阅读说明：本技术 一种基于机器学习的火龙果种植全周期植物营养配置方法 (Machine learning-based pitaya planting full-period plant nutrition configuration method ) 是由 温标堂 梁海玲 龙宣佑 江万里 覃敬源 伍祚斌 黄文娟 阳继辉 朱文国 李蝶 黄 于 2021-07-27 设计创作，主要内容包括：本发明公开一种基于机器学习的火龙果种植全周期植物营养配置方法,包括以下步骤：建立土壤影响神经网络模型；获得土壤影响营养系数；建立生长状态网络模型；获得生长状态影响营养系数；建立气象影响神经网络模型；获得气象影响营养系数；建立灌溉影响神经网络模型；获得灌溉影响营养系数；建立施肥影响网络模型；获得施肥影响营养系数；建立地形影响网络模型；获得地形影响营养系数；构建成数据集；划分所述数据集的70％用做模型构建,所述数据集的30％用做模型验证；构建数据集与火龙果营养N、P、K比例的回归模型,获得不同生长阶段火龙果需要相应的N、P、K的比例。本发明具有很好的针对性和预测性,抗干扰能力强,准确性高,诊断速度快。(The invention discloses a machine learning-based pitaya planting full-period plant nutrition configuration method, which comprises the following steps of: establishing a soil influence neural network model; obtaining a soil influence nutrition coefficient; establishing a growth state network model; obtaining growth state influence nutrition coefficients; establishing a meteorological influence neural network model; obtaining a weather influence nutrition coefficient; establishing an irrigation influence neural network model; obtaining irrigation influence nutrition coefficients; establishing a fertilization influence network model; obtaining fertilization influence nutrition coefficients; establishing a terrain influence network model; obtaining a landform influence nutrition coefficient; constructing a data set; dividing 70% of the data set to be used as model construction, and dividing 30% of the data set to be used as model verification; and (3) constructing a regression model of the ratio of the data set to the nutrition N, P, K of the pitaya to obtain the corresponding N, P, K ratio required by the pitaya in different growth stages. The invention has good pertinence and predictability, strong anti-interference capability, high accuracy and high diagnosis speed.)

1. A full-period plant nutrition configuration method for dragon fruit planting based on machine learning is characterized by comprising the following steps:

establishing a soil influence neural network model; acquiring soil data of the environment where the dragon fruits are located and inputting the soil data into a soil influence neural network model to obtain a soil influence nutrition coefficient;

establishing a growth state network model; acquiring growth image data of the dragon fruits and inputting the growth image data into a growth state network model to obtain growth state influence nutrition coefficients;

establishing a meteorological influence neural network model; acquiring meteorological data of the environment where the dragon fruits are located and inputting the meteorological data into a meteorological influence neural network model to obtain a meteorological influence nutrition coefficient;

establishing an irrigation influence neural network model; acquiring irrigation data in the dragon fruit planting process and inputting the irrigation data into an irrigation influence neural network model to obtain an irrigation influence nutrition coefficient;

establishing a fertilization influence network model; acquiring fertilization data in the dragon fruit planting process and inputting the fertilization data into a fertilization influence network model to obtain fertilization influence nutrition coefficients;

establishing a terrain influence network model; acquiring topographic data of the environment of the dragon fruit and inputting the topographic data into a topographic influence network model to obtain a topographic influence nutrition coefficient;

respectively filtering the soil influence nutrition coefficient, the growth state influence nutrition coefficient, the weather influence nutrition coefficient, the irrigation influence nutrition coefficient, the fertilization influence nutrition coefficient and the terrain influence nutrition coefficient to construct a data set; dividing 70% of the data set to be used as model construction, and dividing 30% of the data set to be used as model verification; constructing a regression model of the ratio of the data set to the nutrition N, P, K of the pitaya to obtain the corresponding ratio N, P, K required by the pitaya in different growth stages;

the soil influence neural network model is used for acquiring a large amount of soil data corresponding to soil influence nutrition coefficient values after soil influence nutrition coefficient values are set for pitaya nutrition according to soil data by a plant nutrition expert; the soil influence neural network model takes soil data as input and takes a soil influence nutrition coefficient as output;

the growth state network model is used for acquiring a large amount of growth image data corresponding to growth state influence nutrition coefficient values after the growth state influence nutrition coefficient values of the pitaya nutrition are set by plant nutrition experts according to growth image data; the growth state network model takes growth image data as input and takes growth state influence nutrition coefficients as output;

the meteorological influence neural network model is used for acquiring a large amount of meteorological data corresponding to meteorological influence nutrition coefficient values after the meteorological influence nutrition coefficient values of the pitaya nutrition are set by plant nutrition experts according to meteorological data; the meteorological influence neural network model takes meteorological data as input and a meteorological influence nutrition coefficient as output;

the irrigation influence neural network model is used for acquiring a large amount of irrigation data corresponding to irrigation influence nutrition coefficient values after the irrigation influence nutrition coefficient values of the pitaya nutrition are set according to irrigation data by plant nutrition experts; the irrigation influence neural network model takes irrigation data as input and takes an irrigation influence nutrition coefficient as output;

the fertilization influence network model is used for collecting a large amount of fertilization data corresponding to fertilization influence nutrition coefficient values after the fertilization influence nutrition coefficient values of the nutrition of the dragon fruits are set according to the fertilization data by plant nutrition experts; the fertilization influence network model takes fertilization data as input and fertilization influence nutrition coefficients as output;

the terrain influence network model is used for acquiring a large amount of terrain data corresponding to terrain influence nutrition coefficient values after the terrain influence nutrition coefficient values of the pitaya nutrition are set by a plant nutrition expert according to the terrain data; the terrain influence network model takes terrain data as input and takes a terrain influence nutrition coefficient as output.

2. The machine learning-based pitaya planting full-period plant nutrition configuration method as claimed in claim 1, wherein the machine learning-based pitaya planting full-period plant nutrition configuration method comprises the following steps: the soil data comprises soil structure, soil air content, earth surface average temperature, different soil layer average humidity, soil pH value, soil conductivity, soil element components, soil organic matter content and soil microorganism content.

3. The machine learning-based pitaya planting full-period plant nutrition configuration method as claimed in claim 1, wherein the machine learning-based pitaya planting full-period plant nutrition configuration method comprises the following steps: the growth image data comprises images of branches and leaves of the dragon fruit, images of flowers of the dragon fruit and images of fruits of the dragon fruit.

4. The machine learning-based pitaya planting full-period plant nutrition configuration method as claimed in claim 1, wherein the machine learning-based pitaya planting full-period plant nutrition configuration method comprises the following steps: the meteorological data comprise illumination duration, illumination intensity, rainfall, atmospheric temperature, average temperature, limiting temperature, wind speed and evaporation capacity.

5. The machine learning-based pitaya planting full-period plant nutrition configuration method as claimed in claim 1, wherein the machine learning-based pitaya planting full-period plant nutrition configuration method comprises the following steps: the irrigation data includes irrigation water type, quality, irrigation time, irrigation amount and irrigation frequency.

6. The machine learning-based pitaya planting full-period plant nutrition configuration method as claimed in claim 1, wherein the machine learning-based pitaya planting full-period plant nutrition configuration method comprises the following steps: the fertilization information comprises the type, nutrient content, fertilization time, fertilization amount, fertilization concentration and fertilization frequency of the fertilizer.

7. The machine learning-based pitaya planting full-period plant nutrition configuration method as claimed in claim 1, wherein the machine learning-based pitaya planting full-period plant nutrition configuration method comprises the following steps: the topographic information comprises the terrain, the altitude, the gradient and the slope direction of the dragon fruit.

8. The machine learning-based pitaya planting full-period plant nutrition configuration method as claimed in claim 1, wherein the machine learning-based pitaya planting full-period plant nutrition configuration method comprises the following steps: the method is characterized in that the soil influence nutrition coefficient, the growth state influence nutrition coefficient, the weather influence nutrition coefficient, the irrigation influence nutrition coefficient, the fertilization influence nutrition coefficient and the terrain influence nutrition coefficient are respectively filtered, and the specific filtering method comprises the following steps: only one set of the same or similar coefficient data is retained, and the other sets of coefficient data are removed.

9. A system for applying the method for preparing plant nutrients for a whole period of dragon fruit planting according to any one of claims 1-8, comprising:

the soil data collection module is a soil moisture content instrument and is used for collecting soil data of the environment where the dragon fruits are located;

the image processing module is used for collecting growth image data of the dragon fruits and carrying out image processing;

the meteorological data collection module is a meteorological station and is used for collecting meteorological data of the environment where the dragon fruits are located;

the irrigation data collection module is equipment provided with an irrigation system and is used for collecting irrigation data in the dragon fruit planting process;

the fertilization data collection module is equipment provided with a fertilization system and is used for collecting irrigation data in the dragon fruit planting process;

the topographic data collecting module is used for collecting topographic data of the environment where the dragon fruits are located;

and the machine learning module is used for storing the soil influence neural network model, the growth state network model, the meteorological influence neural network model, the irrigation influence neural network model, the fertilization influence network model and the terrain influence network model and respectively outputting the soil influence nutrition coefficient, the growth state influence nutrition coefficient, the meteorological influence nutrition coefficient, the irrigation influence nutrition coefficient, the fertilization influence nutrition coefficient and the terrain influence nutrition coefficient.

The database module is used for storing a large amount of soil data, growth image data, meteorological data, irrigation data and topographic data;

the calculation module is used for respectively filtering the soil influence nutrition coefficient, the growth state influence nutrition coefficient, the meteorological influence nutrition coefficient, the irrigation influence nutrition coefficient, the fertilization influence nutrition coefficient and the terrain influence nutrition coefficient to construct a data set; dividing 70% of the data set to be used as model construction, and dividing 30% of the data set to be used as model verification; constructing a regression model of the ratio of the data set to the nutrition N, P, K of the pitaya to obtain the corresponding ratio N, P, K required by the pitaya in different growth stages;

the soil data collection module, the image processing module, the meteorological data collection module, the irrigation data collection module, the fertilization data collection module, the terrain data collection module and the database module are respectively connected with the machine learning module; the machine learning module is connected with the computing module.

10. The system of claim 9, wherein: still include the camera, the camera is installed on unmanned aerial vehicle and can wireless connection image processing module.

Technical Field

The invention relates to the technical field of dragon fruit planting, in particular to a machine learning-based full-period plant nutrition configuration method for dragon fruit planting.

Background

Dragon fruit [ Hylocereus undatus spp ], also called red dragon fruit, sweet fruit, sweetheart fruit, sesamol fruit and the like, is a perennial climbing fleshy plant of the family Cactaceae (Cactaceae) scale genus (Hylocereus undatus), belongs to tropical and subtropical fruit trees, originates from tropical rainforest and desert zone in central america, is artificially cultivated in central america, israel, vietnam, thailand, usa and the like, is cultivated in taiwan in China more frequently, has been cultivated for decades, has developed in the world of china in the recent years in Guangxi, Guangdong, Fujian and the like, has developed area of about 32 ten thousand mu, wherein the Fujian is about 2 thousand mu and is in the momentum of rapid development. The dragon fruit integrates fruits, flowers, vegetables and health care into a whole, and has high economic value. The root system of the dragon fruit is developed, has no obvious main root and is shallow and is mostly distributed in a surface soil layer of 2-15 cm; the stem vines are provided with climbing roots which grow upwards on the shed frame or other columnar supports; the plant grows vigorously, has strong germination force and branch force, and can grow all the year round; the leaves are not available, photosynthesis is completed by stem vines, and fleshy stem vines are thick and are in a triangular column shape or a quadrangular prism shape. The dragon fruit has the advantages of strong adaptability, fast growth, early fruiting, large fruit, high yield, no big or small years and the like.

The pitaya needs to be provided with different N, P, K proportion nutrition formulas in the whole planting period. At present, the nutrition allocation method for judging the dragon fruits is various, such as diagnosis by a soil analysis method, an appearance diagnosis method, a chemical analysis method and the like. The soil analysis method has good pertinence and predictability, but the soil interference factors are more, and the result accuracy is lower; the appearance diagnosis method is limited by experience and is easy to generate misdiagnosis; chemical analysis requires a long examination time. In order to improve the accuracy of diagnosis, a method for preparing a nutrient formula of pitaya capable of judging different growth stages is urgently needed.

Disclosure of Invention

The invention aims to provide a method for configuring plant nutrition in a full-period pitaya planting based on machine learning, solves the technical problems of more interference, low accuracy, easy misdiagnosis and long detection time of the general method for configuring plant nutrition in the full-period pitaya planting, and has the advantages of good pertinence and predictability, strong anti-interference capability, high accuracy and high diagnosis speed.

In order to achieve the aim, the full-period plant nutrition configuration method based on machine learning for dragon fruit planting is provided, and comprises the following steps:

establishing a soil influence neural network model; acquiring soil data of the environment where the dragon fruits are located and inputting the soil data into a soil influence neural network model to obtain a soil influence nutrition coefficient;

establishing a growth state network model; acquiring growth image data of the dragon fruits and inputting the growth image data into a growth state network model to obtain growth state influence nutrition coefficients;

establishing a meteorological influence neural network model; acquiring meteorological data of the environment where the dragon fruits are located and inputting the meteorological data into a meteorological influence neural network model to obtain a meteorological influence nutrition coefficient;

establishing an irrigation influence neural network model; acquiring irrigation data in the dragon fruit planting process and inputting the irrigation data into an irrigation influence neural network model to obtain an irrigation influence nutrition coefficient;

establishing a fertilization influence network model; acquiring fertilization data in the dragon fruit planting process and inputting the fertilization data into a fertilization influence network model to obtain fertilization influence nutrition coefficients;

establishing a terrain influence network model; acquiring topographic data of the environment of the dragon fruit and inputting the topographic data into a topographic influence network model to obtain a topographic influence nutrition coefficient;

respectively filtering the soil influence nutrition coefficient, the growth state influence nutrition coefficient, the weather influence nutrition coefficient, the irrigation influence nutrition coefficient, the fertilization influence nutrition coefficient and the terrain influence nutrition coefficient to construct a data set; dividing 70% of the data set to be used as model construction, and dividing 30% of the data set to be used as model verification; constructing a regression model of the ratio of the data set to the nutrition N, P, K of the pitaya to obtain the corresponding ratio N, P, K required by the pitaya in different growth stages;

the soil influence neural network model is used for acquiring a large amount of soil data corresponding to soil influence nutrition coefficient values after soil influence nutrition coefficient values are set for pitaya nutrition according to soil data by a plant nutrition expert; the soil influence neural network model takes soil data as input and takes a soil influence nutrition coefficient as output;

the growth state network model is used for acquiring a large amount of growth image data corresponding to growth state influence nutrition coefficient values after the growth state influence nutrition coefficient values of the pitaya nutrition are set by plant nutrition experts according to growth image data; the growth state network model takes growth image data as input and takes growth state influence nutrition coefficients as output;

the meteorological influence neural network model is used for acquiring a large amount of meteorological data corresponding to meteorological influence nutrition coefficient values after the meteorological influence nutrition coefficient values of the pitaya nutrition are set by plant nutrition experts according to meteorological data; the meteorological influence neural network model takes meteorological data as input and a meteorological influence nutrition coefficient as output;

the irrigation influence neural network model is used for acquiring a large amount of irrigation data corresponding to irrigation influence nutrition coefficient values after the irrigation influence nutrition coefficient values of the pitaya nutrition are set according to irrigation data by plant nutrition experts; the irrigation influence neural network model takes irrigation data as input and takes an irrigation influence nutrition coefficient as output;

the fertilization influence network model is used for collecting a large amount of fertilization data corresponding to fertilization influence nutrition coefficient values after the fertilization influence nutrition coefficient values of the nutrition of the dragon fruits are set according to the fertilization data by plant nutrition experts; the fertilization influence network model takes fertilization data as input and fertilization influence nutrition coefficients as output;

the terrain influence network model is used for acquiring a large amount of terrain data corresponding to terrain influence nutrition coefficient values after the terrain influence nutrition coefficient values of the pitaya nutrition are set by a plant nutrition expert according to the terrain data; the terrain influence network model takes terrain data as input and takes a terrain influence nutrition coefficient as output.

In particular, the soil data comprises soil structure, soil air content, earth surface average temperature, different soil layer average humidity, soil pH value, soil conductivity, soil element composition, soil organic matter content and soil microorganism content.

Specifically, the growth image data includes an image of branches and leaves of the dragon fruit, an image of flowers of the dragon fruit, and an image of fruits of the dragon fruit.

In particular, the meteorological data includes a duration of illumination, an intensity of illumination, an amount of rainfall, an atmospheric temperature, an average temperature, a limit temperature, a wind speed, and an amount of evaporation.

In particular, the irrigation data includes irrigation water type, quality, irrigation time, irrigation usage and irrigation frequency.

In particular, the fertilization information comprises the type, nutrient content, fertilization time, fertilization amount, fertilization concentration and fertilization frequency of the fertilizer.

Specifically, the topographic information includes the terrain, the altitude, the gradient and the slope direction of the dragon fruit.

Particularly, the soil nutrient-affecting coefficient, the growth state nutrient-affecting coefficient, the weather nutrient-affecting coefficient, the irrigation nutrient-affecting coefficient, the fertilization nutrient-affecting coefficient and the terrain nutrient-affecting coefficient are respectively filtered, and the specific filtering method comprises the following steps: only one set of the same or similar coefficient data is retained, and the other sets of coefficient data are removed.

A system applying the nutrition configuration method for the plants in the whole period of dragon fruit planting comprises the following steps:

the soil data collection module is a soil moisture content instrument and is used for collecting soil data of the environment where the dragon fruits are located;

the image processing module is used for collecting growth image data of the dragon fruits and carrying out image processing;

the meteorological data collection module is a meteorological station and is used for collecting meteorological data of the environment where the dragon fruits are located;

the irrigation data collection module is equipment provided with an irrigation system and is used for collecting irrigation data in the dragon fruit planting process;

the fertilization data collection module is equipment provided with a fertilization system and is used for collecting irrigation data in the dragon fruit planting process;

the topographic data collecting module is used for collecting topographic data of the environment where the dragon fruits are located;

and the machine learning module is used for storing the soil influence neural network model, the growth state network model, the meteorological influence neural network model, the irrigation influence neural network model, the fertilization influence network model and the terrain influence network model and respectively outputting the soil influence nutrition coefficient, the growth state influence nutrition coefficient, the meteorological influence nutrition coefficient, the irrigation influence nutrition coefficient, the fertilization influence nutrition coefficient and the terrain influence nutrition coefficient.

The database module is used for storing a large amount of soil data, growth image data, meteorological data, irrigation data and topographic data;

the calculation module is used for respectively filtering the soil influence nutrition coefficient, the growth state influence nutrition coefficient, the meteorological influence nutrition coefficient, the irrigation influence nutrition coefficient, the fertilization influence nutrition coefficient and the terrain influence nutrition coefficient to construct a data set; dividing 70% of the data set to be used as model construction, and dividing 30% of the data set to be used as model verification; constructing a regression model of the ratio of the data set to the nutrition N, P, K of the pitaya to obtain the corresponding ratio N, P, K required by the pitaya in different growth stages;

the soil data collection module, the image processing module, the meteorological data collection module, the irrigation data collection module, the fertilization data collection module, the terrain data collection module and the database module are respectively connected with the machine learning module; the machine learning module is connected with the computing module.

In particular, still include the camera, the camera is installed on unmanned aerial vehicle and can wireless connection image processing module.

The invention has the beneficial effects that:

the invention relates to a machine learning-based pitaya planting full-period plant nutrition configuration method, which is a new identification method with high implementation efficiency, can completely replace the traditional pitaya nutrition configuration method, and simultaneously utilizes soil data, growth condition data, climate data, irrigation data, fertilization data and terrain data to analyze influence coefficients, calibrate and finally obtain corresponding N, P, K proportions required by pitaya in different growth stages. The method does not depend on experienced agricultural personnel any more, automatically identifies the plant nutrition formulas of the dragon fruits at different growth stages through machine learning, quickly improves nutrition supply judgment of the dragon fruits at different stages, pertinently improves nutrition and quality of the dragon fruits, has good pertinence and predictability, strong anti-interference capability, high accuracy and high diagnosis speed, and accordingly improves the satisfaction degree of consumers.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

FIG. 1 is a system block diagram of a method of practicing the embodiments of the invention;

FIG. 2 is a flow chart of a method of an embodiment of the present invention.

Detailed Description

The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings so that the advantages and features of the present invention can be more easily understood by those skilled in the art, and the scope of the present invention will be more clearly and clearly defined.

It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.

It is to be understood that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," "outer," and the like are used in a generic and descriptive sense only and not for purposes of limitation, the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," "outer," and the like are used in the generic and descriptive sense only and not for purposes of limitation, as the term is used in the generic and descriptive sense, and not for purposes of limitation, unless otherwise specified or implied, and the specific reference to a device or element is intended to be a reference to a particular element, structure, or component. Furthermore, the terms "first," "second," "third," and the like are used solely to distinguish one from another and are not to be construed as indicating or implying relative importance.

Furthermore, the terms "horizontal", "vertical", "overhang" and the like do not imply that the components are required to be absolutely horizontal or overhang, but may be slightly inclined. For example, "horizontal" merely means that the direction is more horizontal than "vertical" and does not mean that the structure must be perfectly horizontal, but may be slightly inclined.

In the description of the present invention, it should also be noted that, unless otherwise explicitly specified or limited, the terms "disposed," "mounted," "connected," and "connected" are to be construed broadly and may, for example, be fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.

As shown in fig. 1, the method for configuring plant nutrition for a full period of dragon fruit planting based on machine learning of the embodiment includes the following steps:

establishing a soil influence neural network model; and acquiring soil data of the environment where the dragon fruits are located and inputting the soil data into the soil influence neural network model to obtain the soil influence nutrition coefficient. The soil influence neural network model is used for acquiring a large amount of soil data corresponding to the soil influence nutrition coefficient value after the plant nutrition expert sets the soil influence nutrition coefficient value of the pitaya nutrition according to the soil data; the soil influence neural network model takes soil data as input and takes a soil influence nutrition coefficient as output. The soil data comprises soil structure, soil air content, earth surface average temperature, different soil layer average humidity, soil pH value, soil conductivity, soil element composition, soil organic matter content and soil microorganism content.

Establishing a growth state network model; and acquiring growth image data of the dragon fruits and inputting the growth image data into the growth state network model to obtain growth state influence nutrition coefficients. The growth state network model is used for acquiring a large amount of growth image data corresponding to growth state influence nutrition coefficient values after the growth state influence nutrition coefficient values of the pitaya nutrition are set by plant nutrition experts according to growth image data; the growth state network model takes growth image data as input and takes growth state influence nutrition coefficients as output. The growth image data includes images of branches and leaves of the dragon fruit, images of flowers of the dragon fruit, and images of fruits of the dragon fruit. And judging the nutrient abundance condition of the plants according to the colors, textures and shapes of branches, leaves, flowers and fruits of the dragon fruits.

Establishing a meteorological influence neural network model; acquiring meteorological data of the environment where the dragon fruits are located and inputting the meteorological data into a meteorological influence neural network model to obtain a meteorological influence nutrition coefficient; the meteorological influence neural network model is used for acquiring a large amount of meteorological data corresponding to meteorological influence nutrition coefficient values after the meteorological influence nutrition coefficient values of the pitaya nutrition are set by plant nutrition experts according to meteorological data; the meteorological influence neural network model takes meteorological data as input and meteorological influence nutrition coefficients as output. The meteorological data comprises illumination duration, illumination intensity, rainfall, atmospheric temperature, average temperature, limiting temperature, wind speed and evaporation capacity.

Establishing an irrigation influence neural network model; and acquiring irrigation data in the dragon fruit planting process, inputting the irrigation data into the irrigation influence neural network model, and acquiring an irrigation influence nutrition coefficient. The irrigation influence neural network model is used for acquiring a large amount of irrigation data corresponding to irrigation influence nutrition coefficient values after the irrigation influence nutrition coefficient values of the pitaya nutrition are set according to irrigation data by plant nutrition experts; the irrigation impact neural network model takes irrigation data as input and takes an irrigation impact nutrition coefficient as output. Irrigation data includes irrigation water type, quality, irrigation time, irrigation usage and irrigation frequency.

Establishing a fertilization influence network model; and acquiring fertilization data in the dragon fruit planting process, and inputting the fertilization data into the fertilization influence network model to obtain fertilization influence nutrition coefficients. The fertilization influence network model is used for collecting a large amount of fertilization data corresponding to fertilization influence nutrition coefficient values after the fertilization influence nutrition coefficient values of the nutrition of the dragon fruits are set according to the fertilization data by plant nutrition experts; the fertilization influence network model takes fertilization data as input and fertilization influence nutrition coefficients as output. The fertilization information comprises the type, nutrient content, fertilization time, fertilization amount, fertilization concentration and fertilization frequency of the fertilizer.

Establishing a terrain influence network model; acquiring topographic data of the environment of the dragon fruit and inputting the topographic data into a topographic influence network model to obtain a topographic influence nutrition coefficient; the terrain influence network model is used for acquiring a large amount of terrain data corresponding to terrain influence nutrition coefficient values after the terrain influence nutrition coefficient values of the pitaya nutrition are set by a plant nutrition expert according to the terrain data; the terrain influence network model takes terrain data as input and takes a terrain influence nutrition coefficient as output. The topographic information comprises the terrain, the altitude, the gradient and the slope direction of the dragon fruit.

And respectively filtering the soil influence nutrition coefficient, the growth state influence nutrition coefficient, the meteorological influence nutrition coefficient, the irrigation influence nutrition coefficient, the fertilization influence nutrition coefficient and the terrain influence nutrition coefficient to construct a data set. The specific method for filtering comprises the following steps: only one set of the same or similar coefficient data is retained, and the other sets of coefficient data are removed.

70% of the partitioned data set was used for model building and 30% of the data set was used for model validation.

And (3) constructing a regression model of the ratio of the data set to the nutrition N, P, K of the pitaya to obtain the corresponding N, P, K ratio required by the pitaya in different growth stages.

As shown in fig. 2, the system for applying the nutrition configuration method for plants in a full period of dragon fruit planting in the embodiment includes:

and the soil data collection module is a soil moisture content instrument and is used for collecting soil data of the environment where the dragon fruits are located.

And the image processing module is used for collecting growth image data of the dragon fruits and carrying out image processing. Still include right camera, the camera is installed on unmanned aerial vehicle and can wireless connection image processing module, will shoot image wireless transmission to image processing module.

And the meteorological data collection module is a meteorological station and is used for collecting meteorological data of the environment where the dragon fruits are located.

And the irrigation data collection module is equipment provided with an irrigation system and used for collecting irrigation data in the dragon fruit planting process.

And the fertilization data collection module is equipment provided with a fertilization system and used for collecting irrigation data in the dragon fruit planting process.

And the topographic data collecting module is used for collecting topographic data of the environment where the dragon fruits are located.

And the machine learning module is used for storing the soil influence neural network model, the growth state network model, the meteorological influence neural network model, the irrigation influence neural network model, the fertilization influence network model and the terrain influence network model and respectively outputting the soil influence nutrition coefficient, the growth state influence nutrition coefficient, the meteorological influence nutrition coefficient, the irrigation influence nutrition coefficient, the fertilization influence nutrition coefficient and the terrain influence nutrition coefficient.

And the database module is used for storing a large amount of soil data, growth image data, meteorological data, irrigation data and topographic data.

The calculation module is used for respectively filtering the soil influence nutrition coefficient, the growth state influence nutrition coefficient, the meteorological influence nutrition coefficient, the irrigation influence nutrition coefficient, the fertilization influence nutrition coefficient and the terrain influence nutrition coefficient to construct a data set; dividing 70% of the data set to be used as model construction, and dividing 30% of the data set to be used as model verification; and (3) constructing a regression model of the ratio of the data set to the nutrition N, P, K of the pitaya to obtain the corresponding N, P, K ratio required by the pitaya in different growth stages.

The soil data collection module, the image processing module, the meteorological data collection module, the irrigation data collection module, the fertilization data collection module, the terrain data collection module and the database module are respectively connected with the machine learning module. The machine learning module is connected with the calculation module.

Although the embodiments of the present invention have been described with reference to the accompanying drawings, various changes or modifications may be made by the patentees within the scope of the appended claims, and within the scope of the invention, as long as they do not exceed the scope of the invention described in the claims. The principles and embodiments of the present invention are explained herein using specific examples, which are presented only to assist in understanding the method and its core concepts of the present invention. It should be noted that there are no specific structures but a few objective structures due to the limited character expressions, and that those skilled in the art may make various improvements, decorations or changes without departing from the principle of the invention or may combine the above technical features in a suitable manner; such modifications, variations, combinations, or adaptations of the invention using its spirit and scope, as defined by the claims, may be directed to other uses and embodiments.