Self-adaptive garden landscape irrigation method and system based on artificial intelligence

Abstract

The invention relates to a self-adaptive garden landscape irrigation method and system based on artificial intelligence. The method comprises the following steps: acquiring a current garden plant monitoring live image and historical weather parameters of a position where a garden is located; the historical weather parameters comprise a soil humidity value per hour in a preset period and a rainfall condition in a future preset time period; matching types and growth stages of the garden plants according to the monitored actual images of the garden plants to obtain regular water absorption characteristic data of the garden plants; according to the historical weather parameters, obtaining predicted soil humidity data in a future preset time period; and generating a corresponding irrigation scheme based on the regular water absorption characteristic data of the garden plants and the predicted soil humidity data. By adopting the method, the accuracy of irrigation water volume prediction can be improved by combining the water absorption characteristics of plants and weather rainfall conditions.

Description

Technical Field

[0001] The present invention belongs to the technical field of garden irrigation, and particularly relates to an artificial intelligence-based adaptive garden landscape irrigation method and system. Background Art

[0002] With the development of urbanization, the demand for urban garden maintenance has been increasing, and garden maintenance technologies based on Internet of Things (IoT) big data technology have gradually emerged in the field of garden maintenance technology.

[0003] The garden maintenance technology based on IoT big data technology makes full use of the advantages of the IoT and big data. By connecting various sensors, it collects data on soil, atmospheric environment, and plant growth indicators over a past period of time, summarizes the rules of garden water resource consumption, predicts the amount of water for the next irrigation, and reduces the problems of water resource waste and insufficient irrigation while taking care of the normal growth of the garden.

[0004] However, the current IoT big data technology only makes predictions based on the past rules of garden water resource consumption, lacks the estimation of irrigation water volume by combining overall environmental perception, ignores influencing factors such as the water absorption requirements of different garden plant species and future meteorological precipitation, resulting in inaccurate prediction of irrigation water volume. Summary of the Invention

[0005] Based on this, it is necessary to provide an artificial intelligence-based adaptive garden landscape irrigation method and system for the above technical problems.

[0006] In a first aspect, the present application provides an artificial intelligence-based adaptive garden landscape irrigation method, including:

[0007] Obtain the current actual monitoring image of garden plants and the historical weather parameters of the location where the garden is located; the historical weather parameters include the hourly soil humidity values within a preset period and the precipitation situation within a future preset time period;

[0008] According to the actual monitoring image of garden plants, match the types and growth stages of garden plants to obtain the regular water absorption characteristic data of garden plants;

[0009] According to the historical weather parameters, obtain the predicted soil humidity data within a future preset time period;

[0010] Generate a corresponding irrigation plan based on the regular water absorption characteristic data of garden plants and the predicted soil humidity data.

[0011] Further, according to the actual monitoring image of garden plants, matching the types and growth stages of garden plants to obtain the regular water absorption characteristic data of garden plants includes:

[0012] Input the real-time images of garden plants into the trained garden plant classification model to obtain the types and growth stages of garden plants;

[0013] According to the types and growth stages of garden plants, determine the water requirements of garden plants, and set the water requirements of garden plants as the regular water absorption characteristic data of garden plants.

[0014] Furthermore, the trained garden plant classification model is obtained through the following method:

[0015] Obtain the plant pictures, plant types and growth stages in the garden plant database;

[0016] Extract the features of the plant pictures to obtain the plant detail feature maps;

[0017] Using the plant detail feature maps as the feature values and the plant types and growth stages as the label values, construct a plant classification model sample set;

[0018] Use the plant classification model sample set to train the garden plant classification model;

[0019] After completing the preset number of training times, obtain the trained garden plant classification model.

[0020] Furthermore, according to the historical weather parameters, obtain the predicted soil humidity data within a preset future time period, including:

[0021] The predicted soil humidity data is calculated by the following formula:

[0022] W = a×R p +b×W o

[0023] Where, W is the soil humidity data at the preset future time, a is the rainfall coefficient, R p is the predicted precipitation, b is the evaporation coefficient, and W o is the soil humidity value at the current moment.

[0024] Furthermore, the method also includes:

[0025] Based on the hourly soil humidity values within a preset period, obtain the linear part of the hourly soil humidity values within the preset period based on the linear regression model;

[0026] According to the linear part of the hourly soil humidity values within the preset period, calculate the slope of the linear part, and define the slope as the evaporation coefficient.

[0027] Furthermore, based on the regular water absorption characteristic data of garden plants and the predicted soil humidity data, generate the corresponding irrigation plan, including:

[0028] According to the regular water absorption characteristic data of garden plants, calculate the total water absorption of garden plants within a preset time period, and determine the total water absorption of garden plants as the total water absorption data of garden plants;

[0029] According to the total water absorption data of garden plants and the predicted soil humidity data, determine the amount of irrigation water required within a future preset time period;

[0030] Furthermore, the method further includes:

[0031] After implementing the irrigation plan, obtain the actual soil humidity data and the actual precipitation data;

[0032] According to the soil humidity data and the actual precipitation data, calculate the updated rainfall coefficient;

[0033] Wherein, the rainfall coefficient is calculated by the following formula:

[0034]

[0035] Wherein, a is the rainfall coefficient, W R is the actual soil humidity value, b is the evaporation coefficient, W o is the soil humidity value at the current moment, R r is the actual precipitation.

[0036] In a second aspect, the present application further provides an artificial intelligence-based adaptive garden landscape irrigation system, including:

[0037] A garden environment data acquisition module, configured to acquire the current monitoring live image of garden plants and the historical weather parameters of the location where the garden is located; the historical weather parameters include the hourly soil humidity value within a preset period and the precipitation situation within a future preset time period;

[0038] A garden plant feature extraction module, configured to match the type and growth stage of garden plants according to the monitoring live image of garden plants, and obtain the regular water absorption characteristic data of garden plants;

[0039] A garden soil humidity prediction module, configured to obtain the predicted soil humidity data within a future preset time period according to the historical weather parameters;

[0040] A garden irrigation plan determination module, configured to generate a corresponding irrigation plan based on the regular water absorption characteristic data of garden plants and the predicted soil humidity data.

[0041] In a third aspect, the present application further provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, it implements any one of the artificial intelligence-based adaptive garden landscape irrigation methods described in the first aspect of the present application.

[0042] Fourthly, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, any one of the artificial intelligence-based adaptive garden landscape irrigation methods described in the first aspect of the present application is implemented.

[0043] The above artificial intelligence-based adaptive garden landscape irrigation method and system obtain the current real-time monitoring images of garden plants and the historical weather parameters of the location where the garden is located; the historical weather parameters include the hourly soil humidity values within a preset period and the precipitation conditions within a preset future time period; according to the real-time monitoring images of garden plants, the types and growth stages of garden plants are matched to obtain the regular water absorption characteristic data of garden plants; according to the historical weather parameters, the predicted soil humidity data within a preset future time period is obtained; based on the regular water absorption characteristic data of garden plants and the predicted soil humidity data, a corresponding irrigation plan is generated, combining the water absorption characteristics of different types of plants and the precipitation conditions in the future for a period of time, realizing environmental perception, and based on the perception of the changes in the surrounding environment, realizing a more accurate estimation of the irrigation water volume required for garden plants, and improving the accuracy of the estimation of the irrigation water volume. Description of the Drawings

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required to be used in the description of the embodiments or related technologies. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0045] Figure 1 It is an implementation environment diagram provided for an exemplary embodiment of the present application;

[0046] Figure 2 It is a flowchart of an artificial intelligence-based adaptive garden landscape irrigation method provided for an exemplary embodiment of the present application;

[0047] Figure 3 It is a flowchart of the information entry process of common garden plants provided for an exemplary embodiment of the present application;

[0048] Figure 4 It is a flowchart of a soil humidity content prediction method provided for an exemplary embodiment of the present application;

[0049] Figure 5 It is a structural block diagram of an artificial intelligence-based adaptive garden landscape irrigation system provided for an exemplary embodiment of the present application; Detailed Embodiments

[0050] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0051] An artificial intelligence-based adaptive garden landscape irrigation method provided by an embodiment of the present application can be applied to an application environment as Figure 1 shown, including an irrigation terminal 101, an image acquisition device 102, a server 103, an irrigation execution device 104, and a communication network 105. The irrigation terminal 101 has functions of information acquisition, processing, storage, and output. It can obtain relevant precipitation and soil humidity information, and can obtain information from the image acquisition device 102 and the server 103, process it to generate an irrigation plan, and output the plan to the irrigation execution device 104 for execution. The image acquisition device 102 has functions of information acquisition and transmission. It can acquire image information and transmit it to the irrigation terminal 101. The server 103 can transmit the information required by the irrigation terminal 101 to the irrigation terminal 101 according to the request of the irrigation terminal 101. The irrigation execution device 104 has the function of performing irrigation behavior and can irrigate garden plants according to the irrigation plan generated by the irrigation terminal 101.

[0052] The irrigation terminal 101 communicates with the image acquisition device 102, the server 103, and the irrigation execution device 104 through the communication network 105. The communication network 105 can be a wireless network or a wired network.

[0053] In an exemplary embodiment, as Figure 2 shown, an artificial intelligence-based adaptive garden landscape irrigation method is provided. Taking the irrigation terminal 101 in Figure 1 as an example, the method includes the following steps S101 to S104. Among them:

[0054] Step S101, obtaining the current monitoring live image of garden plants and the historical weather parameters of the location where the garden is located; the historical weather parameters include the hourly soil humidity value within a preset period and the precipitation situation within a preset future time period.

[0055] Specifically, the irrigation terminal acquires the image information of garden plants, the soil moisture content information per hour in a past preset period at the location of the garden plants, and the meteorological rainfall conditions in a future preset time period. Schematically, the acquired image information of garden plants can be in the raster image format for subsequent processing by models or algorithms on the pictures. Optionally, the user can set the preset period and the future preset time respectively according to the local environmental conditions and in combination with work habits. Schematically, environmental perception is achieved by acquiring the soil moisture content information per hour in a past preset period at the location of the garden plants and the meteorological rainfall conditions in a future preset time period.

[0056] Step S102: Match the types and growth stages of garden plants according to the monitored live images of garden plants to obtain the regular water absorption characteristic data of garden plants.

[0057] Specifically, the irrigation terminal matches the plant species and the corresponding growth stage of the garden plants according to the image information of the garden plants, and acquires the water volume data required for the corresponding growth stage according to the plant species and the corresponding growth stage of the garden plants. Optionally, the garden plant database can be a collection of common garden plant information stored by the user in the irrigation terminal. The common garden plant information can include plant names, plant varieties, plant growth stages and corresponding plant characteristics, and the theoretical water absorption required for each growth stage of the plants.

[0058] Step S103: Obtain the predicted soil moisture data within a future preset time period according to historical weather parameters.

[0059] Specifically, according to historical weather parameters, by summarizing the variation law of the soil moisture content per hour in the past and combining with the predicted future precipitation, the soil moisture content in a future period of time is predicted. Schematically, the future period of time can be consistent with the prediction time of the predicted precipitation situation.

[0060] Step S104: Generate a corresponding irrigation plan based on the regular water absorption characteristic data of garden plants and the predicted soil moisture data.

[0061] Specifically, compare the obtained theoretical water absorption of the plants with the predicted soil wetness content in a future period of time to obtain the required irrigation water volume, determine the irrigation plan, and irrigate the plants according to the irrigation plan, and the irrigation process is completed. Schematically, if the theoretical water absorption is greater than the soil wetness content, then the plants need to be irrigated, and the relevant calculation is performed on the theoretical water absorption and the soil wetness content to obtain the required irrigation water volume. If the theoretical water absorption is less than the soil wetness content, then the plants do not need to be irrigated. Optionally, the irrigation plan can include the irrigation water volume, the irrigation location and the irrigation time.

[0062] In this embodiment, by extracting plant characteristics, determining the plant species and growth stage, and determining the theoretical water absorption amount, combined with the past change law of soil moisture content and future precipitation conditions, the irrigation water amount is estimated more accurately.

[0063] In one of the embodiments, according to the real-time monitoring images of garden plants, the types and growth stages of garden plants are matched to obtain the regular water absorption characteristic data of garden plants, including:

[0064] S201, input the real-time monitoring images of garden plants into the trained garden plant classification model to obtain the types and growth stages of garden plants.

[0065] Specifically, the irrigation terminal extracts features from the obtained plant image information, and matches the plant species and corresponding growth stage that conform to the features in the garden plant database. Schematically, the garden plant classification model can match the image features and the plant features in the database according to the input image information of the garden plants, and output the plant species and growth stage of the garden plants.

[0066] S202, according to the types and growth stages of garden plants, determine the water requirement of garden plants, and set the water requirement of garden plants as the regular water absorption characteristic data of garden plants.

[0067] Among them, the irrigation terminal obtains the water absorption amount data required for the corresponding growth stage by consulting the information of the plant species in the garden plant database according to the determined plant species and growth stage of the garden plants.

[0068] In this embodiment, the irrigation terminal determines the theoretical water absorption amount of the plant by matching the garden plant image and the corresponding plant species and growth stage, improving the accuracy of predicting the water requirement of the plant.

[0069] In one of the embodiments, the trained garden plant classification model is obtained by the following method:

[0070] S301, obtain the plant pictures, plant species and growth stages in the garden plant database.

[0071] Specifically, the irrigation terminal obtains the plant photos of the same plant species at different growth stages from the garden plant database.

[0072] S302, extract features from the plant pictures to obtain plant detail feature maps.

[0073] Specifically, the plant pictures can be preprocessed first to obtain plant grayscale pictures, and based on the convolutional neural network system, the grayscale pictures are used to extract features to obtain plant detail feature maps that highlight the plant contour features.

[0074] S303. Construct a sample set for the plant classification model with the detailed plant feature maps as the eigenvalue and the plant species and growth stages as the label values.

[0075] Specifically, pair the detailed plant feature maps with their corresponding plant species and growth stages. Use the detailed plant feature maps as the eigenvalue and the plant species and growth stages as the label values to form a sample for the plant classification model, and then construct a complete sample set for the plant classification model.

[0076] S304. Use the sample set for the plant classification model to train the landscape plant classification model.

[0077] S305. After completing the preset number of training times, obtain the trained landscape plant classification model.

[0078] It should be noted that the landscape plant classification model can be a multi-classification model. Softmax and other multi-classification models can be selected as the initial model. Use the eigenvalue in the sample set for the plant classification model, that is, the detailed plant feature maps, as the input data, and the plant species and growth stages as the label values to train the landscape plant classification model. After the preset number of training times, obtain the final landscape plant classification model. The final achieved effect is that the irrigation terminal obtains a photo of a landscape plant and inputs the photo of the landscape plant into the landscape plant classification model, and the species and growth stage of the landscape plant in the photo can be obtained.

[0079] In this embodiment, by constructing a landscape plant classification model, the information on the species and growth stage of the plant can be directly obtained from the acquired landscape plant image, reducing the time and error of manual comparison and improving the efficiency and accuracy of plant species identification.

[0080] In one embodiment, according to historical weather parameters, obtain the predicted soil moisture data for a preset future time period, including:

[0081] The predicted soil moisture data is calculated by the following formula:

[0082] W = a×R p + b×W o

[0083] where W is the soil moisture data for the preset future time, a is the rainfall coefficient, R p is the predicted precipitation, b is the evaporation coefficient, and W o is the soil moisture value at the current moment.

[0084] Specifically, the soil humidity data for a preset future time refers to the predicted soil humidity value for the preset future time, which is a factor affecting the irrigation water volume in the irrigation plan. Schematically, the rainfall coefficient is a coefficient that improves the accuracy of the predicted precipitation amount to reduce the error between the predicted precipitation amount and the actual precipitation amount, and the rainfall coefficient is recalculated after the implementation of the irrigation plan. Optionally, the evaporation coefficient is a coefficient that characterizes the natural evaporation of the water content in the soil and is used to measure the trend of soil humidity change in the absence of precipitation. The soil humidity value at the current moment refers to the soil humidity value obtained at the current moment, and the soil humidity value for the preset future time is predicted based on the soil humidity value at the current moment.

[0085] In this embodiment, by constructing a formula for predicting soil humidity data, the law of soil humidity change and future precipitation conditions are combined to predict the future soil humidity value. More influencing factors are combined to predict the soil humidity value, improving the accuracy of predicting the soil humidity value.

[0086] In one of the embodiments, the method further includes:

[0087] S501, based on the hourly soil humidity values within a preset period, obtain the linear part of the hourly soil humidity values within the preset period based on a linear regression model;

[0088] Among them, taking the hourly soil humidity values within the preset period as the variable y to be determined, and the hour corresponding to the soil humidity value as the independent variable x, a data set is formed. The data set is input into the linear regression model to obtain the linear part of the data set and the fitting curve of the linear part. Schematically, the linear regression model can be a one-dimensional linear regression model. Optionally, the linear part of the data set reduces the sudden change in soil humidity caused by rainfall and increases the accuracy of determining the natural evaporation time period.

[0089] S502, calculate the slope of the linear part according to the linear part of the hourly soil humidity values within the preset period, and define the slope as the evaporation coefficient.

[0090] Specifically, according to the fitting curve of the linear part, the linear fitting equation is obtained, the curve slope is obtained from the equation, and the curve slope is defined as the evaporation coefficient to characterize the proportion of the natural evaporation part of the soil humidity in the total humidity in the absence of precipitation.

[0091] In this embodiment, by summarizing the law of natural evaporation of soil humidity value over time and quantifying this law to obtain the evaporation coefficient, a quantitative factor is provided for predicting the required future soil humidity value.

[0092] In one of the embodiments, based on the regular water absorption characteristic data of garden plants and the predicted soil humidity data, a corresponding irrigation plan is generated, including:

[0093] S601. Calculate the total water absorption of landscape plants within a preset time period based on the regular water absorption characteristic data of landscape plants, and determine the total water absorption of landscape plants as the total water absorption data of landscape plants.

[0094] S602. Determine the required irrigation water volume within a future preset time period based on the total water absorption data of landscape plants and the predicted soil humidity data.

[0095] Specifically, calculate the total water absorption within the future preset time period according to the theoretical water absorption of landscape plants and the length of the future preset time set by the user, and numerically compare the total water absorption with the predicted soil humidity value. If the total water absorption is greater than the measured soil humidity value, calculate the difference between the total water absorption and the predicted soil humidity value, and define this difference as the required irrigation water volume.

[0096] In this embodiment, by calculating the total water absorption of plants within the preset time and comparing the total water absorption with the predicted soil humidity value, and calculating the irrigation water volume according to the comparison result, it combines the water absorption characteristics of different plant species and the prediction of soil humidity, and calculates the irrigation water volume more accurately.

[0097] In one of the embodiments, the method further includes:

[0098] After implementing the irrigation plan, obtain the actual soil humidity data and the actual precipitation data.

[0099] Calculate the updated rainfall coefficient according to the soil humidity data and the actual precipitation data.

[0100] Among them, the rainfall coefficient is calculated by the following formula:

[0101]

[0102] Among them, a is the rainfall coefficient, W r is the actual soil humidity value, b is the evaporation coefficient, W o is the soil humidity value at the current moment, R r is the actual precipitation.

[0103] Specifically, the rainfall coefficient is a coefficient that measures the accuracy of predicted precipitation and is corrected based on the actual precipitation and actual soil moisture. The change in the value of the rainfall coefficient can reflect the trend of the local rainfall scale and improve the accuracy of the impact of predicted precipitation on soil moisture. Schematically, the actual soil moisture value is the actual moisture value of the soil obtained after implementing the irrigation plan. Optionally, the evaporation coefficient is used to reflect the proportion of the natural evaporation of the soil moisture value in the total moisture value without precipitation, reflecting the change trend of the soil moisture without precipitation. Schematically, the soil moisture value at the current moment refers to the soil moisture data at the moment when the predicted soil moisture data is calculated. Optionally, the actual precipitation refers to the actual precipitation value within a preset time.

[0104] In this embodiment, the irrigation terminal updates the rainfall coefficient in a timely manner by obtaining the actual precipitation and actual soil moisture after implementing the irrigation plan, improving the accuracy of the predicted soil moisture data.

[0105] To further illustrate the solution of the application embodiment in this embodiment, a specific example is given below for illustration:

[0106] (I) Process of Entering Information of Common Garden Plants

[0107] Refer to Figure 3 , and the specific process is as follows:

[0108] Step S01: Enter basic information into the system.

[0109] Specifically, it includes: entering the plant name, plant characteristics, plant species, plant growth stage, and water requirement at each growth stage of common garden plants into the irrigation terminal.

[0110] Step S02: Enter plant photos.

[0111] Specifically, it includes: entering the plant photos taken at different growth stages into the irrigation terminal to form a plant information data, which contains the following key fields: (plant name, plant characteristics, plant species, plant growth stage and corresponding photos).

[0112] (II) Method for Predicting Soil Moisture Content

[0113] Refer to Figure 4 , and the specific process is as follows:

[0114] S01: Obtain environmental information.

[0115] Specifically, it includes: obtaining environmental information, including real-time photos of garden plants taken, soil moisture per hour in the garden in the past day, and rainfall in the garden location in the next day.

[0116] S02: Determine the water absorption characteristics of plants.

[0117] Specifically, it includes: extracting the characteristics of garden plants from the real-time photos of garden plants taken, comparing them with the plant characteristics in the obtained garden plant database, determining the plant species and growth stage of the garden plants in the photos, and obtaining the corresponding plant regular water absorption data.

[0118] S03: Predict the soil moisture value.

[0119] With the soil moisture per hour in the past day and the rainfall in the garden location in the next day, the soil moisture value in the next day is budgeted. Among them, the formula for budgeting the soil moisture value is:

[0120] W = a×R p +b×W o

[0121] where W is the soil moisture data at a future preset time, a is the rainfall coefficient, R p is the predicted precipitation, b is the evaporation coefficient, and W o is the soil moisture value at the current moment.

[0122] According to the above formula, the soil moisture value in the next day is calculated.

[0123] S04: Generate an irrigation plan.

[0124] Specifically, it includes: calculating the total water absorption of plants within one day according to the plant regular water absorption data, comparing the total water absorption of plants within one day with the soil moisture value after one day. If the total water absorption is less than the soil moisture value after one day, the plant does not need irrigation; otherwise, if the total water absorption is greater than the soil moisture value after one day, then calculate the difference between the total water absorption and the soil moisture value, set the difference as the irrigation water volume, and set one day later as the irrigation time. Combining with the location of the plant, an irrigation plan is formed.

[0125] In the above artificial intelligence-based adaptive garden landscape irrigation method, the total water absorption of plants is accurately obtained by identifying the images of plants. At the same time, combining the natural change law of soil moisture and the influence of future precipitation on soil moisture, environmental perception is realized. Based on the perception of the changes in the surrounding environment, more accurate prediction of the irrigation water volume required by plants is realized, and the waste of irrigation water resources is reduced.

[0126] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are sequentially shown according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0127] Based on the same inventive concept, an embodiment of the present application also provides an irrigation system for implementing the above-mentioned artificial intelligence-based adaptive garden landscape irrigation method. The implementation solutions provided by this system to solve problems are similar to the implementation solutions described in the above method. Therefore, the specific limitations in one or more of the following irrigation system embodiments can refer to the limitations on the artificial intelligence-based adaptive garden landscape irrigation method in the above text, and will not be repeated here.

[0128] In an exemplary embodiment, as Figure 5 shown, an artificial intelligence-based adaptive garden landscape irrigation system 500 is provided, including:

[0129] A garden environment data acquisition module 501, configured to acquire the current actual monitoring image of garden plants and the historical weather parameters of the location where the garden is located; the historical weather parameters include the soil humidity value per hour within a preset period and the precipitation situation within a preset future time period;

[0130] A garden plant feature extraction module 502, configured to match the types and growth stages of garden plants according to the current actual monitoring image of garden plants, and obtain the regular water absorption feature data of garden plants;

[0131] A garden soil humidity prediction module 503, configured to obtain the predicted soil humidity data within a preset future time period according to the historical weather parameters;

[0132] A garden irrigation plan determination module 504, configured to generate a corresponding irrigation plan based on the regular water absorption feature data of garden plants and the predicted soil humidity data.

[0133] Further, the system further includes:

[0134] Matching the types and growth stages of garden plants according to the current actual monitoring image of garden plants to obtain the regular water absorption feature data of garden plants, including:

[0135] Input the real-time images of garden plants into the trained garden plant classification model to obtain the types and growth stages of garden plants;

[0136] Determine the water requirement of the garden plants according to the types and growth stages of the garden plants, and set the water requirement of the garden plants as the regular water absorption characteristic data of the garden plants.

[0137] Furthermore, the system also includes:

[0138] The trained garden plant classification model is obtained through the following method:

[0139] Obtain the plant pictures, types and growth stages in the garden plant database;

[0140] Extract features from the plant pictures to obtain the plant detail feature maps;

[0141] Construct a plant classification model sample set with the plant detail feature maps as the eigenvalue and the plant types and growth stages as the label values;

[0142] Use the plant classification model sample set to train the garden plant classification model;

[0143] After completing the preset number of training times, obtain the trained garden plant classification model.

[0144] Furthermore, the system also includes:

[0145] Obtain the predicted soil humidity data within a preset future time period according to the historical weather parameters, including:

[0146] The predicted soil humidity data is calculated by the following formula:

[0147] W = a×R p + b×W o

[0148] where W is the soil humidity data at the preset future time, a is the rainfall coefficient, R p is the predicted precipitation, b is the evaporation coefficient, and W o is the current soil humidity value.

[0149] Furthermore, the system also includes:

[0150] Based on the linear regression model, obtain the linear part of the hourly soil humidity values within a preset period according to the hourly soil humidity values within the preset period;

[0151] Calculate the slope of the linear part according to the linear part of the hourly soil humidity values within the preset period, and define the slope as the evaporation coefficient.

[0152] Furthermore, the system also includes:

[0153] Based on the regular water absorption characteristic data of garden plants and the predicted soil humidity data, a corresponding irrigation plan is generated, including:

[0154] According to the regular water absorption characteristic data of garden plants, calculate the total water absorption of garden plants within a preset time period, and determine the total water absorption data of garden plants;

[0155] According to the total water absorption data of garden plants and the predicted soil humidity data, determine the required irrigation water volume within a future preset time period;

[0156] Furthermore, the system further includes:

[0157] After implementing the irrigation plan, obtain the actual soil humidity data and the actual precipitation data;

[0158] According to the soil humidity data and the actual precipitation data, calculate the updated rainfall coefficient;

[0159] Among them, the rainfall coefficient is calculated by the following formula:

[0160]

[0161] Among them, a is the rainfall coefficient, W_R is the actual soil humidity value, b is the evaporation coefficient, W o is the soil humidity value at the current moment, R r is the actual precipitation.

[0162] In one embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of an adaptive garden landscape irrigation method based on artificial intelligence as described above are implemented.

[0163] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0164] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiments described above are only illustrative. The components described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the present disclosure solution. Those of ordinary skill in the art can understand and implement it without creative work.

[0165] The above-described embodiments merely represent several implementation manners of the embodiments of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent of the embodiments of the application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the embodiments of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the embodiments of the present application.

Claims

1. An adaptive garden landscape irrigation method based on artificial intelligence, characterized in that: The method comprises: Acquire the current real-time monitoring image of the garden plants and the historical weather parameters of the location of the garden; the historical weather parameters include the hourly soil moisture value within a preset period and the precipitation conditions within a future preset time period; According to the real-time monitoring image of the garden plants, the types and growth stages of the garden plants are matched to obtain regular water absorption characteristic data of the garden plants; Obtaining predicted soil moisture data within a future preset time period based on the historical weather parameters; Based on the garden plant regular water absorption characteristic data and the predicted soil moisture data, a corresponding irrigation plan is generated.

2. The method according to claim 1, characterized in that The method of matching the types and growth stages of the garden plants according to the real-time monitoring images of the garden plants to obtain the regular water absorption characteristic data of the garden plants includes: Inputting the real-time monitoring image of the garden plants into a trained garden plant classification model to obtain the type and growth stage of the garden plants; The plant water requirement of the garden plant is determined according to the type and growth stage of the garden plant, and the plant water requirement of the garden plant is set as the regular water absorption characteristic data of the garden plant.

3. The method according to claim 2, characterized in that The trained garden plant classification model is obtained by the following method: Obtain plant images, plant species and growth stages from the garden plant database; Extracting features from the plant image to obtain a plant detail feature map; Using the plant detail feature map as feature value and the plant species and growth stage as label value, a plant classification model sample set is constructed; Using the plant classification model sample set to train the garden plant classification model; After completing the preset number of training times, the trained garden plant classification model is obtained.

4. The method according to claim 1, characterized in that: The step of obtaining predicted soil moisture data within a preset time period in the future according to the historical weather parameters includes: The predicted soil moisture data is calculated by the following formula: W=a×R p +b×W o Among them, W is the soil moisture data at a preset time in the future, a is the rainfall coefficient, and R p is the expected precipitation, b is the evaporation coefficient, W o is the soil moisture value at the current moment.

5. The method according to claim 4, characterized in that: The method also includes: According to the hourly soil moisture values ​​within the preset period, based on a linear regression model, a linear part of the hourly soil moisture values ​​within the preset period is obtained; According to the linear part of the hourly soil moisture value within the preset period, the slope of the linear part is calculated, and the slope is defined as the evaporation coefficient.

6. The method according to claim 1, characterized in that The generating a corresponding irrigation plan based on the garden plant regular water absorption characteristic data and the predicted soil moisture data comprises: Calculating the total water absorption of the garden plant in a preset time period according to the regular water absorption characteristic data of the garden plant, and determining the total water absorption as the total water absorption data of the garden plant; Determine the amount of irrigation water required in a future preset time period according to the total water absorption data of the garden plants and the predicted soil moisture data; According to the irrigation water volume, the positions of the garden plants are matched and an irrigation plan within a preset time is generated.

7. The method according to claim 6, characterized in that The method also includes: After implementing the irrigation scheme, obtaining actual soil moisture data and actual precipitation data; Calculating the updated rainfall coefficient according to the soil moisture data and the actual precipitation data; The rainfall coefficient is calculated by the following formula: Where a is the rainfall coefficient, W R is the actual soil moisture value, b is the evaporation coefficient, W o is the soil moisture value at the current moment, R r is the actual precipitation.

8. An adaptive garden landscape irrigation system based on artificial intelligence, characterized in that: include: The garden environment data acquisition module is used to obtain the current garden plant monitoring real-time image and the historical weather parameters of the location of the garden; the historical weather parameters include the hourly soil moisture value within a preset period and the precipitation conditions within a future preset time period; A garden plant feature extraction module is used to match the type and growth stage of the garden plant according to the real-time monitoring image of the garden plant to obtain the regular water absorption feature data of the garden plant; A garden soil moisture prediction module obtains predicted soil moisture data within a future preset time period based on the historical weather parameters; The garden irrigation scheme determination module is used to generate a corresponding irrigation scheme based on the garden plant regular water absorption characteristic data and the predicted soil moisture data.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: 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