A method and system for monitoring agricultural products

Abstract

The embodiment of the specification provides a kind of agricultural product monitoring method, the method comprises: obtaining the variety information of agricultural product, planting area information and growth image;Based on variety information, planting area information and growth image, the growth parameter of agricultural product is predicted by vigor prediction model;Wherein, growth parameter includes one or more of the growth cycle, plant height, crown width and flower amount of agricultural product;Vigor prediction model includes feature extraction layer and vigor prediction layer, feature extraction layer is used to process growth image, and determine image feature;Vigor prediction layer is used to process image feature, variety information, planting area information, and determine growth parameter;When growth parameter does not satisfy preset condition, send early warning reminder to target terminal.

Description

A method and system for monitoring agricultural products

[0001] Case Analysis

[0002] This application is a divisional application of Chinese application filed on April 6, 2022, with application number 202210353960.0 and entitled "A Method and System for Predicting the Growth of Agricultural Products". Technical Field

[0003] This manual relates to the field of agricultural products, and in particular to a method and system for monitoring agricultural products. Background Technology

[0004] In agricultural production, different varieties of agricultural products are adapted to different regional growth conditions (such as soil conditions and climate conditions). Growth conditions are often the main influencing factor on the growth of agricultural products, and timely screening and eliminating agricultural products with poor growth conditions is the key to ensuring agricultural yield.

[0005] Therefore, it is desirable to provide a method for monitoring the growth of agricultural products, which can be used to monitor agricultural products with poor growth and, if necessary, provide early warnings. Summary of the Invention

[0006] This specification provides one or more embodiments of a method for monitoring agricultural products. The method includes: acquiring variety information, planting area information, and growth images of the agricultural product; predicting growth parameters of the agricultural product using a growth prediction model based on the variety information, the planting area information, and the growth images; wherein the growth parameters include one or more of the following: growth cycle, plant height, canopy width, and flowering quantity; the growth prediction model includes a feature extraction layer and a growth prediction layer, wherein the feature extraction layer processes the growth images to determine image features; the growth prediction layer processes the image features, the variety information, and the planting area information to determine the growth parameters; and when the growth parameters do not meet preset conditions, sending an early warning reminder to a target terminal.

[0007] This specification provides one or more embodiments of an agricultural product monitoring system, comprising: an acquisition module for acquiring variety information, planting area information, and growth images of the agricultural product; a prediction module for predicting growth parameters of the agricultural product based on the variety information, the planting area information, and the growth images using a growth prediction model; wherein the growth parameters include one or more of the following: growth cycle, plant height, canopy width, and flowering quantity; the growth prediction model includes a feature extraction layer and a growth prediction layer, wherein the feature extraction layer processes the growth images to determine image features; the growth prediction layer processes the image features, the variety information, and the planting area information to determine the growth parameters; and an early warning module for sending an early warning reminder to a target terminal when the growth parameters do not meet preset conditions.

[0008] This specification provides one or more embodiments of an agricultural product monitoring device, including a processor for executing an agricultural product monitoring method.

[0009] This specification provides one or more embodiments of a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes an agricultural product monitoring method. Attached Figure Description

[0010] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0011] Figure 1 is a schematic diagram of an application scenario of an agricultural product growth monitoring system according to some embodiments of this specification;

[0012] Figure 2 is an exemplary block diagram of an agricultural product growth monitoring system according to some embodiments of this specification;

[0013] Figure 3 is an exemplary flowchart of a method for monitoring the growth of agricultural products according to some embodiments of this specification;

[0014] Figure 4 is a schematic diagram of a growth prediction model according to some embodiments of this specification;

[0015] Figure 5 is a schematic diagram illustrating the determination of confidence levels for growth parameters according to some embodiments of this specification. Detailed Implementation

[0016] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0017] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0018] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0019] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0020] Figure 1 is a schematic diagram of an application scenario of an agricultural product growth monitoring system according to some embodiments of this specification.

[0021] In some embodiments, the application scenario 100 of the agricultural product growth monitoring system may include a processor 110, a network 120, a storage device 130, a monitoring device 140, a terminal device 150, and agricultural products 160. Application scenario 100 can collect relevant information about agricultural products (e.g., variety information, planting area information, growth images, etc.) by implementing the methods and / or processes disclosed in this specification, determine the growth parameters of the agricultural products based on the relevant information, and provide feedback and reminders based on the production parameters. This allows for accurate identification of the growth status of agricultural products and facilitates the timely detection of poorly growing agricultural products.

[0022] Processor 110 can be used to process data and / or information from at least one component of application scenario 100 or an external data source (e.g., a cloud data center). Processor 110 can be connected to storage device 130, monitoring device 140, and / or terminal device 150 via network 120 to access and / or receive data and information. For example, processor 110 can receive relevant information (e.g., variety information, planting area information, growth images, etc.) collected by monitoring device 140 on agricultural product 160 via network 120. In other embodiments, processor 110 can send production parameters of agricultural product 160 (e.g., growth cycle, plant height, canopy width, flowering quantity, etc.) to terminal device 150 via network 120. In some embodiments, processor 110 can be a single processor or a group of processors. The group of processors can be centralized or distributed (e.g., processor 110 can be a distributed system), and can be dedicated or simultaneously provided by other devices or systems. In some embodiments, processor 110 can be locally connected to network 120 or remotely connected to network 120. In some embodiments, processor 110 can be implemented on a cloud platform. By way of example only, the cloud platform may include private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, internal cloud, multi-layered cloud, etc., or any combination thereof. In some embodiments, the processor 110 may be located in places including but not limited to farm control rooms, agricultural production management centers, etc. In some embodiments, the processor 110 is equipped with a collaborative platform for directing and coordinating the various tasks of agricultural production workers. Agricultural production workers may include agricultural product breeders, agricultural product growers, agricultural product comprehensive management personnel, agricultural product technical experts, etc.

[0023] Network 120 can facilitate the exchange of information and / or data. In some embodiments, one or more components of application scenario 100 (e.g., storage device 130, monitoring device 140, terminal device 150) can transmit information and / or data to another component of application scenario 100 via network 120. Network 120 may include a local area network (LAN), a wide area network (WAN), a wired network, a wireless network, or any combination thereof. In some embodiments, network 120 may be any one or more of wired or wireless networks. In some embodiments, network 120 may include one or more network access points. For example, network 120 may include wired or wireless network access points, such as base stations and / or network switching points, through which one or more components of application scenario 100 can connect to network 120 to exchange data and / or information.

[0024] Storage device 130 can be used to store data and / or instructions. Data may include information relating to users, terminal device 150, monitoring device 140, etc. In some embodiments, storage device 130 may store data and / or instructions used by processor 110 to execute or use in order to perform the exemplary methods described herein. For example, storage device 130 may store historical agricultural product information. As another example, storage device 130 may store one or more machine learning models. In some embodiments, storage device 130 may be part of processor 110. In some embodiments, storage device 130 may include mass storage, removable storage, volatile read-write memory, read-only memory (ROM), etc., or any combination thereof. In some embodiments, storage device 130 may be implemented on a cloud platform. In some embodiments, storage device 130 may be connected to network 120 to communicate with one or more components of application scenario 100 (e.g., processor 110, terminal device 150, monitoring device 140).

[0025] The monitoring device 140 refers to a device used to acquire relevant information about agricultural products (such as variety information, planting area information, growth images, etc.). In some embodiments, the monitoring device 140 can be implemented by various detection devices. For example, it may include a drone 140-1 and a soil unmanned detection device 140-2. In some embodiments, the drone 140-1 can collect growth images of agricultural products. In some embodiments, the soil unmanned detection device 140-2 can collect soil information of the area. For more details on acquiring relevant information about agricultural products through the monitoring device 140, please refer to Figure 3 and its related description, which will not be repeated here.

[0026] Terminal device 150 may refer to one or more terminal devices or software used by a user. In some embodiments, the user (e.g., agricultural grower, agricultural production technology expert, etc.) may be the owner of terminal device 150. In some embodiments, terminal device 150 may include mobile device 150-1, tablet computer 150-2, laptop computer 150-3, vehicle-mounted device, etc., or any combination thereof. In some embodiments, terminal device 150 may include a signal transmitter and a signal receiver, configured to communicate with monitoring device 140 to obtain relevant information about agricultural products. In some embodiments, terminal device 150 may be fixed and / or mobile. For example, terminal device 150 may be directly installed on processor 110 and / or monitoring device 140, becoming part of processor 110 and / or monitoring device 140. As another example, terminal device 150 may be a mobile device that agricultural production workers can carry to a location far from processor 110, monitoring device 140, and agricultural products 160, and terminal device 150 may connect to and / or communicate with processor 110 and / or monitoring device 140 via network 120. In some embodiments, terminal device 150 may receive user requests and send information related to the requests to processor 110 via network 120. For example, terminal device 150 may receive a user requesting the sending of information related to agricultural products and / or growth parameters, and send the information related to the requests to processor 110 via network 120. Terminal device 150 may also receive information from processor 110 via network 120. For example, terminal device 150 may receive information related to monitoring device 140 or agricultural product 160 from processor 110. One or more of the identified relevant information may be displayed on terminal device 150. As another example, processor 110 may send growth parameters (e.g., growth cycle, plant height, canopy width, flowering quantity, etc.) or reminder information (e.g., growth parameters do not meet preset conditions, etc.) generated based on information related to agricultural products to terminal device 150.

[0027] Agricultural product 160 refers to products produced in agriculture (e.g., radishes, tomatoes, cabbage, peanuts, corn, wheat, chickens, milk, eggs, etc.). In some embodiments, relevant information about agricultural product 160 can be collected by monitoring device 140 and transmitted to processor 110 and / or terminal device 150 via network 120. In some embodiments, relevant information about agricultural product 160 may include variety information, planting area information, growth images, etc. In some embodiments, relevant information about agricultural product 160 may also include animal breed information, breeding area information, location information, etc. For example, growth images of agricultural product 160 can be collected by drone 140-1, and the growth images can be transmitted to processor 110 via network 120. In some embodiments, relevant information about agricultural product 160 (e.g., variety information, planting area information, growth images, etc.) can be transmitted to processor 110 to determine growth parameters (e.g., growth cycle, plant height, canopy width, flowering quantity, etc.) and / or feedback reminders (e.g., poor tomato growth, etc.), and transmitted to terminal device 150 via network 120. In some embodiments, the growth parameters of agricultural product 160 can also be transmitted to processor 110 to generate improvement plans to provide guidance to agricultural product production workers.

[0028] It should be noted that the above description of the application scenarios of the agricultural product growth monitoring system is for ease of description only and should not be construed as limiting this specification to the scope of the embodiments described. It is understood that those skilled in the art, after understanding the principle of the system, may arbitrarily combine the various components or connect sub-components with other components without departing from this principle. In some embodiments, the processor and storage device disclosed in Figure 1 may be different units within a single component, or a single part may implement the functions of two or more of the aforementioned parts. For example, the various components may share a single storage unit, or each component may have its own separate storage unit. Such variations are all within the scope of protection of this specification.

[0029] Figure 2 is a block diagram of an agricultural product growth monitoring system according to some embodiments of this specification.

[0030] In some embodiments, the agricultural product growth monitoring system 200 may include an acquisition module 210 and a prediction module 220.

[0031] The acquisition module 210 can be used to acquire variety information, planting area information, and growth images of agricultural products. In some embodiments, the planting area information includes soil and climate information of the planting area. For more information on variety information, planting area information, and growth images, please refer to Figure 3 and its related description; further details will not be provided here.

[0032] The prediction module 220 can be used to predict the growth parameters of the agricultural product based on the variety information, the planting area information, and the growth image, using a growth prediction model. The growth parameters include one or more of the following: growth cycle, plant height, canopy width, and flowering quantity. For more information on the growth prediction model, growth cycle, plant height, canopy width, and flowering quantity, please refer to Figure 3 and its related description; further details are omitted here.

[0033] In some embodiments, the agricultural product growth monitoring system 200 may further include a determination module 230 and an early warning module 240.

[0034] The determination module 230 can be used to determine the confidence level of the growth parameters. For more information on confidence levels, please refer to Figure 5 and its related description; further details will not be provided here.

[0035] The early warning module 240 can be used to send an early warning reminder to the target terminal when the growth parameters do not meet the preset conditions. For more information on the preset conditions and early warning reminders, please refer to Figure 5 and its related description; further details will not be provided here.

[0036] In some embodiments, the agricultural product growth monitoring system 200 may further include a transmission module (not shown in the figure). The transmission module can be used to transmit relevant information about the agricultural product and / or information that the growth parameters do not meet preset conditions to the agricultural product production management and control center or corresponding terminal equipment. The transmission method can be wired, such as through open wires, cables, and optical fibers, or wireless, such as through microwave, satellite, scattering, ultra-shortwave, shortwave, Wi-Fi, Bluetooth, and infrared.

[0037] It should be noted that the above description of the agricultural product growth monitoring system and its modules is for convenience only and should not be construed as limiting this specification to the scope of the embodiments described. It is understood that those skilled in the art, after understanding the principles of the system, may arbitrarily combine the various modules or construct subsystems connected to other modules without departing from these principles. In some embodiments, the acquisition module 210, prediction module 220, determination module 230, and early warning module 240 disclosed in FIG1 may be different modules within a single system, or a single module may implement the functions of two or more of the aforementioned modules. For example, the modules may share a single storage module, or each module may have its own separate storage module. Such variations are all within the scope of protection of this specification.

[0038] Figure 3 is a schematic diagram of a method for monitoring the growth of agricultural products according to some embodiments of this specification. In some embodiments, process 300 can be executed by processor 110. As shown in Figure 3, process 300 may include the following steps:

[0039] Step 310: Obtain variety information, planting area information, and growth images of the agricultural product. In some embodiments, step 310 can be performed by the acquisition module 210.

[0040] Variety information for agricultural products refers to information about the types of plants cultivated in agriculture. In some embodiments, variety information may include, but is not limited to, the plant type, basic growth parameters, and growth habits of the agricultural product. Plant type can refer to a specific category of a plant classified according to different standards. For example, the plant category of an agricultural product may include soybeans, potatoes, tomatoes, etc. Plant type may also include more specific species information for each crop. For example, soybean variety information includes yellow soybeans, green soybeans, red soybeans, black soybeans, etc.; for example, tuber variety information includes sweet potatoes, potatoes, purple sweet potatoes, etc.; for example, wheat variety information includes winter wheat, spring wheat, etc.; for example, corn variety may be Xianyu 335, etc. Basic growth parameters refer to basic data that can reflect the agricultural product during the growth process, and may include data on the stems, leaves, flowers, or fruits of the agricultural product at different stages. For example, basic growth parameters can include the number of leaves, plant height, and the number and size of flowers of a winter wheat variety at different stages: sowing, emergence, tillering, overwintering, greening, jointing, booting, heading, flowering, grain-filling, and maturity. Growth habits refer to the inherent adaptive attributes formed by the long-term interaction between agricultural products and the environment, which may include preference for warm and drought-tolerant conditions, preference for sunlight, preference for high temperatures and abundant rainfall, preference for fertile soil but insensitivity to waterlogging, and susceptibility to pests and diseases. For example, wheat prefers warm temperatures and is drought-tolerant. In some embodiments, variety information of agricultural products can be obtained in various ways. For example, variety information can be input through a user terminal. Another example is the acquisition of images of agricultural products by drones, followed by image recognition to obtain variety information.

[0041] Planting area information refers to information related to the planting area of ​​agricultural products. A planting area can refer to the region where an agricultural product is grown. For example, an area where winter wheat is grown. The planting area can be determined based on the preset planting area of ​​the agricultural product. The location and size of the planting area can be determined based on the user's preset planting location and planting area for the agricultural product.

[0042] In some embodiments, planting area information may include soil and climate information of the planting area.

[0043] Soil information refers to relevant information about the soil in the planting area. Soil information may include, but is not limited to, soil type, soil texture, soil depth, soil moisture content, and the content of various elements in the soil. For example, soil information may include that the soil type in this area is lateritic red soil, the soil depth is approximately 5m, the texture is sandy, the moisture content is 38%, and the average manganese content is 271.54 μg / g, the average zinc content is 52.65 μg / g, the average copper content is 31.78 μg / g, and the average nickel content is 15.81 μg / g.

[0044] In some embodiments, soil information in agricultural product planting areas can be collected using monitoring devices. For example, the soil in agricultural product planting areas can be periodically tested using monitoring devices to obtain soil information. Here, "monitoring device" refers to a device capable of monitoring soil. For example, the monitoring device may include a soil content detection device. In some embodiments, the monitoring device may include an unmanned soil monitoring device. The unmanned soil monitoring device can test the soil in agricultural product planting areas to obtain soil information for that area. The unmanned soil monitoring device can periodically (e.g., twice a month) test the soil in agricultural product planting areas to obtain soil information for that area.

[0045] In some embodiments, when the growth parameters of agricultural products in the planting area do not meet preset conditions, the detection frequency and / or detection area of ​​the unmanned soil monitoring device can be adjusted. For example, when the growth parameters of agricultural products in the planting area do not meet preset conditions, the detection frequency of the unmanned soil monitoring device can be adjusted from 2 times / month to 4 times / month. As another example, when the growth parameters of agricultural products in the planting area do not meet preset conditions, the detection area of ​​the unmanned soil monitoring device can be adjusted from the planting area to the area represented by a new edge formed by extending the edge of the planting area by 50cm. For more details regarding the failure of agricultural product growth parameters in the planting area to meet preset conditions, please refer to Figure 5 and its related description; further details will not be elaborated here.

[0046] In some embodiments, the corresponding detection frequency and detection items can be set according to the variety information of agricultural products. For example, when crops are susceptible to waterlogging and require fertilization, the focus can be on detecting soil moisture and soil nutrients. Another example is when crops thrive in alkaline soil, soil pH can be tested.

[0047] Climate information can refer to information about the climate conditions of the area where agricultural products are grown. Climate information may include, but is not limited to, information about the temperature, humidity, wind speed, rainfall, and air pressure of the growing area.

[0048] In some embodiments, climate information can be obtained through the internet, broadcasting, television, or other means. In some embodiments, climate information can be obtained through third-party platforms. These third-party platforms may include meteorological websites, agricultural meteorological information networks, etc. Climate information can be obtained periodically, or the acquisition cycle and content of climate information can be set based on the variety information of agricultural products.

[0049] A growth image refers to visual information about the current growth of an agricultural product. Growth images of agricultural products can include images of their nutrient structure and reproductive structure. Nutrient structure images can be images of roots, stems, leaves, etc., while reproductive structure images can be images of flowers, fruits, seeds, etc.

[0050] In some embodiments, images of agricultural product growth can be acquired through various methods. For example, images of agricultural products can be obtained by periodically photographing them using surveillance cameras. In some embodiments, images of agricultural products can be captured using drones. A drone is an unmanned aircraft controlled by radio remote control equipment and its own program control device, which can be operated autonomously, either completely or intermittently, by humans or computers.

[0051] Step 320: Based on variety information, planting area information, and growth images, predict the growth parameters of the agricultural product using a growth prediction model. These growth parameters include one or more of the following: growth cycle, plant height, canopy width, and number of flowers. In some embodiments, step 320 can be performed by the prediction module 220.

[0052] Growth parameters refer to data that reflect changes in the growth of agricultural products. In some embodiments, growth parameters may include one or more of the following: growth cycle, plant height, canopy width, and flowering rate. Growth cycle refers to the time required for an agricultural product to grow from sowing to harvest. For example, the growth cycle of corn is generally 95-130 days. Plant height refers to the distance from the root collar to the top of the plant. For example, the plant height of sugarcane can generally reach 3 meters. Canopy width refers to the width of the plant in each direction. For example, the canopy width of pomegranate is 100 centimeters. Flowering rate refers to the number of young shoots that develop into flowers. For example, the flowering rate of lychee can generally reach 90%. In some embodiments, growth parameters may also include other relevant parameters of the agricultural product. For example, growth parameters may also include ear position.

[0053] In some embodiments, the variety information, planting area information, and growth images of agricultural products can be input into the growth prediction model, and the output can be the growth parameters of the agricultural product. For example, if the input is the variety information, planting area information, and growth images of potatoes, the output can be the predicted growth cycle of the potato as 70 days, the plant height as 90 cm, and the flowering rate as 60%.

[0054] In some embodiments, the growth prediction model can be a deep neural network model. The model can be input with information about the variety of the agricultural product, its planting area, and growth images, and output with the growth parameters of that product. For example, if the input is potato variety information, planting area information, and growth images, the output could be a predicted growth cycle of 70 days, a plant height of 90 cm, and a flowering rate of 60%. In some embodiments, the growth prediction model can be trained based on historical planting data. Training samples can include information about the variety of the agricultural product, its planting area, and historical growth images. The labels of the training samples can be the growth parameters of the agricultural product. Both the training samples and their labels can be directly obtained from historical planting data. The labeled training samples are input into the initial growth prediction model, and the parameters of the initial growth prediction model are updated through training. When the trained model meets preset conditions, training ends, and the trained growth prediction model is obtained.

[0055] As shown in Figure 4, the growth prediction model 440 may further include a feature extraction layer 440-1 and a growth prediction layer 440-2 connected sequentially. The growth image 410 can be input into the feature extraction layer 440-1, outputting image features 441. Then, the image features 441, planting area information 420, and variety information 430 are input into the growth prediction layer 440-2, outputting growth parameters 450. The feature extraction layer 440-1 can be a convolutional neural network model, and the growth prediction layer 440-2 can be a deep neural network model. Image features include color features, shape features, and spatial relationship features of the growth image. Color features refer to the surface properties of crops. For example, the color features of chili peppers are red, yellow, and green; shape features are the outer boundary features of crops. For example, the shape feature of rice is the panicle, and the shape feature of celery is the stem; spatial relationship features refer to the relationship features of crops relative to a certain standard. For example, the number of leek leaves and the size of potato roots.

[0056] In some embodiments, the feature extraction layer and the growth prediction layer can be jointly trained. Training data samples include variety information, planting area information, and historical growth images of agricultural products. The labels of the training samples include the growth parameters of the agricultural products. The historical growth images from the training samples are input into the feature extraction layer of the growth prediction model. The output of the feature extraction layer, along with the variety information and planting area information of the agricultural products from the training samples, are input into the growth prediction layer of the growth prediction model. A loss function is constructed based on the output and labels of the growth prediction layer, and the parameters of both the feature extraction layer and the growth prediction layer are iteratively updated simultaneously based on the loss function until preset conditions are met and training is complete, resulting in a trained feature extraction layer and growth prediction layer.

[0057] In some embodiments, when training the growth prediction model, the growth parameters can be divided into multiple segments, and labels can be constructed based on the range of the corresponding growth parameters for each agricultural product. For example, plant height can be set to (0-10, 10-20, 20-30, 30-40, 40-50, 50-60, 60-70, 70-80, 80-90, 90-100). When the plant height of a certain agricultural product is 75cm, the label can be set to (0, 0, 0, 0, 0, 0, 1, 0, 0).

[0058] It should be understood that for multiple agricultural products of the same variety that are planted, grown, and fertilized simultaneously in a certain planting area, the growth parameters of all agricultural products in that planting area can be determined by predicting the growth parameters of one or a few of the agricultural products.

[0059] In some embodiments, the confidence level of the predicted growth parameters can also be determined. The confidence level can characterize the degree of reliability of the predicted growth parameters of the agricultural product.

[0060] In some embodiments, the output of the growth prediction model may further include a first confidence level, which can be directly used as the confidence level of the growth parameter. As shown in Figure 4, the output of the growth prediction model 440 may also include a first confidence level 460. In some embodiments, the output of the growth prediction model may be represented by a vector, where the position of each element in the vector can represent the range corresponding to the growth parameter, and the value of each element can represent the probability that the growth parameter is within the corresponding range. The range corresponding to the growth parameter with the highest probability can be determined as the final growth parameter, and the highest probability can be determined as the first confidence level. For example, the output of the growth prediction model may be (0, 0, 0, 0, 0, 0, 0.15, 0.8, 0.05, 0), where each element in the vector represents a plant height of 0-10cm, 10-20cm, 20-30cm, 30-40cm, 40-50cm, 50-60cm, 60-70cm, 70-80cm, 80-90cm, and 90-100cm. Therefore, the plant height of this agricultural product can be predicted to be 70-80cm, with a first confidence level of 0.8. Correspondingly, when training the growth prediction model, the labels can be represented as vectors. The labels can be obtained manually based on historical planting data, and the specific values ​​of the vectors within the labels can be obtained statistically. For example, by statistically analyzing 100 agricultural products with similar variety information, planting area information, and growth images, the number of agricultural products within the range corresponding to each growth parameter can be determined, thereby determining the specific values ​​of the vectors.

[0061] In some cases, the crop growth images and / or planting area information input into the growth prediction model may be inaccurate, leading to inaccurate first confidence scores obtained by the model. Therefore, in some embodiments, the confidence scores of growth parameters can be determined by combining the second confidence scores of the crop growth images and / or planting area information with the first confidence scores output by the growth prediction model. Further details regarding the above embodiments are provided in Figure 5 and its related description, and will not be repeated here.

[0062] In some embodiments, the growth prediction model can be augmented based on its output at each iteration. During augmented training, the loss function for training the growth prediction model can be adjusted based on the second confidence level. In some embodiments, the second confidence level can be converted into an influence factor and added to the loss function. For more details on the second confidence level, please refer to Figure 5 and its related description; further explanation is omitted here.

[0063] Loss functions are functions that can represent the risk or loss of random events, including perceptual loss functions, cross-entropy loss functions, and mean squared error loss functions. Taking the mean squared error loss function as an example, it can calculate the Euclidean distance between the predicted value and the true value. The closer the predicted value and the true value are, the smaller their mean squared error. The influence factor refers to the reference influence probability corresponding to agricultural product planting area information and growth image. For a set of training data, if the sampling accuracy is low, the proportion of the loss value in the total loss is reduced; if the sampling accuracy is high, the proportion of the loss value in the total loss is increased. Taking the mean squared error loss function as an example, as shown in formula (1), the influence factor can be added before the mean squared error loss term during reinforcement training:

[0064]

[0065] in, Let the mean squared error loss function be used. It is the impact factor.

[0066] Some embodiments in this specification utilize agricultural product variety information, planting area information, and growth images to predict agricultural product growth parameters using machine learning models, thereby gaining insight into the growth status of the agricultural products. Furthermore, prediction based on machine learning models can reduce labor costs and improve prediction efficiency. Additionally, it allows for the determination of the confidence level of growth parameters, thus ensuring the accuracy of the prediction results.

[0067] In some embodiments, the growth parameters of agricultural products can be monitored to determine whether they meet preset conditions for the agricultural product. When monitoring the growth parameters of agricultural products, process 300 may further include the following steps:

[0068] Step 330: When the growth parameters do not meet the preset conditions, the early warning module can send an early warning reminder to the target terminal. In some embodiments, step 330 can be executed by the early warning module 240.

[0069] Preset conditions refer to the conditions that are pre-defined to determine the normal growth of agricultural products. Preset conditions may include, but are not limited to, one or more requirements for the growth cycle, plant height, canopy width, and number of flowers. For example, when tomatoes are in the flowering and fruit-setting stage, the preset conditions are: plant height of 18-25cm, canopy width of 0.8-1.5m, 4-7 petals, and 3-7 flowers per inflorescence stalk.

[0070] In some embodiments, the preset conditions can be set by agricultural production workers (e.g., agricultural technology experts, agricultural growers, etc.) based on their past experience. In some embodiments, the early warning module can store the normal growth parameters corresponding to different varieties of agricultural products at different growth stages in a storage device and / or a database. When it is necessary to obtain the preset conditions for a certain agricultural product (e.g., tomato), the early warning module retrieves the corresponding preset conditions from the storage device and / or database as needed. In some embodiments, the preset conditions can also be obtained by the early warning module accessing the Internet.

[0071] In some embodiments, preset conditions can be adjusted based on the confidence level of growth parameters. In some embodiments, when the confidence level of a growth parameter is higher than a threshold (e.g., 90%, 70%), the warning module can lower the specific requirement index in the preset conditions. The warning module can determine the growth parameter requirement in the adjusted preset conditions based on the proportion of confidence level higher than the threshold. Taking a winter wheat as an example, the preset plant height requirement is 70cm, the confidence threshold is 90%, and the actual confidence level is 95.6%, which is 5.6% higher than the threshold. Then, the wheat plant height requirement can be reduced proportionally or by an appropriate reduction. As an example only, a proportional reduction means a reduction of 5.6%, resulting in a plant height requirement of approximately 66cm; an appropriate reduction means a reduction of 1 / 2, i.e., a reduction of 2.8%, resulting in a plant height requirement of approximately 68cm.

[0072] A warning reminder is a notification message indicating that the growth parameters may not meet preset conditions. In some embodiments, the content of the warning reminder may include, but is not limited to, current weather information (e.g., sunny, 31℃; heavy rain, 20℃; light snow, -5℃, etc.), the current growth cycle of the agricultural product, the reasons why the growth parameters do not meet the preset conditions (e.g., excessively low temperature, excessively low soil nitrogen content, etc.), and the specific abnormal conditions of the corresponding growth parameters (e.g., insufficient plant height, no flowering, small fruit diameter, yellowing leaves, etc.) and a description of the probability of the abnormal conditions (e.g., highly likely, possibly, etc.). The probability description of the abnormal conditions can be determined by the confidence level of the growth parameters. For example, a confidence level greater than 80% corresponds to "highly likely," a confidence level between 50% and 80% corresponds to "possibly," and a confidence level less than 50% corresponds to "maybe." As an example only, the warning reminder content could be: "The current temperature is 18℃, sunny, and the apple tree is in its flowering period. Due to the excessively low phosphorus and potassium content in the soil, the apple tree is highly likely not to flower."

[0073] In some embodiments, the warning alert may be a combination of one or more forms, including but not limited to SMS, text push, images, videos, voice, and broadcasts.

[0074] When growth parameters do not meet preset conditions, the early warning module can send an early warning alert to the target terminal (e.g., the mobile phone of the staff responsible for the agricultural product production, the control console of the agricultural product production management center, etc.) to notify the user (e.g., the agricultural product production staff) that the predicted growth of the agricultural product is abnormal. In some embodiments, when growth parameters meet preset conditions, information may not be sent to the target terminal, or the status information of the agricultural product may be sent to the target terminal. For example, the content of the agricultural product may include that the agricultural product is currently growing well and is expected to mature within 20 days.

[0075] The methods described in some embodiments of this manual, by alerting users when the growth parameters of agricultural products do not meet preset conditions, help users to promptly grasp abnormal growth conditions of agricultural products, and to handle and maintain them. It also facilitates subsequent investigation into whether the planting or growth conditions of similar agricultural products are abnormal, thus avoiding greater losses.

[0076] Figure 5 is a flowchart illustrating the determination of confidence levels for growth parameters according to some embodiments of this specification. In some embodiments, process 500 may be executed by determination module 230. As shown in Figure 5, process 500 may include the following steps:

[0077] Step 510: Determine the second confidence level of the crop growth images and / or planting area information.

[0078] The second confidence level refers to the confidence level of the crop growth images and / or the confidence level of the planting area information.

[0079] The confidence level of agricultural product growth images can be determined in various ways. In some embodiments, the confidence level of the growth images can be determined based on weather information at the time of drone capture. For example, if the drone capture is taken during fog with very low visibility, the confidence level is low; if the drone capture is taken during strong winds that affect the image balance and thus the image quality, the confidence level is also low.

[0080] In some embodiments, the confidence level of a growth image can be determined based on its quality, thereby determining the confidence level of the current growth image of the agricultural product. For example, a low-resolution growth image results in a low confidence level. Methods for evaluating the resolution of growth images may include the Tenengrad gradient method, the Laplacian gradient method, variance methods, etc.

[0081] The confidence level of planting area information can be determined in several ways. For example, it can be based on the accuracy of the monitoring device. For instance, if the accuracy of the monitoring device is 97%, the corresponding confidence level of the planting area information can be determined to be 97%.

[0082] When the second confidence level includes the confidence level of the crop growth image or the confidence level of the planting area information, the confidence level of the crop growth image or the confidence level of the planting area information can be directly used as the second confidence level. When the second confidence level includes both the confidence level of the crop growth image and the confidence level of the planting area information, the average of the confidence levels of the crop growth image and the confidence level of the planting area information can be used as the second confidence level.

[0083] Step 520: Determine the first confidence level of the growth prediction model output.

[0084] In some embodiments, the first confidence level can be obtained through a growth prediction model. For more information on the first confidence level, please refer to Figure 4 and its related description; further details will not be provided here.

[0085] Step 530: Determine the confidence level of the growth parameters based on the first confidence level and the second confidence level.

[0086] In some embodiments, the confidence level of the growth parameter can be determined by combining the first confidence level and the second confidence level.

[0087] In some embodiments, the first confidence level and the second confidence level may correspond to different weights. In some embodiments, the weight of the first confidence level may be determined based on the accuracy coefficient of the growth prediction model, and the weight of the second confidence level may be determined based on the weight of the first confidence level.

[0088] The accuracy coefficient refers to the proportion of correctly predicted growth parameters among all predicted growth parameters. In some embodiments, the accuracy coefficient can be expressed as:

[0089]

[0090] Where V is the accuracy coefficient of the growth prediction model, V T For example, V is the correct growth parameter. F Examples of incorrect growth parameters. As shown in formula (3), the accuracy coefficient can be directly determined as the weight of the first confidence level:

[0091] S1=V (3) The weight of the second confidence level can be expressed as:

[0092] The confidence level of the growth parameters after fusion (S1 = 1 - S2(4)) can be expressed as:

[0093] P = S2 * P2 + S1 * P1 (5) where P1 is the first confidence level, S1 is the weight of the first confidence level, P2 is the second confidence level, S2 is the weight of the second confidence level, and P is the confidence level of the growth parameter.

[0094] In some embodiments, the frequency of acquiring regional information on agricultural products and the accuracy of drone photography can be adjusted based on the confidence level of growth parameters.

[0095] In some embodiments, when the confidence level of growth parameters is below a confidence threshold, the detection frequency of the soil testing device and the accuracy of growth images captured by the drone can be increased. The confidence threshold can be preset empirically.

[0096] Some embodiments in this specification integrate variety information, planting area information, and growth images of agricultural products, and utilize a growth prediction model to predict the growth parameters of agricultural products, thereby more accurately identifying the growth status of agricultural products. Simultaneously, by enhancing the training of the growth prediction model and adjusting its loss function, the prediction accuracy can be improved. By determining the confidence level of growth parameters and adjusting the growth parameters based on the confidence level, agricultural products with poor growth can be monitored, ensuring the quality of agricultural products.

[0097] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0098] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0099] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented by hardware devices, they can also be implemented solely by software solutions, such as installing the described system on an existing processor or mobile device.

[0100] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0101] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0102] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0103] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A method for monitoring agricultural products, characterized in that, The method includes: acquiring variety information, planting area information, and growth images of agricultural products; predicting growth parameters of the agricultural products based on the variety information, planting area information, and growth images using a growth prediction model; wherein the growth parameters include one or more of the following: growth cycle, plant height, canopy width, and flowering quantity; the growth prediction model includes a feature extraction layer and a growth prediction layer, wherein the feature extraction layer is used to process the growth images to determine image features; the growth prediction layer is used to process the image features, the variety information, and the planting area information to determine the growth parameters; and the growth parameters are then determined. The confidence level of the growth parameter is used to adjust preset conditions; wherein, determining the confidence level of the growth parameter includes: determining a first confidence level output by the growth prediction model; determining a second confidence level of the growth image and / or the planting area information of the agricultural product; and fusing the first confidence level and the second confidence level by weighted summation as the confidence level of the growth parameter; wherein, the weight of the first confidence level is determined based on the accuracy coefficient of the growth prediction model, and the weight of the second confidence level is determined based on the weight of the first confidence level; when the growth parameter does not meet the preset conditions, a warning reminder is sent to the target terminal.

2. The method according to claim 1, characterized in that, The planting area information includes soil and climate information of the planting area for the agricultural products.

3. The method according to claim 2, characterized in that, The soil information is detected by an unmanned soil detection device, including: determining the detection frequency and / or detection items of the unmanned soil detection device based on the variety information; and adjusting the detection frequency of the unmanned soil detection device when the growth parameters do not meet the preset conditions.

4. An agricultural product monitoring system, characterized in that, The system includes: an acquisition module for acquiring variety information, planting area information, and growth images of agricultural products; and a prediction module for predicting growth parameters of the agricultural products based on the variety information, planting area information, and growth images using a growth prediction model; wherein the growth parameters include one or more of the following: growth cycle, plant height, canopy width, and flowering quantity; the growth prediction model includes a feature extraction layer and a growth prediction layer, wherein the feature extraction layer processes the growth images to determine image features; and the growth prediction layer processes the image features, variety information, and planting area information to determine the growth parameters; and a determination module. The system is used to determine the confidence level of the growth parameters, which is used to adjust preset conditions. The determining module is further used to: determine the first confidence level output by the growth prediction model; determine the second confidence level of the growth image and / or the planting area information of the agricultural product; and fuse the first confidence level and the second confidence level by weighted summation as the confidence level of the growth parameters. The weight of the first confidence level is determined based on the accuracy coefficient of the growth prediction model, and the weight of the second confidence level is determined based on the weight of the first confidence level. An early warning module is also used to send an early warning reminder to the target terminal when the growth parameters do not meet the preset conditions.

5. The system according to claim 4, characterized in that, The planting area information includes soil and climate information of the planting area for the agricultural products.

6. The system according to claim 5, characterized in that, The soil information is detected by an unmanned soil detection device, including: determining the detection frequency and / or detection items of the unmanned soil detection device based on the variety information; and adjusting the detection frequency of the unmanned soil detection device when the growth parameters do not meet the preset conditions.

7. An agricultural product monitoring device, comprising a processor, the processor being configured to execute the agricultural product monitoring method according to any one of claims 1 to 3.

8. A computer-readable storage medium storing computer instructions, wherein when a computer reads the computer instructions in the storage medium, the computer executes the agricultural product monitoring method as described in any one of claims 1 to 3.

Images