CN114661932A

CN114661932A - Information processing method, storage medium, and electronic device

Info

Publication number: CN114661932A
Application number: CN202210211167.7A
Authority: CN
Inventors: 董保华; 满远斌; 裘虬; 王海滨
Original assignee: Alibaba China Co Ltd
Current assignee: Alibaba China Co Ltd
Priority date: 2022-03-03
Filing date: 2022-03-03
Publication date: 2022-06-24

Abstract

The invention discloses an information processing method, a storage medium and an electronic device. Wherein, the method comprises the following steps: the method comprises the steps of obtaining internet of things data of crops to be monitored, wherein the internet of things data is used for determining growth information and environmental information of the crops to be monitored in real time; carrying out image recognition on the crop to be monitored to obtain a recognition result; acquiring current growth information corresponding to the crop to be monitored from a knowledge map associated with the crop to be monitored based on the data of the Internet of things and the identification result; and pushing current growth information. The invention solves the technical problems of low efficiency and poor flexibility caused by the dependence of the crop growth information identification and the relevant decision on the manual processing of production workers or technical experts in the related technology.

Description

Information processing method, storage medium, and electronic device

Technical Field

The present invention relates to the field of computer technologies, and in particular, to an information processing method, a storage medium, and an electronic device.

Background

In agriculture and forestry production, environmental information (such as soil environment, meteorological environment and the like) and growth information (such as phenological period, environmental information, pest information and the like) of crops are generally required to be identified and corresponding management decisions are made, and the existing identification and decision scheme mainly depends on production workers to carry out manual processing or recourse to technical experts to identify and give decision suggestions. As the production mode changes and the production scale increases, the diversity and complexity of crop production increases, and the identification and decision scheme increasingly depends on technical experts. However, technical expert resources are limited, and large-scale crop growth information identification and related decisions are difficult to perform in time.

In the related art, no effective solution is provided at present how to efficiently help production workers to accurately identify crop growth information and make relevant decisions.

Disclosure of Invention

The embodiment of the invention provides an information processing method, a storage medium and electronic equipment, which are used for at least solving the technical problems of low efficiency and poor flexibility caused by that crop growth information identification and related decisions depend on manual processing of production workers or technical experts in related technologies.

According to an aspect of an embodiment of the present invention, there is provided an information processing method including: the method comprises the steps of obtaining internet of things data of crops to be monitored, wherein the internet of things data is used for determining growth information and environmental information of the crops to be monitored in real time; carrying out image recognition on the crop to be monitored to obtain a recognition result; acquiring current growth information corresponding to the crop to be monitored from a knowledge map associated with the crop to be monitored based on the data of the Internet of things and the identification result; and pushing current growth information.

According to another aspect of the embodiments of the present invention, there is also provided an information processing method, including: the method comprises the steps of obtaining internet of things data of the kiwi fruit crops to be monitored, wherein the internet of things data is used for determining growth information and environmental information of the kiwi fruit crops to be monitored in real time; carrying out image recognition on the kiwi fruit crop to be monitored to obtain a recognition result; acquiring current growth information corresponding to the kiwi fruit crop to be monitored from a kiwi fruit knowledge map associated with the kiwi fruit crop to be monitored based on the data of the Internet of things and the identification result; and pushing current growth information.

According to another aspect of the embodiments of the present invention, there is also provided an information processing method, including: the method comprises the steps of obtaining internet of things data of a fruit tree to be monitored, wherein the internet of things data is used for determining growth information and environmental information of the fruit tree to be monitored in real time; carrying out image recognition on the fruit tree to be monitored to obtain a recognition result; based on the data of the Internet of things and the recognition result, obtaining current growth information corresponding to the fruit tree to be monitored from a fruit tree knowledge map associated with the fruit tree to be monitored; and pushing current growth information.

According to another aspect of the embodiments of the present invention, there is also provided an information processing apparatus including: the system comprises a first acquisition module, a second acquisition module and a monitoring module, wherein the first acquisition module is used for acquiring the Internet of things data of crops to be monitored, and the Internet of things data is used for determining the growth information and the environmental information of the crops to be monitored in real time; the identification module is used for carrying out image identification on the crops to be monitored to obtain an identification result; the second acquisition module is used for acquiring current growth information corresponding to the crop to be monitored from a knowledge map associated with the crop to be monitored based on the data of the Internet of things and the identification result; and the pushing module is used for pushing the current growth information.

According to another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium, where the computer-readable storage medium includes a stored program, and when the program runs, the apparatus on which the computer-readable storage medium is located is controlled to execute any one of the information processing methods.

According to another aspect of the embodiments of the present invention, there is also provided an electronic device, including: a processor; and a memory, connected to the processor, for providing instructions to the processor for processing the following processing steps: the method comprises the steps of obtaining internet of things data of crops to be monitored, wherein the internet of things data is used for determining growth information and environmental information of the crops to be monitored in real time; carrying out image recognition on the crop to be monitored to obtain a recognition result; acquiring current growth information corresponding to the crop to be monitored from a knowledge map associated with the crop to be monitored based on the data of the Internet of things and the identification result; and pushing current growth information.

In the embodiment of the invention, the data of the internet of things of the crop to be monitored is acquired, wherein the data of the internet of things is used for determining the growth information and the environmental information of the crop to be monitored in real time, the identification result is obtained by carrying out image identification on the crop to be monitored, the current growth information corresponding to the crop to be monitored is acquired from the knowledge map associated with the crop to be monitored based on the data of the internet of things and the identification result, and the current growth information is further pushed.

It is easy to notice that according to the embodiment of the application, the internet of things data of crops are obtained through the internet of things technology, the current growth information corresponding to the crops is obtained and pushed to a user by combining an image recognition method and a crop related knowledge graph, and the purposes of determining crop damage and prevention and control strategies based on image recognition and knowledge graph reasoning are achieved, so that the technical effects of reducing dependence on technical experts and improving efficiency and flexibility in crop growth information recognition and related decisions are achieved, and the technical problems of low efficiency and poor flexibility caused by dependence on production workers or manual processing of the technical experts in the crop growth information recognition and the related decisions in the related technology are solved.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:

fig. 1 shows a hardware configuration block diagram of a computer terminal (or mobile device) for implementing an information processing method;

FIG. 2 is a flow chart of a method of information processing according to an embodiment of the present invention;

FIG. 3 is a flow chart of another information processing method according to an embodiment of the present invention;

fig. 4 is a schematic diagram of the structure of an alternative intelligent kiwi planting system according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of an alternative deep network model training process according to an embodiment of the invention;

FIG. 6 is a flow chart of another information processing method according to an embodiment of the present invention;

fig. 7 is a schematic configuration diagram of an information processing apparatus according to an embodiment of the present invention;

fig. 8 is a schematic configuration diagram of another information processing apparatus according to an embodiment of the present invention;

fig. 9 is a schematic configuration diagram of another information processing apparatus according to an embodiment of the present invention;

fig. 10 is a block diagram of another computer terminal according to an embodiment of the present invention.

Detailed Description

In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

First, some terms or terms appearing in the description of the embodiments of the present application are applicable to the following explanations:

internet of Things (Internet of Things, IOT for short): the intelligent sensing, identification and management of objects and processes are realized by various devices and technologies of various information sensors, radio frequency identification technologies, global positioning systems, infrared sensors and laser scanners, acquiring any object or process needing video acquisition and connection interaction in real time, acquiring various required information of sound, light, heat, electricity, mechanics, chemistry, biology, position and the like, and realizing ubiquitous connection of the objects and people through various possible network accesses.

Image recognition: refers to a technique of processing, analyzing and understanding an image with a computer to recognize various patterns of objects and objects. Image recognition is a practical application of deep learning algorithms.

Knowledge graph: a knowledge graph is a structured semantic knowledge base used to quickly describe concepts and their interrelationships in the physical world. The knowledge graph obtains simple and clear entity, relation and entity triples by effectively processing, processing and integrating data of the complicated and intricate documents, and finally, quick response and reasoning on knowledge are realized by aggregating a large amount of knowledge. Due to the strong semantic processing capability and the open interconnection capability, the knowledge graph is widely applied to the fields of intelligent search, intelligent question answering, personalized recommendation, information analysis, fraud prevention and the like.

Intelligent early warning: the method is a process of collecting related data information according to the characteristics of a research object, monitoring the abnormal movement and the variation trend of the research object, evaluating the degree of the state value of the research object deviating from the early warning value through an intelligent algorithm, and sending out an early warning signal according to the degree.

Example 1

There is also provided, in accordance with an embodiment of the present invention, an information processing method embodiment, it should be noted that the steps illustrated in the flowchart of the accompanying drawings may be performed in a computer system such as a set of computer-executable instructions, and that, although a logical order is illustrated in the flowchart, in some cases, the steps illustrated or described may be performed in an order different than here.

The method provided by the first embodiment of the present application may be executed in a mobile terminal, a computer terminal, or a similar computing device. Fig. 1 shows a hardware configuration block diagram of a computer terminal (or mobile device) for implementing an information processing method. As shown in fig. 1, the computer terminal 10 (or mobile device 10) may include one or more (shown as 102a, 102b, … …, 102 n) processors 102 (the processors 102 may include, but are not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. Besides, the method can also comprise the following steps: a display, an input/output interface (I/O interface), a Universal Serial BUS (USB) port (which may be included as one of the ports of the BUS), a network interface, a power source, and/or a camera. It will be understood by those skilled in the art that the structure shown in fig. 1 is only an illustration and is not intended to limit the structure of the electronic device. For example, the computer terminal 10 may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.

It should be noted that the one or more processors 102 and/or other data processing circuitry described above may be referred to generally herein as "data processing circuitry". The data processing circuitry may be embodied in whole or in part in software, hardware, firmware, or any combination thereof. Further, the data processing circuit may be a single stand-alone processing module, or incorporated in whole or in part into any of the other elements in the computer terminal 10 (or mobile device). As referred to in the embodiments of the application, the data processing circuit acts as a processor control (e.g. selection of a variable resistance termination path connected to the interface).

The memory 104 may be used to store software programs and modules of application software, such as program instructions/data storage devices corresponding to the information processing method in the embodiment of the present invention, and the processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, implementing the information processing method described above. The memory 104 may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory located remotely from the processor 102, which may be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The transmission device 106 is used for receiving or transmitting data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network adapter (NIC) that can be connected to other Network devices through a base station to communicate with the internet. In one example, the transmission device 106 can be a Radio Frequency (RF) module, which is used to communicate with the internet in a wireless manner.

The display may be, for example, a touch screen type Liquid Crystal Display (LCD) that may enable a user to interact with a user interface of the computer terminal 10 (or mobile device).

It should be noted here that in some alternative embodiments, the computer device (or mobile device) shown in fig. 1 described above may include hardware elements (including circuitry), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that fig. 1 is only one example of a particular specific example and is intended to illustrate the types of components that may be present in the computer device (or mobile device) described above.

Under the above operating environment, the present application provides an information processing method as shown in fig. 2. Fig. 2 is a flowchart of an information processing method according to an embodiment of the present invention, as shown in fig. 2, the information processing method including:

step S202, obtaining Internet of things data of the crops to be monitored, wherein the Internet of things data is used for determining growth information and environmental information of the crops to be monitored in real time;

step S204, carrying out image recognition on the crop to be monitored to obtain a recognition result;

step S206, acquiring current growth information corresponding to the crop to be monitored from a knowledge map associated with the crop to be monitored based on the data of the Internet of things and the identification result;

and step S208, pushing the current growth information.

Alternatively, the crop to be monitored may be an agricultural crop or a forestry crop. Agricultural crops may include food crops (e.g., rice, corn, beans, potatoes, wheat, etc.) and cash crops (e.g., oil crops, vegetable crops, medicinal crops, melon and fruit crops, etc.), among others. Forestry crops may include industrial raw forest crops (e.g., camellia oleifera, sapium sebiferum, tung oil tree, sumac, etc.), medicinal forest crops, spice trees, and the like.

Optionally, the data of the internet of things may include various information of sound, light, heat, electricity, mechanics, chemistry, biology, location, and the like of the crop to be monitored. The data of the internet of things can be used for determining the growth information and the environmental information of the crops to be monitored in real time, wherein the growth information can comprise the height of the crops, the shape of the crops, the color of the crops, the water content of the crops and the like; the environmental information may include soil humidity, air temperature, soil ph, illumination intensity, average daily illumination duration, content of each element in the soil, and the like.

Optionally, the obtaining of the internet of things data of the crop to be monitored may be: the relevant internet of things data is collected in real time by using various internet of things devices or internet of things technologies (such as information sensors, radio frequency identification technologies, global positioning systems, infrared sensors, laser scanners and the like).

Optionally, based on an image recognition technology, image recognition may be performed on the crop to be monitored, so as to obtain the recognition result. Based on the internet of things data and the identification result of the crop to be monitored, the current growth information corresponding to the crop to be monitored can be obtained from the knowledge map associated with the crop to be monitored. Further, the acquired current growth information may be pushed to the user to guide production.

Optionally, the current growth information corresponding to the crop to be monitored may be used to describe a growth condition of the crop to be monitored at the current time. For example: the current growth information may include: information of the current phenological period (including phenological period type, degree of coincidence between the current phenological period and an expected phenological period, and the like), information of current plant diseases and insect pests (including plant disease and insect pest type, plant disease and insect pest degree, and the like), and the like.

Optionally, pest and disease information in the current growth information may include disease information and pest and disease information. The disease may also include biological diseases (e.g., fungal diseases, bacterial diseases, viral diseases, etc.) and non-biological diseases (e.g., drought-causing diseases, waterlogging-causing diseases, cold-causing diseases, nutrient-imbalance diseases, etc.). The insect pest usually refers to the pest (such as beetle, locust, moth and butterfly larva, aphid, mite, fly, plant bug, etc.) to destroy the crop to be monitored.

Specifically, other method steps included in the information processing method may also refer to the further description below for the embodiments of the present application, which are not described herein again.

It should be noted that the information processing method provided by the invention can be applied to any application scene relating to monitoring of the internet of things, hazard identification, hazard early warning and prevention and control strategy recommendation in the fields of agriculture, forestry, natural resources and ecological environment, water conservancy, weather, secondary disasters and the like.

The information processing method of the above-described embodiment is further described below.

In an alternative embodiment, in step S204, the image recognition is performed on the crop to be monitored to obtain a recognition result, and the method includes the following steps:

s241, shooting a crop to be monitored to obtain a picture to be processed;

and step S242, analyzing the picture to be processed by using the target crop identification model to obtain an identification result, wherein the identification result is used for determining the crop growth condition presented in the picture to be processed.

In the optional embodiment, the to-be-processed picture may be obtained by shooting the to-be-monitored crop by using an image capturing device (e.g., a camera) of the internet of things.

In the above alternative embodiment, the target crop recognition model may be an artificial neural network for deep learning, and the target crop recognition model may be trained through machine learning using a plurality of sets of training data. Each set of training data in the plurality of sets of training data includes: training pictures and growth status data identified in the training pictures. The training picture can be preset by technicians according to actual application scenes, and can also be updated in real time in the machine learning training process. The growth state data identified in the training picture may include: phenological period data, pest data (such as pest type, pest location, pest degree, etc.), etc.

In the above optional embodiment, the target crop recognition model is used to analyze the to-be-processed picture, so as to obtain the recognition result. The recognition result can be used for determining the crop growth condition presented in the picture to be processed. For example: the crop growth condition can be the degree of coincidence between the current phenological period and the expected phenological period, the presence or absence of a target type of hazard, and the like. The target type of hazard may be a pest, or may be one or more designated hazards within a pest. In a practical application scenario, the target type hazard may be preset by a technician.

In an alternative embodiment, the current growth information includes at least one of: treat the pest and disease information of monitoring crop, wherein, pest and disease information includes: whether the crop to be monitored has plant diseases and insect pests or not and the degree of the plant diseases and insect pests of the crop to be monitored; whether the growth state of the crops to be monitored meets expectations or not; the growth stage of the crop to be monitored; and (3) target control strategy of the crops to be monitored.

Optionally, the pest and disease information of the crop to be monitored may include: whether the crop to be monitored has the pest (when the pest exists, the pest type, the pest occurrence position and the like can also be included), the pest degree (such as slight, medium and serious) of the crop to be monitored, and the like.

Optionally, whether the growth state of the crop to be monitored is expected or not may include: whether the growth stage of the crops to be monitored is in accordance with the expectation (whether the growth stage is advanced or delayed), whether the growth form of the crops to be monitored (such as plant height, stem thickness, leaf size and the like) is in accordance with the expectation, and the like.

Optionally, the growth stage of the crop to be monitored may include: germination stage, seedling stage, green stage, flowering stage, fruit stage, mature stage, etc.

Alternatively, the target control strategy of the crops to be monitored can be a series of control measures aiming at the pest and disease damage information of the crops to be monitored and implementation strategies thereof. The control strategy may include: agricultural control strategies (e.g., deep ploughing, weeding, trimming, fertilizing, breeding, reseeding, etc.), biological control strategies (e.g., methods utilizing beneficial insects, beneficial birds, frogs, microorganisms, methods causing infertility of pests, etc.), physical control strategies (e.g., setting insect nets, light trapping, plasma disinfection, radiation control, etc.), chemical control strategies (e.g., applying chemical pesticides such as insecticides, bactericides, bird repellents, etc.).

In the above optional embodiment, the current growth information corresponding to the crop to be monitored, which is acquired from the knowledge graph based on the data of the internet of things and the recognition result, may include at least one of the following information: the pest and disease information of the crops to be monitored, whether the growth state of the crops to be monitored meets expectations, the growth stage of the crops to be monitored, and the target control strategy of the crops to be monitored.

In an optional embodiment, the information processing method may further include the following method steps:

step S210, carrying out classification training on a plurality of pre-training models respectively by adopting a plurality of groups of training data to obtain a training result corresponding to each pre-training model in the plurality of pre-training models, wherein the training result is used for determining the classification performance and the generalization performance of each pre-training model in the plurality of pre-training models;

step S212, selecting an initial depth network model from a plurality of pre-training models based on a training result;

step S214, carrying out classification training on the initial deep network model by utilizing a plurality of groups of training data to obtain a first loss, and carrying out characteristic reconstruction training on the initial deep network model by utilizing a plurality of groups of training data to obtain a second loss;

step S216, determining a target loss based on the first loss and the second loss;

and step S218, adjusting the network parameters of the initial deep network model based on the target loss to obtain a target crop identification model.

In the above alternative embodiment, the plurality of pre-trained models may be a plurality of convolutional neural networks that are pre-trained. Each set of training data in the plurality of sets of training data includes: the training pictures and the hazard data identified in the training pictures. The multiple groups of training data are adopted to respectively carry out classification training on the multiple pre-training models, and a training result corresponding to each pre-training model in the multiple pre-training models can be obtained. The training results may be used to determine a classification performance and a generalization performance of each of the plurality of pre-trained models.

In particular, the classification performance may be a performance for representing the neural network accurately classifying the training data. Generalization performance may be the performance used to represent a reasonable output of the neural network on data not encountered during training.

In the above optional embodiment, based on the training result corresponding to each of the plurality of pre-training models, an initial deep network model may be selected from the plurality of pre-training models. The initial deep network model may be selected as follows: and selecting a pre-training model of which the classification performance and the generalization performance meet preset conditions (for example, the classification performance is optimal, the generalization performance is optimal, and the like) from the plurality of pre-training models as an initial deep network model.

In the optional embodiment, the first loss may be obtained by performing classification training on the initial deep network model by using the plurality of sets of training data. And performing feature reconstruction training on the initial deep network model by using the plurality of groups of training data to obtain the second loss. Based on the first loss and the second loss, the target loss may be determined. The target loss may be a total loss of the training process.

In the above optional embodiment, through the target loss, the network parameters of the initial deep network model may be adjusted, so as to obtain the target crop identification model. The target crop recognition model can be used for analyzing the to-be-processed picture to obtain a recognition result. In particular, the principle of adjusting the network parameters may be: the total loss of training of the initial deep network is minimized as much as possible.

In an optional embodiment, the information processing method further includes the following method steps:

step S220, collecting first target data and second target data, wherein the first target data is machine data collected from the Internet and associated with crops to be monitored, and the second target data is expert experience data associated with the crops to be monitored;

step S222, constructing a knowledge graph by using the first target data and the second target data.

In the above alternative embodiment, the first target data may be machine data collected from the internet in association with the crop to be monitored. The machine data may include structured data obtained from a database or a particular website, internet of things data obtained using internet of things devices, machine log data, and the like.

In the above alternative embodiment, the second target data may be expert experience data associated with the crop to be monitored. The expert empirical data may be the expertise that is discussed and recorded by a technician in communication with an associated technical expert. The expert experience data may include expertise in the overall process of crop growth to be monitored, knowledge of pests associated with the crop to be monitored, knowledge of the experience of controlling the pests of the crop to be monitored, and the like.

In the above alternative embodiment, first and second objective data are collected, and a knowledge-map associated with the crop to be monitored is constructed using the first and second objective data.

In an optional embodiment, in step S206, when the current growth information includes pest and disease information of the crop to be monitored and a target control strategy of the crop to be monitored, the current growth information is acquired from the knowledge graph based on the data of the internet of things and the identification result, and the method includes the following steps:

step S261, acquiring pest and disease information from the knowledge graph based on the recognition result;

step S262, determining candidate prevention and control means through pest and disease information;

and step S263, determining a target prevention strategy from the candidate prevention and treatment means by using the data of the Internet of things.

In the above optional embodiment, the recognition result is obtained by analyzing the to-be-processed picture acquired by the internet of things device by using the target crop recognition model. When the current growth information includes pest information of the crop to be monitored and a target control strategy of the crop to be monitored, the pest information can be acquired from a knowledge map corresponding to the crop to be monitored based on the identification result. The pest information may include the type of pest, the degree of pest, and the like.

Alternatively, the candidate control means may comprise control means for the pest at a plurality of growth stages (e.g., germination stage, seedling stage, green stage, flowering stage, fruiting stage, maturation stage, etc.) of the crop to be monitored.

In the above optional embodiment, the data of the internet of things may be data acquired by the equipment of the internet of things in real time, and the data of the internet of things may be used to determine growth information and environmental information of the crop to be monitored in real time, where the growth information may include crop height, crop form, crop color, crop water content, and the like; the environmental information may include soil humidity, air temperature, soil ph, illumination intensity, average daily illumination duration, content of each element in the soil, and the like.

In the above optional embodiment, a target prevention and control strategy may be determined from the candidate prevention and control means by using the internet of things data. The target control strategy may be one or more of a plurality of control means of candidate control means that match the growth information and environmental information of the crop to be monitored, and an implementation strategy of the control means.

In an alternative embodiment, in step S263, the target control strategy is determined from the candidate control means by using the data of the internet of things, and the method includes the following steps:

step S2631, determining type information of crops to be monitored, information of growth period stages and information of crop growth environments by using data of the Internet of things;

step S2632, a target control strategy is determined based on the type information, the stage information of the growth cycle and the crop growth environment information.

Optionally, the type information of the crop to be monitored may be used to indicate the type of the crop to be monitored, for example: the type information can be specific crop types (such as kiwi fruit, apple, rice, etc.), and can also be crop types (such as grain, fruit tree, fresh flower, etc.).

Optionally, the information about the stage of the growth cycle of the crop to be monitored may be the current growth stage. For example: germination stage, seedling stage, green stage, flowering stage, fruit stage, mature stage, etc.

Optionally, the crop growth environment information of the crop to be monitored may include soil humidity, air temperature, soil ph, illumination intensity, daily average illumination duration, content of each element in the soil, and the like.

In the optional embodiment, by using the data of the internet of things, the type information of the crop to be monitored, the information of the stage of the growth cycle, and the information of the growth environment of the crop can be determined. Based on the type information, the stage information of the growth cycle and the crop growth environment information, a target control strategy can be determined from candidate control means.

Optionally, an optimal pest control means can be derived through a relationship path reasoning method, so as to obtain the target control strategy. For example: reasoning is carried out through a knowledge map based on a relation path of 'pest and disease information, phenological period, soil conditions and meteorological conditions', and one or more prevention and control means which are optimally matched with all relations in the relation path are deduced.

In an alternative embodiment, in step S208, the current growth information is pushed, including one of the following method steps:

step S281, pushing the current growth information according to the preset business rule;

step S282, pushing current growth information based on a determination result of a target determination model, where the target determination model is used to analyze the data of the internet of things to determine the determination result.

In the above alternative embodiment, the preset service rule may be a rule preset by a technician according to an actual application scenario. The preset business rule can be used for stipulating the time, condition, mode and the like of pushing the current growth information.

In the above optional embodiment, the target determination model may be used to analyze the data of the internet of things to determine the determination result. The data of the internet of things can be used for determining the growth information and the environmental information of the crop to be monitored in real time. The decision may be used to determine whether and how to push the current growth information. The target determination model may be a decision tree model that may be constructed based on internet of things data, pest information, control strategy information, machine data, expert experience data, etc. of the crop to be monitored.

In the above alternative embodiment, one of the methods for pushing the current growth information may be pushing according to the preset business rule. One of the methods of pushing the current growth information may also be pushing based on the determination result of the target determination model.

It is easy to notice that the pest information of the crop to be monitored can be found in real time based on the internet of things equipment, and the pest degree identification and related decision corresponding to the pest information usually need the guidance of experts. However, the production workers cannot or cannot be guided in time due to the limited expert resources, the messy and complicated internet information and the like. By the method provided by the embodiment, machine data of an objective world and expert experience data can be combined, a knowledge graph is constructed, a knowledge graph calculation engine service is provided, and then the pest and disease damage degree and the control strategy corresponding to pest and disease damage information are obtained to be pushed to production workers. By the method, the dependence on technical experts in the crop production process can be reduced, the efficiency and the flexibility are improved, and the production is served.

It should be noted that the present invention is focused on: the method comprises the steps of combining machine data of an objective world with expert experience data, constructing a knowledge map, providing knowledge map calculation engine service, collecting crop data to be monitored in real time through Internet of things data, identifying current growth information corresponding to crops by using an image identification technology, and reasoning pest and disease damage degree and prevention and control strategies corresponding to crop pest and disease damage information according to the knowledge map calculation engine service to push the pest and disease damage degree and prevention and control strategies to production workers. In the process, the data acquisition mode of the internet of things can be any relevant realizable mode, the image recognition algorithm can be any relevant realizable algorithm, and the construction and reasoning method of the knowledge graph can be any relevant realizable method.

In an alternative embodiment, a graphical user interface is provided through the terminal device, the content displayed by the graphical user interface at least partially includes a target crop monitoring scene, and the information processing method further includes the following method steps:

step S291, displaying a plurality of plots and a plurality of types of candidate crops planted on the plots in the graphical user interface;

step S292, responding to a first touch operation acting on a graphical user interface, and determining a crop to be monitored from multiple types of candidate crops;

step S293, responding to a second touch operation acting on the graphical user interface, and selecting a target crop identification model and a target pushing mode of the crop to be monitored;

step S294, responding to a third touch operation acting on the graphical user interface, obtaining a recognition result by using the target crop recognition model, obtaining current growth information corresponding to the crop to be monitored from the knowledge graph based on the internet of things data and the recognition result, and displaying the current growth information in the graphical user interface according to a target pushing manner.

In the above optional embodiment, the user may at least partially obtain the target crop monitoring scene through graphical user interface content displayed by the terminal device. The smallest target of the method provided by the invention can be an agricultural land parcel, and the graphical user interface can display a plurality of land parcels and can also display a plurality of types of candidate crops planted on each land parcel in the plurality of land parcels.

Optionally, in the above graphical user interface, when multiple types of candidate crops planted in each of the multiple plots are displayed, the candidate control strategies corresponding to each type of candidate crops in the multiple types of candidate crops may also be displayed at the same time. For example: in a certain graphical user interface, a user can select 'display crops' or 'display crops and strategies' according to requirements. When the user selects 'show crops', the crops such as kiwi fruits, apples, oranges and the like planted in a plurality of plots are displayed on the graphical user interface; when the user selects 'display crops and strategies', kiwifruit planted in a plurality of plots and planting control strategies thereof, apples and planting control strategies thereof, oranges and planting control strategies thereof and the like are displayed on the graphical user interface.

Optionally, in the graphical user interface, the user may perform a first touch operation on multiple types of candidate crops shown in the graphical user interface. The user can touch one of the candidate crops in the graphical user interface to determine the crop to be monitored.

Optionally, the user may perform a second touch operation on the graphical user interface. The user can determine a target crop recognition model to be used by the crop to be monitored through at least one of the plurality of deep network models displayed in the touch graphic user interface, and the user can determine a target pushing mode to be used through at least one of the plurality of pushing modes displayed in the touch graphic user interface.

Optionally, the user may perform a third touch operation on the graphical user interface. The user can touch an 'identification' button, a 'recommendation' button, a 'generation' button or a 'display' button and the like in the graphical user interface so that the target crop identification model obtains an identification result, obtains current growth information corresponding to the crop to be monitored from the knowledge graph based on the data of the internet of things and the identification result, and displays the current growth information in the graphical user interface according to a target pushing mode.

In particular, the first touch operation, the second touch operation, and the third touch operation may be operations in which a user touches a display screen of the terminal device with a finger and touches the terminal device. The touch operation may include single-point touch, multi-point touch, wherein the touch operation of each touch point may include clicking, long pressing, re-pressing, swiping, and the like. The first touch operation, the second touch operation, and the third touch operation may also be touch operations implemented by an input device such as a mouse and a keyboard.

step S295, responding to the editing operation acting on the target prevention and control strategy, and acquiring a prevention and control operation result corresponding to the target prevention and control strategy;

in step S296, a target prevention and control strategy is optimized based on the prevention and control operation result.

In the above optional embodiment, the target control strategy may be one or more control means that match the growth information and the environmental information of the crop to be monitored, which are determined from a plurality of control means of candidate control means according to data of the internet of things, and an implementation strategy of the control means. The user can edit the target prevention and control strategy (such as modifying part of information in the prevention and control strategy, and completely re-defining the prevention and control strategy).

In the above optional embodiment, when an editing operation of the target prevention and control policy by the user is received, a prevention and control operation result corresponding to the target prevention and control policy is obtained. Based on the control operation result, a target control strategy can be optimized.

Optionally, when the user edits the target prevention and treatment policy, the edited content may be collected and stored. The edited content may be used to dynamically adjust the objective control strategy. For example: the dynamic adjustment can be based on the growth information, environmental information and the like of crops, and the control strategy is adjusted in real time according to the implementation situation and the editing content of the control strategy by the user.

It is easy to notice that the method provided by the embodiment can support the user to edit the crop damage control strategy pushed by the system, which is beneficial for the user to use the method more flexibly and conveniently, and is also beneficial for the knowledge-graph engine in the method to adjust the construction method, the calculation model and the like according to the user editing operation feedback.

Under the above operating environment, the present invention provides an information processing method as shown in fig. 3. Fig. 3 is a flowchart of another information processing method according to an embodiment of the present invention, as shown in fig. 3, the information processing method including:

step S302, Internet of things data of the agricultural crops to be monitored are obtained, wherein the Internet of things data are used for determining growth information and environmental information of the agricultural crops to be monitored in real time;

step S304, carrying out image recognition on the kiwi fruit crop to be monitored to obtain a recognition result;

step S306, acquiring current growth information corresponding to the kiwi fruit crops to be monitored from the kiwi fruit knowledge map associated with the kiwi fruit crops to be monitored based on the Internet of things data and the identification result;

and step S308, pushing the current growth information.

Optionally, the internet of things data may include various information such as sound, light, heat, electricity, mechanics, chemistry, biology, location, etc. of the kiwi fruit crop to be monitored. The data of the internet of things can be used for determining the growth information and the environmental information of the kiwi fruit crops to be monitored in real time, wherein the growth information can comprise the height of the kiwi fruit crops, the shape of the kiwi fruit crops, the color of the kiwi fruit crops, the water content of the kiwi fruit crops and the like; the environmental information may include soil humidity, air temperature, soil ph, illumination intensity, average daily illumination time, content of each element in the soil, and the like.

Optionally, the internet of things data for acquiring the kiwi fruit crops to be monitored may be: the relevant internet of things data is collected in real time by using various internet of things devices or internet of things technologies (such as information sensors, radio frequency identification technologies, global positioning systems, infrared sensors, laser scanners and the like).

Optionally, based on an image recognition technology, image recognition can be performed on the kiwi fruit crop to be monitored, so as to obtain the recognition result. Based on the internet of things data and the identification result of the kiwi fruit crop to be monitored, the current growth information corresponding to the kiwi fruit crop to be monitored can be acquired from the knowledge map associated with the kiwi fruit crop to be monitored. Further, the acquired current growth information may be pushed to the user to guide production.

Optionally, the current growth information corresponding to the kiwi fruit crop to be monitored may be used to describe the growth condition of the kiwi fruit crop to be monitored at the current time. For example: the current growth information may include: information of the current phenological period (including phenological period type, degree of coincidence between the current phenological period and an expected phenological period, and the like), information of current plant diseases and insect pests (including plant disease and insect pest type, plant disease and insect pest degree, and the like), and the like.

Optionally, the pest information in the current growth information may include disease information and pest information. The diseases can also include biological diseases (such as fungal diseases, bacterial diseases, viral diseases, etc.) and non-biological diseases (such as drought-caused diseases, waterlogging-caused diseases, severe cold diseases, nutrient imbalance diseases, etc.). The insect pest usually refers to the pest (such as beetle, locust, moth larva, aphid, mite, fly, stink bug, etc.) to damage the kiwi fruit crop to be monitored.

In the embodiment of the invention, the internet of things data of the kiwi fruit crop to be monitored is acquired, wherein the internet of things data is used for determining the growth information and the environmental information of the kiwi fruit crop to be monitored in real time, the image of the kiwi fruit crop to be monitored is identified to obtain an identification result, the current growth information corresponding to the kiwi fruit crop to be monitored is acquired from the kiwi fruit knowledge map associated with the kiwi fruit crop to be monitored based on the internet of things data and the identification result, and the current growth information is pushed.

It is easy to notice that, through the embodiment of the application, the internet of things data of the kiwi fruit crops is obtained through the internet of things technology, the image recognition method and the kiwi fruit knowledge base related to the kiwi fruit crops are combined, the current growth information and the related decision corresponding to the kiwi fruit crops are obtained and pushed to users, and the purposes of determining the harm of the kiwi fruit crops and the control strategy based on the image recognition and the kiwi fruit knowledge base reasoning are achieved, so that the technical effects of reducing the dependence on kiwi fruit technical experts in the kiwi fruit crop growth information recognition and the related decision and improving the efficiency and the flexibility are achieved, and the technical problems of low efficiency and poor flexibility caused by the fact that farmers or kiwi fruit technical experts rely on the kiwi fruit crop growth information recognition and the related decision in the related technology are solved.

It should be noted that the information processing method provided by the invention can be applied to any application scene relating to monitoring of the internet of things, hazard identification, hazard early warning and prevention and control strategy recommendation in the field of kiwi fruits, but not limited thereto.

In an alternative embodiment, in step S304, performing image recognition on the kiwi fruit crop to be monitored to obtain a recognition result, including the following steps:

step S341, shooting the kiwi fruit crops to be monitored to obtain pictures to be processed;

and step S342, analyzing the picture to be processed by using the target kiwi fruit crop identification model to obtain an identification result, wherein the identification result is used for determining the growth condition of the kiwi fruit presented in the picture to be processed.

In the above optional embodiment, the to-be-processed picture may be obtained by shooting the to-be-monitored kiwi fruit crop with an internet-of-things image acquisition device (such as a camera).

In the above-described alternative embodiment, the target kiwi fruit crop recognition model may be an artificial neural network for deep learning, and the target kiwi fruit crop recognition model may be obtained by machine learning training using multiple sets of training data. Each set of training data in the plurality of sets of training data includes: training pictures and growth status data identified in the training pictures. The training picture can be preset by a technician according to an actual application scene, and can also be updated in real time in the machine learning training process. The growth state data identified in the training picture may include: phenological period data, pest data (such as pest type, pest position, pest degree, etc.), etc.

In the above optional embodiment, the target kiwi fruit crop recognition model is used to analyze the to-be-processed picture, so that the recognition result can be obtained. The identification result can be used for determining the growth condition of the kiwi fruits represented in the picture to be processed. For example: the growth condition of the kiwi fruits can be the consistent degree of the current phenological period and the expected phenological period, whether target type hazards exist or not and the like. The target type of hazard may be a pest, and may be one or more designated hazards within the pest. In a practical application scenario, the target type hazard may be preset by a technician.

Fig. 4 is a schematic diagram of the structure of an alternative intelligent kiwi fruit planting system according to an embodiment of the present invention, and as shown in fig. 4, the intelligent kiwi fruit planting system may include an internet of things real-time acquisition system. The real-time collection system of the Internet of things can collect picture information, soil condition information and climate condition information (equivalent to the data of the Internet of things). The picture information may be a picture showing the kiwi plant to be monitored. The picture information, the soil condition information and the climate condition information acquired by the real-time acquisition system of the internet of things can be used for determining the growth information and the environmental information of the kiwi fruits to be monitored in real time.

Still as shown in fig. 4, the kiwifruit intelligent planting system may further include an image recognition system. The image recognition system can be used for preprocessing (for example, screening, cutting, format adjustment, visual adjustment and the like) the picture information (equivalent to the picture to be processed) acquired by the real-time acquisition system of the Internet of things. The image recognition system can also perform feature extraction based on the kiwi plant image contained in the preprocessed picture to obtain kiwi plant features.

Still as shown in fig. 4, the image recognition system in the intelligent kiwi fruit planting system can also use an image recognition algorithm in a deep network model (equivalent to the above-mentioned target kiwi fruit crop recognition model) of machine learning training to analyze the extracted kiwi fruit plant characteristics, so as to obtain a kiwi fruit plant recognition result. The kiwi fruit plant identification result can be used for determining whether plant diseases and insect pests (equivalent to the target type damage) exist in the pictures containing the kiwi fruit plants acquired by the real-time acquisition system of the internet of things.

It should be noted that, the deep network model in the image recognition system may be obtained by machine learning training using multiple sets of kiwi fruit pest training data. Every group kiwi fruit plant diseases and insect pests training data can include in this multiunit kiwi fruit plant diseases and insect pests training data: the method comprises the following steps of (1) obtaining a picture of the kiwi fruit plant (for example, the picture can display at least one of the following parts of the kiwi fruit plant, namely a trunk, a front surface of a leaf, a back surface of the leaf, a root, a flower, a bud and a fruit) and pest data (for example, pest type and pest degree) identified on the picture.

Particularly, above-mentioned multiunit kiwi fruit plant diseases and insect pests training data also can include the training data of healthy kiwi fruit plant, shows healthy kiwi fruit plant in the plant picture among this training data, and the plant diseases and insect pests data of sign is empty on this picture.

In an alternative embodiment, the current growth information includes at least one of: treat the pest and disease information of monitoring kiwi fruit crop, wherein, the pest and disease information includes: whether the kiwi fruit crops have diseases and insect pests or not is to be monitored, and the disease and insect pest degree of the kiwi fruit crops is to be monitored; whether the growth state of the kiwi fruit crops to be monitored meets expectations or not; the growth stage of the kiwi fruit crop to be monitored; and (3) a target control strategy of the kiwi fruit crops to be monitored.

Optionally, the above-mentioned pest and disease information of kiwi fruit crop to be monitored can include: whether the kiwi fruit crops to be monitored have diseases and insect pests (when the diseases and insect pests exist, the types of the diseases and insect pests, the occurrence positions of the diseases and the insect pests and the like can also be included), the disease and insect pest degrees (such as slight, medium and serious) of the kiwi fruit crops to be monitored and the like.

Optionally, whether the growth status of the kiwi fruit crop to be monitored meets expectations or not may include: whether the growth stage of the kiwi fruit crop to be monitored meets the expectation (whether the growth stage is advanced or delayed), whether the growth form (such as plant height, stem thickness, leaf size and the like) of the kiwi fruit crop to be monitored meets the expectation, and the like.

Optionally, the growth stage of the kiwi fruit crop to be monitored may include: germination stage, seedling stage, green stage, flowering stage, fruit stage, mature stage, etc.

Optionally, the target control strategy of the kiwi fruit crop to be monitored may be a series of control measures aiming at pest and disease information of the kiwi fruit crop to be monitored and an implementation strategy thereof. The control strategy may include: agricultural control strategies (e.g., deep ploughing, weeding, trimming, fertilizing, breeding, reseeding, etc.), biological control strategies (e.g., methods utilizing beneficial insects, beneficial birds, frogs, microorganisms, methods causing infertility of pests, etc.), physical control strategies (e.g., setting insect nets, light trapping, plasma disinfection, radiation control, etc.), chemical control strategies (e.g., applying chemical pesticides such as insecticides, bactericides, bird repellents, etc.).

In the above optional embodiment, the current growth information corresponding to the kiwi fruit crop to be monitored, which is acquired from the kiwi fruit knowledge graph based on the internet of things data and the identification result, may include at least one of the following information: the plant diseases and insect pests information of the kiwi fruit crops to be monitored, the growth state of the kiwi fruit crops to be monitored are in accordance with expectation, the growth stage of the kiwi fruit crops to be monitored, and the target control strategy of the kiwi fruit crops to be monitored.

step S310, carrying out classification training on a plurality of pre-training models respectively by adopting a plurality of groups of training data to obtain a training result corresponding to each pre-training model in the plurality of pre-training models, wherein the training result is used for determining the classification performance and generalization performance of each pre-training model in the plurality of pre-training models;

step S312, selecting an initial depth network model from a plurality of pre-training models based on a training result;

step S314, carrying out classification training on the initial deep network model by using multiple groups of training data to obtain a first loss, and carrying out feature reconstruction training on the initial deep network model by using multiple groups of training data to obtain a second loss;

step S316, determining a target loss based on the first loss and the second loss;

and step S318, adjusting the network parameters of the initial deep network model based on the target loss to obtain a target kiwi fruit crop identification model.

In particular, the classification performance may be a performance for representing an accurate classification of the training data by the neural network. Generalization performance may be the performance used to represent a reasonable output of the neural network on data not encountered during training.

In the above optional embodiment, based on the training result corresponding to each of the plurality of pre-training models, an initial deep network model may be selected from the plurality of pre-training models. The initial deep network model may be selected as follows: and selecting a pre-training model of which the classification performance and the generalization performance meet preset conditions (for example, the classification performance is optimal, the generalization performance is the best, and the like) from the plurality of pre-training models as an initial deep network model.

In the above optional embodiment, through the target loss, the network parameters of the initial deep network model may be adjusted, so as to obtain the target kiwi fruit crop identification model. The target kiwi fruit crop recognition model can be used for analyzing the to-be-processed pictures to obtain recognition results. In particular, the principle of adjusting the network parameters may be: the total loss of training of the initial deep network is minimized as much as possible.

Fig. 5 is a schematic diagram of an optional training process of a deep network model according to an embodiment of the present invention, and as shown in fig. 5, the deep network model training process may be used for training a kiwi fruit pest recognition model based on multitask learning. The multitasking may include a classification task and a feature reconstruction task. The training process may comprise the following method steps:

step S3101, kiwi fruit pest and disease damage data are collected, and a kiwi fruit pest and disease damage training sample set is constructed;

step S3102, forming a pre-training model based on a commonly used convolutional neural network DenseNet121, VGG16 and IncepotionV 3;

step S3103, inputting the data in the kiwi fruit pest training sample set into a pre-training model, and performing feature extraction on the data in the kiwi fruit pest training sample set by using the pre-training model as a feature extractor to obtain a feature vector;

step S3104, adding a layer of fully connected layer 20, and performing classification training on the extracted feature vectors by using DenseNet121, VGG16, and inclusion v3 in the pre-training model (such as the classification task shown in fig. 5);

step S3105, comparing and analyzing classification performance and generalization performance of DenseNet121, VGG16 and Inception V3 in the pre-training model on the kiwi fruit pest and disease damage training data, and selecting a network (which can also be subjected to fine adjustment) with the best generalization performance as an initial network for training the kiwi fruit pest and disease damage recognition model based on multi-task learning;

step S3106, designing a feature reconstruction task, i.e. adding four fully-connected layers (fully-connected layer 256, fully-connected layer 218, fully-connected layer 256, fully-connected layer 1024, as shown by the arrow in fig. 5 in order) and one anti-global average pooling layer;

step S3107, inputting the data in the kiwi fruit pest and disease damage training sample set into a kiwi fruit pest and disease damage identification model based on multi-task learning;

step S3108, classifying and training the initial network by adopting the data in the kiwi fruit pest training sample set to obtain Cross Entropy Loss (which is equivalent to the first Loss), and performing feature reconstruction training on the initial network by adopting the data in the kiwi fruit pest training sample set to obtain Mean square Error (which is equivalent to the second Loss);

and S3109, determining Total Loss (which is equivalent to the target Loss) by the cross entropy Loss and the mean square error, and adjusting network parameters of the initial network according to the Total Loss for training again to finally obtain a kiwi fruit pest and disease identification model (which is equivalent to the target kiwi fruit crop identification model) based on multitask learning.

It is easy to note that there are currently few studies on kiwi fruit crops, few training samples available, and overfitting is likely to occur if a single-task learning method is adopted. Through the steps of the method, the kiwi fruit pest and disease identification model with the optimal performance and based on the multitask learning can be obtained through training based on less kiwi fruit pest and disease sample data, and the accuracy of kiwi fruit pest and disease identification can be further improved.

step S320, collecting first target data and second target data, wherein the first target data are machine data collected from the Internet and associated with the kiwi fruit crops to be monitored, and the second target data are expert experience data associated with the kiwi fruit crops to be monitored;

step S322, constructing the kiwi fruit knowledge graph by using the first target data and the second target data.

In the above alternative embodiment, the first target data may be machine data collected from the internet associated with the kiwi fruit crop to be monitored. The machine data may include structured data obtained from a database or a particular website, internet of things data obtained using internet of things devices, machine log data, and the like.

In the above alternative embodiment, the second objective data may be expert experience data associated with the above-mentioned kiwi fruit crop to be monitored. The expert empirical data may be the expertise of a technician communicating with the relevant technical expert for discussion and recording. The expert experience data can include professional knowledge in the whole growth process of the kiwi fruit crop to be monitored, knowledge of plant diseases and insect pests related to the kiwi fruit crop to be monitored, experience knowledge for preventing and treating the plant diseases and insect pests of the kiwi fruit crop to be monitored, and the like.

In the above optional embodiment, first target data and second target data are collected, and a kiwi fruit knowledge-map associated with the kiwi fruit crop to be monitored is constructed using the first target data and the second target data.

Still as shown in fig. 4, the kiwifruit intelligent planting system may also include a process of building a kiwifruit knowledge-graph engine from expert experience and machine data. The construction process may comprise the following method steps:

step S3201, collecting machine data, specifically, acquiring the related structured data of kiwi fruit planting from related databases and related websites, acquiring the related internet of things data of kiwi fruit planting by using internet of things equipment, and acquiring machine log data;

step S3202, collecting expert experiences, specifically, manually constructing a rule base related to kiwi fruit planting, crowdsourcing by knowledge to obtain part of the expert experiences related to kiwi fruit planting, and exchanging discussion with technical experts to record the expert knowledge related to kiwi fruit planting;

and S3203, constructing the kiwi fruit knowledge graph based on the machine data and the expert experience, and providing kiwi fruit knowledge graph calculation engine service.

Specifically, in step S3203, the method of constructing the kiwi fruit knowledge graph based on the machine data and the expert experience may further include the following steps:

step S3031, performing knowledge modeling based on the machine data and the expert experience, specifically, the knowledge modeling may include: entity management, relationship management and attribute management;

step S3032, performing knowledge extraction based on the machine data and the expert experience, specifically, the knowledge extraction may include: entity extraction, relationship extraction and attribute extraction;

step S3033, performing knowledge fusion based on the machine data and the expert experience, specifically, the knowledge fusion may include: entity fusion, relationship fusion and attribute fusion;

step S3034, performing knowledge storage based on the machine data and the expert experience, specifically, the knowledge storage may include: storing native knowledge, storing inference knowledge and editing storage rules;

step S3035, performing knowledge calculation based on the machine data and the expert experience, specifically, the knowledge calculation may include: knowledge inference, knowledge inconsistency detection, and knowledge path analysis.

It is easy to notice that, compared with the methods provided by the related technologies, the method provided by the embodiment can combine objective world machine data and expert experience data, construct the kiwi fruit knowledge map and provide kiwi fruit knowledge map calculation engine service, can reduce the dependence on technical experts in the kiwi fruit pest identification and control process, improve the efficiency and flexibility, and further serve production.

It is easy to note that one of the beneficial effects of the method provided by the present embodiment may be: the kiwi fruit pest and disease control kiwi fruit knowledge map is constructed, and the blank that no mature and available kiwi fruit pest and disease control kiwi fruit knowledge map exists at present is filled.

In an optional embodiment, in step S306, when the current growth information includes pest and disease information of the kiwi fruit crop to be monitored and a target control strategy of the kiwi fruit crop to be monitored, the current growth information is acquired from the kiwi fruit knowledge base based on the internet of things data and the identification result, and the method includes the following steps:

step S361, acquiring pest and disease information from the kiwi fruit knowledge map based on the recognition result;

step S362, determining candidate prevention and treatment means according to the pest and disease information;

and step S363, determining a target prevention strategy from the candidate prevention and treatment means by using the internet of things data.

In the above optional embodiment, the recognition result is obtained by analyzing the to-be-processed picture acquired by the internet of things device by using the target kiwi fruit crop recognition model. When the current growth information includes the pest information of the kiwi fruit crop to be monitored and the target control strategy of the kiwi fruit crop to be monitored, the pest information can be acquired from the kiwi fruit knowledge map corresponding to the kiwi fruit crop to be monitored based on the identification result. The pest information may include the type of pest, the degree of pest, and the like.

Optionally, the candidate control means may include control means for the pest at multiple growth stages (such as germination stage, seedling stage, green stage, flowering stage, fruit stage, mature stage, etc.) of the kiwi fruit crop to be monitored.

In the above optional embodiment, the data of the internet of things may be data acquired by the equipment of the internet of things in real time, and the data of the internet of things may be used to determine growth information and environmental information of the kiwi fruit crop to be monitored in real time, where the growth information may include crop height, crop form, crop color, crop water content, and the like; the environmental information may include soil humidity, air temperature, soil ph, illumination intensity, average daily illumination duration, content of each element in the soil, and the like.

In the above optional embodiment, a target prevention and control strategy may be determined from the candidate prevention and control means by using the internet of things data. The target control strategy can be one or more of a plurality of control means of candidate control means matched with the growth information and the environmental information of the kiwi fruit crop to be monitored, and an implementation strategy of the control means.

In an optional embodiment, in step S363, determining a target control strategy from candidate control means by using the internet of things data includes the following method steps:

step S3631, determining type information, growth period stage information and crop growth environment information of the kiwi fruit crops to be monitored by using the data of the Internet of things;

step S3632, a target control strategy is determined based on the type information, the stage information of the growth cycle and the crop growth environment information.

Optionally, the type information of the kiwi fruit crop to be monitored may be used to indicate the type of the kiwi fruit crop to be monitored, for example: the type information can be specific crop types (such as kiwi fruit, apple, rice, etc.), and can also be crop types (such as grain, fruit tree, fresh flower, etc.).

Optionally, the information of the stage of the growth cycle of the kiwi fruit crop to be monitored may be the current growth stage. For example: germination stage, seedling stage, green stage, flowering stage, fruit stage, mature stage, etc.

Optionally, the crop growth environment information of the kiwi fruit crop to be monitored may include soil humidity, air temperature, soil ph value, illumination intensity, daily average illumination duration, content of each element in soil, and the like.

In the above optional embodiment, the type information, the stage information of the growth cycle and the crop growth environment information of the kiwi fruit crop to be monitored can be determined by using the data of the internet of things. Based on the type information, the stage information of the growth cycle and the crop growth environment information, a target control strategy can be determined from candidate control means.

Optionally, an optimal pest control means can be derived through a relationship path reasoning method, so as to obtain the target control strategy. For example: through the kiwi fruit knowledge map, reasoning is carried out based on a relation path of 'pest and disease information, phenological period, soil conditions and meteorological conditions', and one or more prevention and control means which are optimally matched with all relations in the relation path are deduced.

Still as shown in fig. 4, the kiwi fruit intellectual property map engine in the kiwi fruit intelligent planting system can also obtain a corresponding prevention and control means according to the recognition result of the image recognition system. If the image recognition system recognizes that the kiwifruit suffers from canker, a plurality of control means for the canker under various environmental conditions in each growth period of the kiwifruit can be obtained through the kiwifruit knowledge map engine. Through knowledge calculation, the optimal prevention and treatment means can be deduced.

For example: the kiwi fruit knowledge base map engine can deduce that the best prevention and treatment means for the canker at the moment is pesticide spraying (can also specifically comprise the type, dosage, method and the like of pesticide spraying) based on the current phenological period of the kiwi fruit as the 'flowering-English period'.

For another example: the kiwi fruit knowledge map engine can deduce that the best prevention and treatment means for the canker is to carry out two times of liquid medicine spraying or liquid medicine drying (or specifically comprises the type, dosage and method of the sprayed or dried liquid medicine, the interval time of the two times and the like) after fruit picking and before leaf falling based on the current phenological period of the kiwi fruit as the 'mature period'.

It is easy to note that one of the beneficial effects of the method provided by the present embodiment may be: the damage condition and the optimal control strategy of the kiwi fruit plant diseases and insect pests under different conditions (such as phenological period, soil conditions, weather conditions and the like) can be deduced based on the kiwi fruit plant disease and insect pest knowledge map, and then the production is served.

In an alternative embodiment, in step S308, the current growth information is pushed, which includes one of the following method steps:

step S381, pushing the current growth information according to the preset business rule;

step S382, pushing current growth information based on a determination result of a target determination model, where the target determination model is used to analyze the data of the internet of things to determine the determination result.

In the above optional embodiment, the target determination model may be used to analyze the data of the internet of things to determine the determination result. The data of the internet of things can be used for determining growth information and environmental information of the kiwi fruit crops to be monitored in real time. The decision may be used to determine whether and how to push the current growth information. The target determination model may be a decision tree model that may be constructed based on internet of things data, pest information, control strategy information, machine data, expert experience data, etc. of the kiwi fruit crop to be monitored.

Still as shown in fig. 4, the intelligent kiwi fruit planting system further comprises an intelligent early warning system. The intelligent early warning system can be used for pushing pest control strategies acquired by the kiwi fruit knowledge map engine to users. The intelligent early warning system can comprise the following two early warning modes.

The first early warning mode: and pushing early warning based on the kiwi fruit planting rule (equivalent to the preset business rule) configured by the user. For example: the user-configured kiwi fruit planting rules can include: the early warning system has the advantages that when the diseases and pests occur, the early warning is pushed for the diseases and pests in appointed weather, the early warning is pushed for the diseases and pests in appointed temperature, the early warning is pushed for the diseases and pests in appointed weather, and the early warning is pushed for the diseases and pests in appointed weather and appointed weather.

The second early warning mode: and pushing the early warning based on the target judgment model. And aiming at each pest and disease information of the kiwi fruits, a target judgment model is constructed based on data of information such as weather, soil, phenological period, temperature and the like. And determining whether to push pest early warning or not by using the target judgment model. For example: when the kiwifruit suffers from canker in the flowering-English period, the best prevention and treatment means is pesticide spraying, but the target judgment model can judge that the pesticide is not suitable for spraying at the current time and make a decision on ' not pushing canker early warning information at the current time ' and make a decision on pushing again after the rain stops ' on the basis that the current weather information is ' rainy day '.

It is easy to note that one of the beneficial effects of the method provided by the present embodiment may be: through the preset service rule and the target judgment model, the early warning push strategy of the optimal control strategy of various kiwi fruit plant diseases and insect pests under different conditions (such as the phenological period, the soil condition, the weather condition and the like) is constructed.

It is easy to notice that the pest and disease information of the kiwi fruit crops to be monitored can be found in real time based on the internet of things equipment, and the pest and disease degree identification and prevention strategy decision corresponding to the pest and disease information usually needs guidance of experts. However, the production workers cannot or cannot be guided in time due to the limited expert resources, the messy and complicated internet information and the like. By the method provided by the embodiment, machine data of an objective world and expert experience data can be combined, the kiwi fruit knowledge map is constructed, an engine service for calculating the kiwi fruit knowledge map is provided, and then the pest and disease degree and the control strategy corresponding to pest and disease information are obtained to be pushed to production workers. By the method, the dependence on technical experts in the crop production process can be reduced, the efficiency and the flexibility are improved, and the production is served.

It should be noted that the present invention is focused on: the method comprises the steps of combining machine data of an objective world with expert experience data, constructing a kiwi fruit knowledge graph, providing kiwi fruit knowledge graph calculation engine service, collecting kiwi fruit crop data to be monitored in real time through Internet of things data, identifying current growth information corresponding to crops by using an image identification technology, and reasoning pest and disease damage degree and prevention and control strategies corresponding to crop pest and disease damage information according to the kiwi fruit knowledge graph calculation engine service so as to push the kiwi fruit knowledge graph data to production workers. In the process, the data acquisition mode of the internet of things can be any relevant realizable mode, the image recognition algorithm can be any relevant realizable algorithm, and the building and reasoning method of the kiwi fruit knowledge graph can be any relevant realizable method.

In an optional embodiment, a graphical user interface is provided through the terminal device, the content displayed by the graphical user interface at least partially includes a kiwi fruit crop monitoring scene, and the information processing method further includes the following method steps:

step S391, displaying a plurality of agricultural plots and a plurality of types of candidate agricultural crops planted on the agricultural plots in a graphical user interface;

s392, responding to a first touch operation acting on a graphical user interface, and determining a kiwi fruit crop to be monitored from multiple types of candidate agricultural crops;

step S393, responding to a second touch operation acting on the graphical user interface, and selecting a target crop recognition model and a target pushing mode of the kiwi fruit crop to be monitored;

and step S394, responding to a third touch operation acting on the graphical user interface, acquiring an identification result by using the target crop identification model, acquiring current growth information corresponding to the kiwi fruit to be monitored from the kiwi fruit knowledge map based on the Internet of things data and the identification result, and displaying the current growth information in the graphical user interface according to a target pushing mode.

In the above alternative embodiment, the user may at least partially obtain the above agricultural crop monitoring scene through graphical user interface content displayed by the terminal device. The smallest target of the method provided by the invention can be an agricultural land, and the graphical user interface can display a plurality of agricultural lands and can also display a plurality of types of candidate agricultural crops planted on each agricultural land in the plurality of agricultural lands.

Optionally, in the above graphical user interface, the user may perform a first touch operation on multiple types of candidate agricultural crops shown in the graphical user interface. The user can touch kiwi fruit crops in multiple types of candidate agricultural crops in the graphical user interface to determine the kiwi fruit crops to be monitored.

Optionally, the user may perform a second touch operation on the graphical user interface. The user can determine a target crop recognition model to be used by the kiwi fruit crop to be monitored through at least one of the multiple deep network models displayed in the touch graphic user interface, and can determine a target pushing mode to be used through at least one of the multiple pushing modes displayed in the touch graphic user interface.

Optionally, the user may perform a third touch operation on the graphical user interface. The user can touch an 'identification' button, a 'recommendation' button, a 'generation' button or a 'display' button and the like in the graphical user interface so that the target crop identification model obtains an identification result, obtains current growth information corresponding to the kiwi fruit crop to be monitored from the kiwi fruit knowledge graph based on the internet of things data and the identification result, and displays the current growth information in the graphical user interface according to a target pushing mode.

Under the above operating environment, the present invention provides an information processing method as shown in fig. 6. Fig. 6 is a flowchart of another information processing method according to an embodiment of the present invention, as shown in fig. 6, the information processing method including:

step S602, Internet of things data of the fruit tree to be monitored is obtained, wherein the Internet of things data is used for determining growth information and environmental information of the fruit tree to be monitored in real time;

s604, carrying out image recognition on the fruit tree to be monitored to obtain a recognition result;

step S606, acquiring current growth information corresponding to the fruit tree to be monitored from a fruit tree knowledge map associated with the fruit tree to be monitored based on the data of the Internet of things and the identification result;

step S608, pushing the current growth information.

Alternatively, the fruit trees to be monitored may include pome fruit trees, stone fruit trees, berry fruit trees, nut fruit trees, persimmon fruit trees, and the like.

Optionally, the internet of things data may include various information of sound, light, heat, electricity, mechanics, chemistry, biology, location, and the like of the fruit tree to be monitored. The data of the internet of things can be used for determining growth information and environmental information of the fruit tree to be monitored in real time, wherein the growth information can comprise the height of the fruit tree, the shape of the fruit tree, the color of the fruit tree (including the color of branches, the color of leaves, the color of flowers, the color of fruits and the like), the water content of the fruit tree and the like; the environmental information may include soil humidity, air temperature, soil ph, illumination intensity, average daily illumination time, content of each element in the soil, and the like.

Optionally, the obtaining of the internet of things data of the fruit tree to be monitored may be: the relevant internet of things data is collected in real time by using various internet of things devices or internet of things technologies (such as information sensors, radio frequency identification technologies, global positioning systems, infrared sensors, laser scanners and the like).

Optionally, based on an image recognition technology, image recognition may be performed on the fruit tree to be monitored to obtain the recognition result. Based on the internet of things data and the recognition result of the fruit tree to be monitored, the current growth information corresponding to the fruit tree to be monitored can be obtained from the knowledge map associated with the fruit tree to be monitored. Further, the acquired current growth information may be pushed to the user to guide production.

Optionally, the current growth information corresponding to the fruit tree to be monitored may be used to describe a growth condition of the fruit tree to be monitored at the current time. For example: the current growth information may include: information of the current phenological period (including phenological period type, degree of coincidence between the current phenological period and an expected phenological period, and the like), information of current plant diseases and insect pests (including plant disease and insect pest type, plant disease and insect pest degree, and the like), and the like.

Optionally, the pest information in the current growth information may include disease information and pest information. The diseases can also include biological diseases (such as fungal diseases, bacterial diseases, viral diseases, etc.) and non-biological diseases (such as drought-caused diseases, waterlogging-caused diseases, severe cold diseases, nutrient imbalance diseases, etc.). The insect pest usually refers to the pest (such as beetle, locust, moth larva, aphid, mite, fly, stink bug, etc.) to damage the fruit tree to be monitored.

In the embodiment of the invention, the internet of things data of the fruit tree to be monitored is acquired, wherein the internet of things data is used for determining the growth information and the environmental information of the fruit tree to be monitored in real time, the identification result is obtained by carrying out image identification on the fruit tree to be monitored, the current growth information corresponding to the fruit tree to be monitored is acquired from the fruit tree knowledge map associated with the fruit tree to be monitored based on the internet of things data and the identification result, and the current growth information is further pushed.

It is easy to notice that according to the embodiment of the application, the internet of things data of the fruit tree is obtained through the internet of things technology, the current growth information corresponding to the fruit tree is obtained by combining the image recognition method and the fruit tree related fruit tree knowledge map, and is pushed to a user, so that the aims of determining the harm of the fruit tree and preventing and treating strategies based on image recognition and fruit tree knowledge map reasoning are fulfilled, the technical effects of reducing the dependence on fruit tree technical experts and improving the efficiency and the flexibility in fruit tree growth information recognition and related decisions are achieved, and the technical problems of low efficiency and poor flexibility caused by the fact that farmers or fruit tree technical experts are relied on fruit tree growth information recognition and related decisions in the related technology are solved.

It should be noted that the information processing method provided by the invention can be applied to any application scenario involving internet of things monitoring, hazard identification, hazard early warning and prevention and control strategy recommendation in the field of forestry, but not limited thereto.

It should be noted that, for simplicity of description, the above-mentioned method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the invention. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required by the invention.

Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.

Example 2

According to an embodiment of the present invention, there is also provided an apparatus embodiment for implementing the information processing method, and fig. 7 is a schematic structural diagram of an information processing apparatus according to an embodiment of the present invention, as shown in fig. 7, the apparatus includes: a first acquiring module 701, a recognition module 702, a second acquiring module 703, and a pushing module 704, wherein,

the system comprises a first acquisition module 701, a second acquisition module and a monitoring module, wherein the first acquisition module is used for acquiring the data of the internet of things of crops to be monitored, and the data of the internet of things is used for determining the growth information and the environmental information of the crops to be monitored in real time; the identification module 702 is configured to perform image identification on a crop to be monitored to obtain an identification result; the second obtaining module 703 is configured to obtain, based on the data of the internet of things and the identification result, current growth information corresponding to the crop to be monitored from a knowledge graph associated with the crop to be monitored; and a pushing module 704, configured to push current growth information.

Optionally, the identifying module 702 is further configured to: shooting a crop to be monitored to obtain a picture to be processed; and analyzing the picture to be processed by using the target crop identification model to obtain an identification result, wherein the identification result is used for determining the crop growth condition presented in the picture to be processed.

Alternatively, fig. 8 is a schematic structural diagram of another information processing apparatus according to an embodiment of the present invention, and as shown in fig. 8, the apparatus includes, in addition to all modules shown in fig. 7: the training module 705 is configured to perform classification training on the multiple pre-training models respectively by using multiple sets of training data to obtain a training result corresponding to each pre-training model in the multiple pre-training models, where the training result is used to determine classification performance and generalization performance of each pre-training model in the multiple pre-training models; selecting an initial depth network model from a plurality of pre-training models based on a training result; carrying out classification training on the initial depth network model by utilizing a plurality of groups of training data to obtain a first loss, and carrying out characteristic reconstruction training on the initial depth network model by utilizing a plurality of groups of training data to obtain a second loss; determining a target loss based on the first loss and the second loss; and adjusting network parameters of the initial depth network model based on the target loss to obtain a target crop identification model.

Alternatively, fig. 9 is a schematic structural diagram of another information processing apparatus according to an embodiment of the present invention, and as shown in fig. 9, the apparatus includes, in addition to all modules shown in fig. 8: a construction module 706 configured to collect first target data and second target data, wherein the first target data is machine data collected from the internet and associated with a crop to be monitored, and the second target data is expert experience data associated with the crop to be monitored; and constructing a knowledge graph by using the first target data and the second target data.

Optionally, the second obtaining module 703 is further configured to: acquiring pest and disease information from the knowledge graph based on the recognition result; determining candidate prevention and treatment means through pest and disease information; and determining a target prevention strategy from the candidate prevention and treatment means by using the data of the Internet of things.

Optionally, the second obtaining module 703 is further configured to: determining type information of crops to be monitored, information of the stage of the growth cycle and information of the growth environment of the crops by using the data of the Internet of things; and determining a target control strategy based on the type information, the information of the stage of the growth cycle and the crop growth environment information.

Optionally, the pushing module 704 is further configured to: pushing current growth information according to a preset business rule; pushing current growth information based on a judgment result of the target judgment model; the target judgment model is used for analyzing the data of the Internet of things to determine a judgment result.

It should be noted here that the first obtaining module 701, the identifying module 702, the second obtaining module 703 and the pushing module 704 correspond to steps S202 to S208 in embodiment 1, and the four modules are the same as the corresponding steps in the implementation example and the application scenario, but are not limited to the disclosure in the first embodiment. It should be noted that the modules described above as part of the apparatus may be run in the computer terminal 10 provided in the first embodiment.

In the embodiment of the invention, a first acquisition module is used for acquiring the internet of things data of the crop to be monitored, wherein the internet of things data is used for determining the growth information and the environmental information of the crop to be monitored in real time, the image recognition is carried out on the crop to be monitored through an identification module to obtain an identification result, a second acquisition module is used for acquiring the current growth information corresponding to the crop to be monitored from a knowledge map associated with the crop to be monitored based on the internet of things data and the identification result, and then a pushing module is used for pushing the current growth information.

It should be noted that, reference may be made to the relevant description in embodiment 1 for a preferred implementation of this embodiment, and details are not described here again.

Example 3

There is also provided, in accordance with an embodiment of the present invention, an embodiment of an electronic device, which may be any one of a group of computing devices. The electronic device includes: a processor and a memory, wherein:

a memory coupled to the processor for providing instructions to the processor for processing the following processing steps: the method comprises the steps of obtaining internet of things data of crops to be monitored, wherein the internet of things data is used for determining growth information and environmental information of the crops to be monitored in real time; carrying out image recognition on the crop to be monitored to obtain a recognition result; acquiring current growth information corresponding to the crop to be monitored from a knowledge map associated with the crop to be monitored based on the data of the Internet of things and the identification result; and pushing current growth information.

In the embodiment of the invention, the data of the Internet of things of the crops to be monitored are obtained, wherein the data of the Internet of things is used for determining the growth information and the environmental information of the crops to be monitored in real time, the identification result is obtained by carrying out image identification on the crops to be monitored, the current growth information corresponding to the crops to be monitored is obtained from the knowledge map associated with the crops to be monitored on the basis of the data of the Internet of things and the identification result, and then the current growth information is pushed.

It is easy to notice that through the embodiment of the application, the internet of things data of crops is obtained through the internet of things technology, the current growth information corresponding to the crops is obtained by combining the image recognition method and the related knowledge map of the crops and is pushed to users, and the purposes of determining crop damage and prevention and treatment strategies based on image recognition and knowledge map reasoning are achieved, so that the technical effects of reducing dependence on technical experts and improving efficiency and flexibility in crop growth information recognition and related decisions are achieved, and the technical problems of low efficiency and poor flexibility caused by dependence on production workers or manual processing of the technical experts in the crop growth information recognition and related decisions in the related technology are solved.

Example 4

The embodiment of the invention can provide a computer terminal which can be any computer terminal device in a computer terminal group. Optionally, in this embodiment, the computer terminal may also be replaced with a terminal device such as a mobile terminal.

Optionally, in this embodiment, the computer terminal may be located in at least one network device of a plurality of network devices of a computer network.

In this embodiment, the computer terminal may execute program codes of the following steps in the information processing method: the method comprises the steps of obtaining internet of things data of crops to be monitored, wherein the internet of things data is used for determining growth information and environmental information of the crops to be monitored in real time; carrying out image recognition on the crop to be monitored to obtain a recognition result; acquiring current growth information corresponding to the crop to be monitored from a knowledge map associated with the crop to be monitored based on the data of the Internet of things and the identification result; and pushing current growth information.

Optionally, fig. 10 is a block diagram of another computer terminal according to an embodiment of the present invention, and as shown in fig. 10, the computer terminal may include: one or more processors 122 (only one of which is shown), memory 124, and peripherals interface 126.

The memory may be used to store software programs and modules, such as program instructions/modules corresponding to the information processing method and apparatus in the embodiments of the present invention, and the processor executes various functional applications and data processing by running the software programs and modules stored in the memory, so as to implement the information processing method described above. The memory may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory may further include memory remotely located from the processor, which may be connected to the computer terminal over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The processor can call the information and application program stored in the memory through the transmission device to execute the following steps: the method comprises the steps of obtaining internet of things data of crops to be monitored, wherein the internet of things data is used for determining growth information and environmental information of the crops to be monitored in real time; carrying out image recognition on the crop to be monitored to obtain a recognition result; acquiring current growth information corresponding to the crop to be monitored from a knowledge map associated with the crop to be monitored based on the data of the Internet of things and the identification result; and pushing current growth information.

Optionally, the processor may further execute the program code of the following steps: shooting a crop to be monitored to obtain a picture to be processed; and analyzing the picture to be processed by using the target crop identification model to obtain an identification result, wherein the identification result is used for determining the crop growth condition presented in the picture to be processed.

Optionally, the processor may further execute the program code of the following steps: the current growth information includes at least one of: treat the pest and disease information of monitoring crop, wherein, pest and disease information includes: whether the crop to be monitored has plant diseases and insect pests or not and the degree of the plant diseases and insect pests of the crop to be monitored; whether the growth state of the crops to be monitored meets expectations or not; the growth stage of the crop to be monitored; and (3) target control strategy of the crops to be monitored.

Optionally, the processor may further execute the program code of the following steps: respectively carrying out classification training on the plurality of pre-training models by adopting a plurality of groups of training data to obtain a training result corresponding to each pre-training model in the plurality of pre-training models, wherein the training result is used for determining the classification performance and the generalization performance of each pre-training model in the plurality of pre-training models; selecting an initial depth network model from a plurality of pre-training models based on a training result; carrying out classification training on the initial deep network model by utilizing a plurality of groups of training data to obtain a first loss, and carrying out characteristic reconstruction training on the initial deep network model by utilizing a plurality of groups of training data to obtain a second loss; determining a target loss based on the first loss and the second loss; and adjusting network parameters of the initial depth network model based on the target loss to obtain a target crop identification model.

Optionally, the processor may further execute the program code of the following steps: collecting first target data and second target data, wherein the first target data is machine data collected from the internet and associated with a crop to be monitored, and the second target data is expert experience data associated with the crop to be monitored; and constructing a knowledge graph by using the first target data and the second target data.

Optionally, the processor may further execute the program code of the following steps: when the current growth information comprises pest and disease information of the crops to be monitored and a target control strategy of the crops to be monitored, acquiring the pest and disease information from the knowledge map based on the identification result; determining candidate prevention and control means through pest and disease information; and determining a target prevention strategy from the candidate prevention and treatment means by using the data of the Internet of things.

Optionally, the processor may further execute the program code of the following steps: determining type information of crops to be monitored, information of the stage of the growth cycle and information of the growth environment of the crops by using the data of the Internet of things; and determining a target control strategy based on the type information, the information of the stage of the growth cycle and the crop growth environment information.

Optionally, the processor may further execute the program code of the following steps: pushing current growth information according to a preset business rule; pushing current growth information based on a judgment result of the target judgment model; the target judgment model is used for analyzing the data of the Internet of things to determine a judgment result.

Optionally, the processor may further execute the program code of the following steps: displaying a plurality of plots and a plurality of types of candidate crops planted on the plots in a graphical user interface; determining a crop to be monitored from the multiple types of candidate crops in response to a first touch operation acting on the graphical user interface; responding to a second touch operation acting on the graphical user interface, and selecting a target crop recognition model and a target pushing mode of the crop to be monitored; responding to a third touch operation acting on the graphical user interface, acquiring an identification result by using the target crop identification model, acquiring current growth information corresponding to the crop to be monitored from the knowledge map based on the Internet of things data and the identification result, and displaying the current growth information in the graphical user interface according to a target pushing mode.

Optionally, the processor may further execute the program code of the following steps: responding to the editing operation acted on the target prevention and control strategy, and acquiring a prevention and control operation result corresponding to the target prevention and control strategy; and optimizing a target prevention strategy based on the prevention operation result.

The processor can call the information and application program stored in the memory through the transmission device to execute the following steps: the method comprises the steps of obtaining internet of things data of the kiwi fruit crops to be monitored, wherein the internet of things data is used for determining growth information and environmental information of the kiwi fruit crops to be monitored in real time; carrying out image recognition on the kiwi fruit crop to be monitored to obtain a recognition result; based on the data of the Internet of things and the identification result, acquiring current growth information corresponding to the kiwi fruit crop to be monitored from the kiwi fruit knowledge map associated with the kiwi fruit crop to be monitored; and pushing the current growth information.

Optionally, the processor may further execute the program code of the following steps: displaying a plurality of agricultural plots and a plurality of types of candidate agricultural crops planted on the agricultural plots in a graphical user interface; determining a kiwi fruit crop to be monitored from a plurality of types of candidate agricultural crops in response to a first touch operation acting on a graphical user interface; responding to a second touch operation acting on the graphical user interface, and selecting a target crop recognition model and a target pushing mode of the kiwi fruit crop to be monitored; responding to a third touch operation acting on the graphical user interface, acquiring an identification result by using the target crop identification model, acquiring current growth information corresponding to the kiwi fruit to be monitored from the kiwi fruit knowledge map based on the Internet of things data and the identification result, and displaying the current growth information in the graphical user interface according to a target pushing mode.

The processor can call the information and application program stored in the memory through the transmission device to execute the following steps: the method comprises the steps of obtaining Internet of things data of a fruit tree to be monitored, wherein the Internet of things data is used for determining growth information and environmental information of the fruit tree to be monitored in real time; carrying out image recognition on the fruit tree to be monitored to obtain a recognition result; based on the data of the Internet of things and the recognition result, obtaining current growth information corresponding to the fruit tree to be monitored from a fruit tree knowledge map associated with the fruit tree to be monitored; and pushing current growth information.

It should be understood by those skilled in the art that the structure shown in fig. 10 is only an example, and the computer terminal may also be a terminal device such as a smart phone (e.g., an Android phone, an iOS phone, etc.), a tablet computer, a palmtop computer, a Mobile Internet Device (MID), a PAD, and the like. Fig. 10 is a diagram illustrating a structure of the electronic device. For example, the computer terminal may also include more or fewer components (e.g., network interfaces, display devices, etc.) than shown in FIG. 10, or have a different configuration than shown in FIG. 10.

Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by a program instructing hardware associated with the terminal device, where the program may be stored in a computer-readable storage medium, and the storage medium may include: flash disks, Read-Only memories (ROMs), Random Access Memories (RAMs), magnetic or optical disks, and the like.

Embodiments of a computer-readable storage medium are also provided according to embodiments of the present invention. Alternatively, in this embodiment, the computer-readable storage medium may be configured to store the program code executed by the information processing method provided in embodiment 1.

Optionally, in this embodiment, the computer-readable storage medium may be located in any one of a group of computer terminals in a computer network, or in any one of a group of mobile terminals.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: the method comprises the steps of obtaining internet of things data of crops to be monitored, wherein the internet of things data is used for determining growth information and environmental information of the crops to be monitored in real time; carrying out image recognition on the crop to be monitored to obtain a recognition result; acquiring current growth information corresponding to the crop to be monitored from a knowledge map associated with the crop to be monitored based on the data of the Internet of things and the identification result; and pushing current growth information.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: shooting a crop to be monitored to obtain a picture to be processed; and analyzing the picture to be processed by using the target crop identification model to obtain an identification result, wherein the identification result is used for determining the crop growth condition presented in the picture to be processed.

Optionally, in this embodiment, a computer-readable storage medium is configured to store program code for performing the steps of: the current growth information includes at least one of: treat the pest information of monitoring crop, wherein, the pest information includes: whether the crop to be monitored has plant diseases and insect pests or not and the degree of the plant diseases and insect pests of the crop to be monitored; whether the growth state of the crops to be monitored meets expectations or not; the growth stage of the crop to be monitored; and (3) target control strategy of the crops to be monitored.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: respectively carrying out classification training on the plurality of pre-training models by adopting a plurality of groups of training data to obtain a training result corresponding to each pre-training model in the plurality of pre-training models, wherein the training result is used for determining the classification performance and the generalization performance of each pre-training model in the plurality of pre-training models; selecting an initial depth network model from a plurality of pre-training models based on a training result; carrying out classification training on the initial depth network model by utilizing a plurality of groups of training data to obtain a first loss, and carrying out characteristic reconstruction training on the initial depth network model by utilizing a plurality of groups of training data to obtain a second loss; determining a target loss based on the first loss and the second loss; and adjusting network parameters of the initial depth network model based on the target loss to obtain a target crop identification model.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: collecting first target data and second target data, wherein the first target data is machine data collected from the internet and associated with a crop to be monitored, and the second target data is expert experience data associated with the crop to be monitored; and constructing a knowledge graph by using the first target data and the second target data.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: when the current growth information comprises pest and disease information of the crops to be monitored and a target control strategy of the crops to be monitored, acquiring the pest and disease information from the knowledge map based on the identification result; determining candidate prevention and treatment means through pest and disease information; and determining a target prevention strategy from the candidate prevention and treatment means by using the data of the Internet of things.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: pushing current growth information according to a preset business rule; and pushing current growth information based on a judgment result of a target judgment model, wherein the target judgment model is used for analyzing the data of the Internet of things to determine the judgment result.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: displaying a plurality of plots and a plurality of types of candidate crops planted on the plots in a graphical user interface; determining a crop to be monitored from the multiple types of candidate crops in response to a first touch operation acting on the graphical user interface; responding to a second touch operation acting on the graphical user interface, and selecting a target crop recognition model and a target pushing mode of the crop to be monitored; responding to a third touch operation acting on the graphical user interface, acquiring an identification result by using the target crop identification model, acquiring current growth information corresponding to the crop to be monitored from the knowledge map based on the Internet of things data and the identification result, and displaying the current growth information in the graphical user interface according to a target pushing mode.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: responding to the editing operation acting on the target prevention and control strategy, and acquiring a prevention and control operation result corresponding to the target prevention and control strategy; and optimizing a target prevention strategy based on the prevention operation result.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: the method comprises the steps of obtaining internet of things data of the kiwi fruit crops to be monitored, wherein the internet of things data is used for determining growth information and environmental information of the kiwi fruit crops to be monitored in real time; carrying out image recognition on the kiwi fruit crop to be monitored to obtain a recognition result; acquiring current growth information corresponding to the kiwi fruit crop to be monitored from a kiwi fruit knowledge map associated with the kiwi fruit crop to be monitored based on the data of the Internet of things and the identification result; and pushing current growth information.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: displaying a plurality of agricultural plots and a plurality of types of candidate agricultural crops planted on the agricultural plots in a graphical user interface; determining a kiwi fruit crop to be monitored from a plurality of types of candidate agricultural crops in response to a first touch operation acting on a graphical user interface; responding to a second touch operation acting on the graphical user interface, and selecting a target crop recognition model and a target pushing mode of the kiwi fruit crop to be monitored; responding to a third touch operation acting on the graphical user interface, acquiring an identification result by using the target crop identification model, acquiring current growth information corresponding to the kiwi fruit to be monitored from the kiwi fruit knowledge map based on the Internet of things data and the identification result, and displaying the current growth information in the graphical user interface according to a target pushing mode.

Optionally, in this embodiment, the computer readable storage medium is configured to store program code for performing the following steps: the method comprises the steps of obtaining internet of things data of a fruit tree to be monitored, wherein the internet of things data is used for determining growth information and environmental information of the fruit tree to be monitored in real time; carrying out image recognition on the fruit tree to be monitored to obtain a recognition result; based on the data of the Internet of things and the recognition result, acquiring current growth information corresponding to the fruit tree to be monitored from a fruit tree knowledge map associated with the fruit tree to be monitored; and pushing current growth information.

The above-mentioned serial numbers of the embodiments of the present invention are only for description, and do not represent the advantages and disadvantages of the embodiments.

In the above embodiments of the present invention, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

In the embodiments provided in the present application, it should be understood that the disclosed technology can be implemented in other ways. The above-described embodiments of the apparatus are merely illustrative, and for example, the division of the units is only one type of division of logical functions, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, units or modules, and may be in an electrical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit may be implemented in the form of hardware, or may also be implemented in the form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.

The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.

Claims

1. An information processing method characterized by comprising:

the method comprises the steps of obtaining internet of things data of crops to be monitored, wherein the internet of things data is used for determining growth information and environmental information of the crops to be monitored in real time;

carrying out image recognition on the crops to be monitored to obtain recognition results;

acquiring current growth information corresponding to the crop to be monitored from a knowledge map associated with the crop to be monitored based on the data of the Internet of things and the identification result;

and pushing the current growth information.

2. The information processing method according to claim 1, wherein the image recognition of the crop to be monitored is performed, and obtaining the recognition result comprises:

shooting the crops to be monitored to obtain pictures to be processed;

and analyzing the picture to be processed by using a target crop recognition model to obtain the recognition result, wherein the recognition result is used for determining the crop growth condition presented in the picture to be processed.

3. The information processing method according to claim 1, wherein the current growth information includes at least one of:

the pest and disease information of the crop to be monitored, wherein the pest and disease information comprises: whether the crop to be monitored has plant diseases and insect pests or not and the plant disease and insect pest degree of the crop to be monitored;

whether the growth state of the crop to be monitored is in accordance with expectations or not;

the growth stage of the crop to be monitored;

and (3) a target control strategy of the crop to be monitored.

4. The information processing method according to claim 1, characterized by further comprising:

collecting first target data and second target data, wherein the first target data is machine data collected from the internet and associated with the crop to be monitored, and the second target data is expert experience data associated with the crop to be monitored;

constructing the knowledge-graph using the first target data and the second target data.

5. The information processing method according to claim 1, wherein when the current growth information includes pest and disease information of the crop to be monitored and a target control strategy of the crop to be monitored, acquiring the current growth information from the knowledge graph based on the internet of things data and the recognition result includes:

acquiring the pest and disease damage information from the knowledge graph based on the identification result;

determining candidate prevention and control means according to the pest and disease information;

determining the target control strategy from the candidate control means using the internet of things data.

6. The information processing method according to claim 5, wherein determining the target control strategy from the candidate control means using the IOT data comprises:

determining the type information of the crop to be monitored, the information of the stage of the growth cycle and the information of the crop growth environment by using the data of the Internet of things;

and determining the target control strategy based on the type information, the information of the stage of the growth cycle and the crop growth environment information.

7. The information processing method according to claim 1, wherein pushing the current growth information includes one of:

pushing the current growth information according to a preset business rule;

pushing the current growth information based on a judgment result of a target judgment model;

the target judgment model is used for analyzing the data of the Internet of things to determine the judgment result.

8. The information processing method of claim 1, wherein a graphical user interface is provided through the terminal device, the content displayed by the graphical user interface at least partially includes a target crop monitoring scene, and the information processing method further comprises:

displaying a plurality of plots and a plurality of types of candidate crops planted on the plurality of plots within the graphical user interface;

responding to a first touch operation acting on the graphical user interface, and determining the crop to be monitored from the multiple types of candidate crops;

responding to a second touch operation acting on the graphical user interface, and selecting a target crop recognition model and a target pushing mode of the crop to be monitored;

responding to a third touch operation acting on the graphical user interface, acquiring the identification result by using the target crop identification model, acquiring current growth information corresponding to the crop to be monitored from the knowledge graph based on the Internet of things data and the identification result, and displaying the current growth information in the graphical user interface according to the target pushing mode.

9. An information processing method, characterized by comprising:

the method comprises the steps of obtaining data of the Internet of things of kiwi fruit crops to be monitored, wherein the data of the Internet of things are used for determining growth information and environmental information of the kiwi fruit crops to be monitored in real time;

carrying out image recognition on the kiwi fruit crops to be monitored to obtain recognition results;

based on the data of the Internet of things and the identification result, acquiring current growth information corresponding to the kiwi fruit crop to be monitored from the kiwi fruit knowledge map associated with the kiwi fruit crop to be monitored;

and pushing the current growth information.

10. The information processing method of claim 9, wherein a graphical user interface is provided through the terminal device, the content displayed by the graphical user interface at least partially includes a kiwi monitoring scene, and the information processing method further comprises:

displaying, within the graphical user interface, a plurality of agricultural plots and a plurality of types of candidate agricultural crops planted on the plurality of agricultural plots;

responding to a first touch operation acting on the graphical user interface, and determining the kiwi fruit crop to be monitored from the multiple types of candidate agricultural crops;

responding to a second touch operation acting on the graphical user interface, and selecting a target crop recognition model and a target pushing mode of the kiwi fruit crop to be monitored;

responding to a third touch operation acting on the graphical user interface, utilizing the target crop identification model to acquire the identification result, acquiring current growth information corresponding to the kiwi fruit crop to be monitored from the kiwi fruit knowledge map based on the Internet of things data and the identification result, and displaying the current growth information in the graphical user interface according to the target pushing mode.

11. An information processing method characterized by comprising:

the method comprises the steps of obtaining Internet of things data of a fruit tree to be monitored, wherein the Internet of things data is used for determining growth information and environmental information of the fruit tree to be monitored in real time;

carrying out image recognition on the fruit tree to be monitored to obtain a recognition result;

based on the data of the Internet of things and the recognition result, obtaining current growth information corresponding to the fruit tree to be monitored from a fruit tree knowledge map associated with the fruit tree to be monitored;

and pushing the current growth information.

12. A computer-readable storage medium, comprising a stored program, wherein when the program runs, the computer-readable storage medium controls an apparatus to execute the information processing method according to any one of claims 1 to 11.

13. An electronic device, comprising:

a processor; and

a memory coupled to the processor for providing instructions to the processor for processing the following processing steps:

step 1, obtaining Internet of things data of crops to be monitored, wherein the Internet of things data is used for determining growth information and environmental information of the crops to be monitored in real time;

step 2, carrying out image recognition on the crops to be monitored to obtain recognition results;

step 3, acquiring current growth information corresponding to the crop to be monitored from a knowledge map associated with the crop to be monitored based on the data of the Internet of things and the identification result;

and 4, pushing the current growth information.