivation management model is the key to improving the growth quality of Lucky Red and promoting its industrial development.

[0003] At present, conventional methods for cultivating Jilihong include greenhouse cultivation, greenhouse cultivation, and small-scale scattered cultivation. Although this method can provide basic environmental protection for Jilihong cultivation and reduce the impact of climate on its growth, it still has obvious shortcomings. Although greenhouse cultivation provides certain environmental protection, the regulation relies on manual labor, lacks automation and precision, and is difficult to cope with complex environmental changes; greenhouse cultivation cannot monitor and adjust key environmental factors such as temperature and humidity in real time due to simple equipment; small-scale scattered cultivation is limited by technology and scale, and it is difficult to maintain stable growth conditions. In addition, the current cultivation process has a low degree of intensiveness and intelligence, and it is impossible to identify and classify the various growth periods of Jilihong, and extract key environmental factors for different stages for regulation. Due to the inability to accurately judge the temperature, humidity, light and other factors required for each growth period, improper regulation will lead to slow growth of Jilihong, dull leaf color, and decreased immunity, thereby affecting its quality and market competitiveness. At the same time, the existing cultivation methods also lack the ability to predict changes in environmental factors. For example, there is no comprehensive regulation of atmospheric temperature and humidity and indoor soil temperature and humidity, and there is a lack of targeted supplementary lighting and shading measures. It is impossible to effectively adjust under different light requirements, resulting in too much light or too little light. In addition, there is a lack of identification methods for pests and diseases, making it difficult to make timely judgments and guide reasonable pesticide application. These factors have further aggravated the management difficulties in the cultivation of Jilihong, seriously affecting its growth and economic benefits. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a method, system, device and storage medium for dynamically controlling a plant growth environment, which can achieve the effect of global management and precise control of the plant growth environment by identifying and systematically analyzing key factors of the plant growth environment.

[0005] In a first aspect, the present invention provides a method for dynamically controlling a plant growth environment, the method comprising:

[0006] Acquire image data and environmental data of plant growth, wherein the image data includes plant image data, and the environmental data includes soil temperature and humidity data and meteorological data;

[0007] Inputting the image data into a growth recognition model to obtain the growth stage and growth state of the plant, the growth recognition model comprising a plant recognition module, a stage recognition module and a state recognition module;

[0008] According to the growth stage and the growth state, spraying the plants with drugs and / or supplementing light;

[0009] According to the growth stage, the response surface method is used to obtain the optimal value of soil temperature and humidity in the plant growth environment, and the environmental data is input into the temperature and humidity prediction model to obtain the predicted value of soil temperature and humidity;

[0010] The predicted soil temperature and humidity value is compared with the optimal soil temperature and humidity value, and the soil temperature and humidity of the plant growth environment are regulated based on the comparison result.

[0011] Furthermore, the step of inputting the image data into a growth recognition model to obtain the growth stage and growth state of the plant includes:

[0012] Inputting the image data into a plant recognition module constructed based on the YOLO model to recognize plant parts and obtain plant part images;

[0013] Inputting the plant part image into a stage recognition module constructed based on a convolutional neural network model to obtain the growth stage of the plant, wherein the growth stage includes a budding stage, a growing stage, and a mature stage;

[0014] The plant part image is input into a state recognition module for feature extraction and feature comparison to obtain the growth state of the plant, which includes the pest state and the disease state.

[0015] Furthermore, the step of inputting the plant part image into a state recognition module for feature extraction and feature comparison to obtain the growth state of the plant includes:

[0016] Using color space conversion and color segmentation technology to extract color features from the plant part image to obtain color feature data;

[0017] Using a texture statistics method to extract texture features from the plant part image to obtain texture feature data;

[0018] Using edge detection technology and contour extraction technology to extract shape features from the plant part image to obtain shape feature data;

[0019] The color feature data, texture feature data and shape feature data are compared with the feature data in the pest and disease database, and the growth status of the plant is obtained based on the comparison result.

[0020] Furthermore, the step of obtaining the optimal soil temperature and humidity of the plant growth environment by using the response surface method according to the growth stage includes:

[0021] A response surface model was established with soil temperature and soil moisture as independent variables and the change rate of plant leaf color area as the dependent variable.

[0022] Solving the response surface model to obtain the soil temperature and humidity combination value that makes the change rate of plant leaf color area reach the optimal value;

[0023] According to the soil temperature and humidity combination values ​​corresponding to different growth stages, an environmental parameter database is constructed;

[0024] According to the growth stage, the corresponding optimal value of soil temperature and humidity is selected from the environmental parameter database.

[0025] Furthermore, the step of extracting texture features from the plant part image using a texture statistics method to obtain texture feature data comprises:

[0026] Converting the plant part image into a grayscale image, calculating a grayscale co-occurrence matrix of the grayscale image, and extracting first texture feature data from the grayscale co-occurrence matrix;

[0027] A neighborhood region is defined for each pixel in the grayscale image, and a local binary pattern of the pixel is calculated in the neighborhood region to obtain a histogram of the local binary pattern, and second texture feature data is calculated according to the histogram;

[0028] The first texture feature data and the second texture feature data are merged to obtain texture feature data.

[0029] Furthermore, the step of extracting shape features from the plant part image using edge detection technology and contour extraction technology to obtain shape feature data includes:

[0030] Using edge detection technology to extract edge information of the plant part image;

[0031] According to each edge point in the edge information, a neighborhood search strategy is adopted to perform a neighborhood search, and a connection strategy is adopted to connect each searched adjacent edge point to obtain edge connection information;

[0032] According to the edge connection information, contour extraction is performed, and the extracted contour is analyzed to obtain shape feature data.

[0033] Furthermore, the step of obtaining the optimal soil temperature and humidity of the plant growth environment by using the response surface method according to the growth stage includes:

[0034] A response surface model was established with soil temperature and soil moisture as independent variables and the change rate of plant leaf color area as the dependent variable.

[0035] Solving the response surface model to obtain the soil temperature and humidity combination value that makes the change rate of plant leaf color area reach the optimal value;

[0036] According to the soil temperature and humidity combination values ​​corresponding to different growth stages, an environmental parameter database is constructed;

[0037] According to the growth stage, the corresponding optimal value of soil temperature and humidity is selected from the environmental parameter database.

[0038] Furthermore, the temperature and humidity prediction model is constructed based on the Transformer-XL model and trained using the Grey Wolf Optimization Algorithm.

[0039] In a second aspect, the present invention provides a dynamic control system for a plant growth environment, the system comprising:

[0040] A data acquisition module, used to acquire image data and environmental data of plant growth, wherein the image data includes plant image data, and the environmental data includes soil temperature and humidity data and meteorological data;

[0041] A growth recognition module, used for inputting the image data into a growth recognition model to obtain the growth stage and growth state of the plant, wherein the growth recognition model includes a plant recognition module, a stage recognition module and a state recognition module;

[0042] A first control module is used to spray the plant with medicine and / or supplement light according to the growth stage and the growth state;

[0043] A temperature and humidity prediction module is used to obtain the optimal soil temperature and humidity of the plant growth environment according to the growth stage by using the response surface method, and input the environmental data into the temperature and humidity prediction model to obtain the predicted value of soil temperature and humidity;

[0044] The second control module is used to compare the predicted soil temperature and humidity value with the optimal soil temperature and humidity value, and to control the soil temperature and humidity of the plant growth environment according to the comparison result.

[0045] In a third aspect, an embodiment of the present invention further provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0046] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of the above method when executed by a processor.

[0047] The present invention provides a method, system, computer device and storage medium for dynamically regulating a plant growth environment. The present invention can optimize the cultivation strategy by establishing a dynamic adjustment mechanism to identify and distinguish different growth periods and growth states of plants, and can improve the ability to cope with complex coupled changes by strengthening the long-term dynamic prediction capability of key factors such as soil temperature and humidity, thereby meeting the regulation requirements of plant cultivation in a multivariate environment. The present invention dynamically regulates the temperature, humidity, light and pest control required for plant growth through refined environmental management, and can achieve global management and precise regulation of the cultivation process, improve the intensiveness and intelligence level of the overall cultivation method, and achieve cost reduction and efficiency improvement of plant cultivation. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is a schematic flow chart of a method for dynamically controlling a plant growth environment in an embodiment of the present invention;

[0049] Figure 2 is another schematic flow chart of the method for dynamically controlling the plant growth environment in an embodiment of the present invention;

[0050] Figure 3 is a schematic structural diagram of a dynamic control system for plant growth environment in an embodiment of the present invention;

[0051] Figure 4 It is a diagram of the internal structure of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] See also Figure 1 The first embodiment of the present invention provides a method for dynamically controlling a plant growth environment, which includes steps S10 to S50:

[0054] Step S10, acquiring image data and environmental data of plant growth, wherein the image data includes plant image data, and the environmental data includes soil temperature and humidity data and meteorological data;

[0055] Step S20, inputting the image data into a growth recognition model to obtain the growth stage and growth state of the plant, wherein the growth recognition model includes a plant recognition module, a stage recognition module and a state recognition module;

[0056] Step S30, spraying the plants with drugs and / or supplementing light according to the growth stage and the growth state;

[0057] Step S40, according to the growth stage, using the response surface method to obtain the optimal value of soil temperature and humidity of the plant growth environment, inputting the environmental data into a temperature and humidity prediction model to obtain a predicted value of soil temperature and humidity;

[0058] Step S50, comparing the predicted soil temperature and humidity value with the optimal soil temperature and humidity value, and regulating the soil temperature and humidity of the plant growth environment according to the comparison result.

[0059] The method for dynamically regulating the plant growth environment provided by the present invention can be used for cultivating various types of plants. The specific regulation steps of the present invention are described in detail below using Jilihong as an example.

[0060] First, the data used in the present invention are collected and preprocessed. The data used include image data of plants and parameter data of the growth environment. Specifically, image data of the growth process of plants are collected by camera equipment, soil temperature data and soil moisture data are collected by sensors, and meteorological data are obtained through weather websites. For the received image data and environmental data, data preprocessing is performed, including specifically including eliminating abnormal data, interpolating missing values, and unifying all data into the same time frequency. The purpose of data preprocessing is to ensure the accuracy and consistency of model input data and provide reliable input for the prediction of subsequent models. In order to further optimize the efficiency and performance of the model, principal component analysis (PCA) is introduced after data preprocessing for feature screening to extract the most important features in the data. The specific steps of preprocessing refer to conventional steps and will not be repeated here.

[0061] See also Figure 2 The processed image data and environmental data are input into different models for analysis. The image data is input into a pre-established growth recognition model to identify the growth stage and growth state of the plant, and the environmental data is input into a pre-established temperature and humidity prediction model to predict the soil temperature and humidity. Specifically, the steps of identifying the growth stage and growth state of the plant through the growth recognition model include:

[0062] Inputting the image data into a plant recognition module constructed based on the YOLO model to recognize plant parts and obtain plant part images;

[0063] Inputting the plant part image into a stage recognition module constructed based on a convolutional neural network model to obtain the growth stage of the plant, wherein the growth stage includes a budding stage, a growing stage, and a mature stage;

[0064] The plant part image is input into a state recognition module for feature extraction and feature comparison to obtain the growth state of the plant, which includes the pest state and the disease state.

[0065] In this embodiment, the growth recognition model consists of three modules, namely, a plant recognition module, a stage recognition module and a state recognition module. Among them, the plant recognition module is constructed using the YOLO model, which is used to identify various parts of the plant, including roots, stems, leaves and other parts. The stage recognition module is constructed using a convolutional neural network model, which is used to perform in-depth analysis of the images of various parts recognized by the plant recognition module to identify the specific stage of the growth cycle of the plant. The growth stages here include the germination stage, the growth stage and the maturity stage. It should be noted that as long as the neural network model can achieve the above functions, it can be used to construct the growth recognition model of this embodiment, and no excessive restrictions are made here.

[0066] The state recognition module in this embodiment uses color, texture and shape analysis technology to determine the current growth state of the plant by extracting and analyzing features of the images of various parts recognized by the plant recognition module. The specific steps include:

[0067] Using color space conversion and color segmentation technology to extract color features from the plant part image to obtain color feature data;

[0068] Using a texture statistics method to extract texture features from the plant part image to obtain texture feature data;

[0069] Using edge detection technology and contour extraction technology to extract shape features from the plant part image to obtain shape feature data;

[0070] The color feature data, texture feature data and shape feature data are compared with the feature data in the pest and disease database, and the growth status of the plant is obtained based on the comparison result.

[0071] In this embodiment, the features of the plant part image are extracted and analyzed from three aspects: color, texture, and shape. For the color feature, color space conversion and color segmentation technology are used for feature extraction. First, the plant part image is converted from the RGB color space to other color spaces that are more suitable for color feature extraction, such as the HSI color space. The purpose of the conversion is to separate color information and brightness information. In the converted color space, color components or combinations that are useful for color feature extraction of the plant image are selected. In the HSI color space, the H component and S component used to describe the color are extracted. Then, the color segmentation threshold is set according to the extracted color component, and the color segmentation algorithm is used to segment the image into regions of different colors. Finally, color features are extracted from the segmented color regions.

[0072] For texture features, texture statistics method is used for feature extraction. The specific steps include:

[0073] Converting the plant part image into a grayscale image, calculating a grayscale co-occurrence matrix of the grayscale image, and extracting first texture feature data from the grayscale co-occurrence matrix;

[0074] A neighborhood region is defined for each pixel in the grayscale image, and a local binary pattern of the pixel is calculated in the neighborhood region to obtain a histogram of the local binary pattern, and second texture feature data is calculated according to the histogram;

[0075] The first texture feature data and the second texture feature data are merged to obtain texture feature data.

[0076] In this embodiment, the plant part image is first converted from a color image to a grayscale image, and then the grayscale co-occurrence matrix is ​​calculated by counting the frequency of occurrence of grayscale value pairs between each pixel and its neighboring pixels. In order to eliminate the difference in image size and grayscale level, the grayscale co-occurrence matrix is ​​normalized, and then the first texture feature, such as energy, contrast, correlation, entropy, etc., is extracted from the normalized grayscale co-occurrence matrix. The first texture feature reflects the texture roughness, directionality, and grayscale distribution of different regions in the image. For each pixel on the grayscale image, its neighborhood area is defined, and the local binary pattern is calculated by comparing the grayscale values ​​of the pixel with its neighboring pixels. For each pixel in the image, the histogram of its local binary pattern is counted, and the second texture feature, such as histogram mean, variance, energy, contrast, etc., is calculated based on the histogram data. The second texture feature reflects the distribution and complexity of the local texture in the image. Finally, the first texture feature and the second texture feature are fused by splicing or weighted summation as the final texture feature data. The texture feature data obtained in this way contains more texture features of the plant image, providing more texture details for subsequent analysis, thereby improving the accuracy of the recognition results.

[0077] For shape features, this embodiment uses edge detection technology and contour extraction technology to extract features, and the specific steps include:

[0078] Using edge detection technology to extract edge information of the plant part image;

[0079] According to each edge point in the edge information, a neighborhood search strategy is adopted to perform a neighborhood search, and a connection strategy is adopted to connect each searched adjacent edge point to obtain edge connection information;

[0080] According to the edge connection information, contour extraction is performed, and the extracted contour is analyzed to obtain shape feature data.

[0081] In this embodiment, the image is first processed by edge detection to extract edge information in the image. The edge information is represented in the form of a binary image, in which edge points are set to white (or high brightness) and non-edge points are set to black (or low brightness). For each edge point, a neighborhood search strategy is used to find adjacent edge points, and adjacent edge points are connected by connection strategies such as line segment connection and curve fitting to form continuous edge segments. On the basis of edge connection, contour information is identified and extracted. Contour extraction includes external contour extraction and internal contour extraction. For example, the edge line of the outer periphery of the object is selected as the contour line, and the boundary tracking algorithm is used to extract the external contour. For wormholes or damage on the blade, internal contour extraction needs to be used. The internal contour can be extracted by the area filling algorithm. For the extracted contour information, the shape feature data is obtained by analyzing the shape, length, curvature, etc. of the contour.

[0082] After obtaining the color feature data, texture feature data and shape feature data through the above steps, the current growth state of the plant is determined by calculating the similarity between the extracted feature data and the standard feature data in the pest and disease database, where the growth state includes the health state, disease state and pest state. For different growth states, automatic spraying of drugs or supplementation of light can be performed through automated mechanical control.

[0083] For the growth stage of the plant identified by the growth recognition model, in this embodiment, a standard environmental parameter database is established by the response surface method, and the corresponding optimal value of the environmental parameter is selected according to the identified growth stage. The specific steps include:

[0084] A response surface model was established with soil temperature and soil moisture as independent variables and the change rate of plant leaf color area as the dependent variable.

[0085] Solving the response surface model to obtain the soil temperature and humidity combination value that makes the change rate of plant leaf color area reach the optimal value;

[0086] According to the soil temperature and humidity combination values ​​corresponding to different growth stages, an environmental parameter database is constructed;

[0087] According to the growth stage, the corresponding optimal value of soil temperature and humidity is selected from the environmental parameter database.

[0088] In this embodiment, the response surface method is used to determine the optimal values ​​of soil temperature and humidity at different growth stages. Taking Geely Red as an example, the value of Geely Red is reflected in the red proportion of the leaves. Therefore, for the response surface model of Geely Red, the independent variables are soil temperature value and soil humidity value, and the dependent variable is the change rate of the color area of ​​the plant leaves. The response surface model is established by analyzing the relationship between temperature and humidity and the change rate of the red area. The solution is obtained by designing experiments or dynamic simulations and other methods to obtain the optimal soil temperature and humidity combination value that makes the change rate of the color area of ​​the plant leaves reach the optimal value at different growth stages. An environmental parameter database is established based on the optimal values ​​of these environmental parameters obtained by the solution, and the corresponding optimal soil temperature and humidity values ​​are extracted from the database through the identified growth stage, wherein the optimal soil temperature and humidity values ​​include the optimal soil temperature value and the optimal soil humidity value.

[0089] In the above data collection process, environmental data for plant growth are also collected, including soil temperature, soil moisture and meteorological data. In this embodiment, the environmental data is analyzed by the temperature and humidity prediction model to predict the temperature and humidity values ​​of the soil in the future. The temperature and humidity prediction model preferably adopts the Transformer-XL model to build it. In order to further improve the prediction accuracy, the training process of the Transformer-XL model is optimized by the Gray Wolf Optimization Algorithm (GWO) to reduce the training consumption of the model. GWO iteratively updates the hyperparameters of the model to ensure that the model is updated in time, thereby ensuring the accuracy and stability of soil temperature and humidity prediction. The specific model construction and training process can refer to the steps of the conventional method, which will not be repeated here.

[0090] Through the above steps, we can get the optimal soil temperature and humidity value and the predicted soil temperature and humidity value corresponding to the current growth stage of the plant. By comparing these two types of data, we can judge whether the temperature and humidity of the soil environment in the future stage meet the optimal value of plant growth. If the optimal value is not reached, the soil environment is dynamically adjusted according to the difference between the two, so as to achieve the optimal environmental parameters for plant growth.

[0091] The present embodiment provides a method for dynamically regulating a plant growth environment. The present invention comprehensively acquires and screens out all environmental factors that affect plant growth, and conducts a systematic principal component analysis to effectively identify key factors and their interrelationships, thereby ensuring the scientific nature and effectiveness of the overall regulation. The present invention optimizes the cultivation strategy by establishing a dynamic adjustment mechanism to identify and distinguish different growth stages and growth states of plants, and enhances the ability to respond to complex coupled changes by strengthening the long-term dynamic prediction of key factors such as soil temperature and humidity, so as to meet the regulation needs of plant cultivation in a multivariate environment. The present invention dynamically regulates the temperature, humidity, light, and pest and disease control required for plant growth through refined environmental management, thereby achieving global management and precise regulation of the cultivation process, improving the intensiveness and intelligence level of the overall cultivation method, and achieving cost reduction and efficiency improvement in plant cultivation.

[0092] See also Figure 3 Based on the same inventive concept, a second embodiment of the present invention provides a dynamic control system for plant growth environment, comprising:

[0093] The data acquisition module 10 is used to acquire image data and environmental data of plant growth, wherein the image data includes plant image data, and the environmental data includes soil temperature and humidity data and meteorological data;

[0094] A growth recognition module 20, used to input the image data into a growth recognition model to obtain the growth stage and growth state of the plant, wherein the growth recognition model includes a plant recognition module, a stage recognition module and a state recognition module;

[0095] A first control module 30 is used to spray the plant with medicine and / or supplement light according to the growth stage and the growth state;

[0096] The temperature and humidity prediction module 40 is used to obtain the optimal value of soil temperature and humidity of the plant growth environment according to the growth stage by using the response surface method, and input the environmental data into the temperature and humidity prediction model to obtain the predicted value of soil temperature and humidity;

[0097] The second control module 50 is used to compare the predicted soil temperature and humidity value with the optimal soil temperature and humidity value, and to control the soil temperature and humidity of the plant growth environment according to the comparison result.

[0098] The technical features and technical effects of the dynamic control system of the plant growth environment proposed in the embodiment of the present invention are the same as those of the method proposed in the embodiment of the present invention, and are not described in detail here. Each module in the above-mentioned dynamic control system of the plant growth environment can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0099] In addition, an embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.

[0100] See also Figure 4 , an internal structure diagram of a computer device in an embodiment, the computer device can specifically be a terminal or a server. The computer device includes a processor, a memory, a network interface, a display and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a dynamic control method for a plant growth environment is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0101] It can be understood by those skilled in the art that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computing device may include more or fewer components than shown in the figure, or combine certain components, or have the same component arrangement.

[0102] In addition, an embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above method are implemented.

[0103] In summary, a method, system, device and storage medium for dynamically controlling a plant growth environment proposed in an embodiment of the present invention, the method acquires image data and environmental data of plant growth, the image data includes plant image data, the environmental data includes soil temperature and humidity data and meteorological data; the image data is input into a growth recognition model to obtain the growth stage and growth state of the plant, the growth recognition model includes a plant recognition module, a stage recognition module and a state recognition module; according to the growth stage and the growth state, the plant is sprayed with medicine and / or supplemented with light; according to the growth stage, the response surface method is used to obtain the optimal value of soil temperature and humidity of the plant growth environment, the environmental data is input into a temperature and humidity prediction model to obtain a predicted value of soil temperature and humidity; the predicted value of soil temperature and humidity is compared with the optimal value of soil temperature and humidity, and according to the comparison result, the soil temperature and humidity of the plant growth environment is regulated. The present invention dynamically regulates the temperature, humidity, light and pest control required for plant growth through refined environmental management, realizes the overall management and precise regulation of the cultivation process, improves the intensive and intelligent level of the overall cultivation method, and realizes the cost reduction and efficiency increase of plant cultivation.

[0104] Each embodiment in this specification is described in a progressive manner, and the same or similar parts of each embodiment can be directly referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. It should be noted that the technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above-mentioned embodiments are not 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.

[0105] The above-mentioned embodiments only express several preferred implementation modes of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in the technical field, several improvements and substitutions can be made without departing from the technical principles of the present invention, and these improvements and substitutions should also be regarded as the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be based on the protection scope of the claims.

Claims

1. A method for dynamically controlling a plant growth environment, characterized in that: include: Acquire image data and environmental data of plant growth, wherein the image data includes plant image data, and the environmental data includes soil temperature and humidity data and meteorological data; Inputting the image data into a growth recognition model to obtain the growth stage and growth state of the plant, the growth recognition model comprising a plant recognition module, a stage recognition module and a state recognition module; According to the growth stage and the growth state, spraying the plants with drugs and / or supplementing light; According to the growth stage, the response surface method is used to obtain the optimal value of soil temperature and humidity in the plant growth environment, and the environmental data is input into the temperature and humidity prediction model to obtain the predicted value of soil temperature and humidity; The predicted soil temperature and humidity value is compared with the optimal soil temperature and humidity value, and the soil temperature and humidity of the plant growth environment are regulated based on the comparison result.

2. The method for dynamically controlling the plant growth environment according to claim 1, characterized in that: The step of inputting the image data into a growth recognition model to obtain the growth stage and growth state of the plant comprises: Inputting the image data into a plant recognition module constructed based on the YOLO model to recognize plant parts and obtain plant part images; Inputting the plant part image into a stage recognition module constructed based on a convolutional neural network model to obtain the growth stage of the plant, wherein the growth stage includes a budding stage, a growing stage, and a mature stage; The plant part image is input into a state recognition module for feature extraction and feature comparison to obtain the growth state of the plant, which includes the pest state and the disease state.

3. The method for dynamically controlling the plant growth environment according to claim 2, characterized in that: The step of inputting the plant part image into a state recognition module for feature extraction and feature comparison to obtain the growth state of the plant comprises: Using color space conversion and color segmentation technology to extract color features from the plant part image to obtain color feature data; Using a texture statistics method to extract texture features from the plant part image to obtain texture feature data; Using edge detection technology and contour extraction technology to extract shape features from the plant part image to obtain shape feature data; The color feature data, texture feature data and shape feature data are compared with the feature data in the pest and disease database, and the growth status of the plant is obtained based on the comparison result.

4. The method for dynamically controlling the plant growth environment according to claim 3, characterized in that: The step of extracting texture features from the plant part image using a texture statistics method to obtain texture feature data comprises: Converting the plant part image into a grayscale image, calculating a grayscale co-occurrence matrix of the grayscale image, and extracting first texture feature data from the grayscale co-occurrence matrix; A neighborhood region is defined for each pixel in the grayscale image, and a local binary pattern of the pixel is calculated in the neighborhood region to obtain a histogram of the local binary pattern, and second texture feature data is calculated according to the histogram; The first texture feature data and the second texture feature data are merged to obtain texture feature data.

5. The method for dynamically controlling the plant growth environment according to claim 3, characterized in that: The step of extracting shape features from the plant part image using edge detection technology and contour extraction technology to obtain shape feature data comprises: Using edge detection technology to extract edge information of the plant part image; According to each edge point in the edge information, a neighborhood search strategy is adopted to perform a neighborhood search, and a connection strategy is adopted to connect each searched adjacent edge point to obtain edge connection information; According to the edge connection information, contour extraction is performed, and the extracted contour is analyzed to obtain shape feature data.

6. The method for dynamically controlling the plant growth environment according to claim 1, characterized in that: The step of obtaining the optimal value of soil temperature and humidity in the plant growth environment by using the response surface method according to the growth stage comprises: A response surface model was established with soil temperature and soil moisture as independent variables and the change rate of plant leaf color area as the dependent variable. Solving the response surface model to obtain the soil temperature and humidity combination value that makes the change rate of plant leaf color area reach the optimal value; According to the soil temperature and humidity combination values ​​corresponding to different growth stages, an environmental parameter database is constructed; According to the growth stage, the corresponding optimal value of soil temperature and humidity is selected from the environmental parameter database.

7. The method for dynamically controlling the plant growth environment according to claim 1, characterized in that: The temperature and humidity prediction model is built based on the Transformer-XL model and trained using the Grey Wolf Optimization Algorithm.

8. A dynamic control system for plant growth environment, characterized in that: include: A data acquisition module, used to acquire image data and environmental data of plant growth, wherein the image data includes plant image data, and the environmental data includes soil temperature and humidity data and meteorological data; A growth recognition module, used for inputting the image data into a growth recognition model to obtain the growth stage and growth state of the plant, wherein the growth recognition model includes a plant recognition module, a stage recognition module and a state recognition module; A first control module is used to spray the plant with medicine and / or supplement light according to the growth stage and the growth state; A temperature and humidity prediction module is used to obtain the optimal soil temperature and humidity of the plant growth environment according to the growth stage by using the response surface method, and input the environmental data into the temperature and humidity prediction model to obtain the predicted value of soil temperature and humidity; The second control module is used to compare the predicted soil temperature and humidity value with the optimal soil temperature and humidity value, and to control the soil temperature and humidity of the plant growth environment according to the comparison result.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images