CN115577866A - Method and device for predicting waiting period, electronic equipment and storage medium - Google Patents
Method and device for predicting waiting period, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN115577866A CN115577866A CN202211576214.4A CN202211576214A CN115577866A CN 115577866 A CN115577866 A CN 115577866A CN 202211576214 A CN202211576214 A CN 202211576214A CN 115577866 A CN115577866 A CN 115577866A
- Authority
- CN
- China
- Prior art keywords
- phenological period
- prediction
- phenological
- remote sensing
- period
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Mining
Abstract
The invention relates to the technical field of agricultural intelligent monitoring, and provides a phenological period prediction method, a phenological period prediction device, electronic equipment and a storage medium. The method comprises the following steps: acquiring canopy coverage, remote sensing image data and meteorological data of a crop area to be predicted, and extracting vegetation indexes from the remote sensing image data to obtain vegetation indexes; and inputting the canopy coverage, the vegetation index and the meteorological data into the phenological period prediction model to obtain a phenological period prediction result output by the phenological period prediction model. According to the method, the phenological period prediction is carried out by combining the canopy coverage, the vegetation index and the meteorological data, so that the accuracy of the phenological period prediction is improved, and the phenological period prediction result comprises a phenological period result and phenological period interval days, so that not only can the current phenological period of the target crop in the crop area to be predicted be obtained through prediction, but also the interval days from the current prediction time to the next phenological period of the current phenological period can be obtained through prediction, and the accuracy of the phenological period prediction is further improved.
Description
Technical Field
The invention relates to the technical field of agricultural intelligent monitoring, in particular to a phenological period prediction method, a phenological period prediction device, electronic equipment and a storage medium.
Background
The phenological period refers to the response of the growth, development and activity of crops and the change of organisms to the phenological period, and is called the phenological period when the response is occurring. The phenological period of the crops is predicted, so that the crops can be guided to be accurately planted, pests can be prevented and controlled, risks can be avoided, and production resources can be reasonably allocated, so that the yield and the quality of the crops are improved.
At present, the phenological period of crops is predicted by an accumulated temperature method, namely, the phenological period of the crops is judged according to the current accumulated temperature of the crops. However, the temperature accumulation method only considers temperature data, the prediction result of the climate period is not accurate, and the climate period prediction method can only predict the climate period result at the current time, which results in low accuracy of the climate period prediction.
Disclosure of Invention
The invention provides a phenological period prediction method, a phenological period prediction device, electronic equipment and a storage medium, which are used for solving the defect of low phenological period prediction accuracy in the prior art and realizing high-accuracy phenological period prediction.
The invention provides a phenological period prediction method, which comprises the following steps:
acquiring canopy coverage, remote sensing image data and meteorological data of a to-be-predicted crop area, and performing vegetation index extraction processing on the remote sensing image data to obtain a vegetation index;
inputting the canopy coverage, the vegetation index and the meteorological data into a phenological period prediction model to obtain a phenological period prediction result output by the phenological period prediction model;
the phenological period prediction result comprises a phenological period result and phenological period interval days, the phenological period result is used for representing the current phenological period where the target crop in the crop area to be predicted is located, and the phenological period interval days are used for representing the interval days between the current prediction time and the next phenological period of the current phenological period.
According to the phenological period prediction method provided by the invention, the canopy coverage, the vegetation index and the meteorological data are input into a phenological period prediction model to obtain a phenological period prediction result output by the phenological period prediction model, and the phenological period prediction method comprises the following steps:
inputting the canopy coverage, the vegetation index and the meteorological data into a feature extraction layer of a phenological period prediction model to obtain a feature vector output by the feature extraction layer;
inputting the feature vector into a first phenological period prediction layer of the phenological period prediction model to obtain a phenological period result output by the first phenological period prediction layer;
and inputting the characteristic vector to a second waiting period prediction layer of the waiting period prediction model to obtain the number of days of a waiting period interval output by the second waiting period prediction layer.
According to the phenological period prediction method provided by the present invention, the step of inputting the feature vector into a first phenological period prediction layer of the phenological period prediction model to obtain phenological period results output by the first phenological period prediction layer includes:
inputting the feature vector into a phenological period classification layer of the phenological period prediction model to obtain a phenological period result output by the phenological period classification layer;
the step of inputting the feature vector into a second waiting period prediction layer of the waiting period prediction model to obtain the number of days of the waiting period interval output by the second waiting period prediction layer comprises the following steps:
and inputting the characteristic vector into a phenological period regression layer of the phenological period prediction model to obtain phenological period interval days output by the phenological period regression layer.
According to the phenological period prediction method provided by the invention, the canopy coverage is obtained based on the following steps:
determining a sowing time of the target crop, a canopy growth coefficient of the target crop, and an initial canopy size of the target crop at the sowing time;
determining canopy coverage of the crop area to be predicted based on the seeding time, the canopy growth coefficient and the initial canopy size.
According to the phenological period prediction method provided by the invention, the remote sensing image data is obtained based on the following steps:
obtaining remote sensing data of the to-be-predicted crop area;
screening the remote sensing data based on the cloud coverage rate and the preset cloud coverage rate of the remote sensing data to obtain remote sensing image data of the object area to be predicted; or the like, or a combination thereof,
inputting the remote sensing data into a cloud layer identification model to obtain a cloud layer area output by the cloud layer identification model, and screening the remote sensing data based on the cloud layer area to obtain remote sensing image data of the object area to be predicted.
According to the method for predicting the phenological period, provided by the invention, the meteorological data comprise at least one of average temperature, maximum temperature, minimum temperature, average humidity, sunshine duration, accumulated precipitation and maximum wind speed.
According to the phenological period prediction method provided by the invention, the acquisition of canopy coverage, remote sensing image data and meteorological data of a to-be-predicted crop area comprises the following steps:
determining a historical data acquisition period based on the current prediction time and a preset duration;
and acquiring canopy coverage, remote sensing image data and meteorological data of the object area to be predicted during the acquisition of the historical data.
The present invention also provides a phenological period prediction device, including:
the system comprises an acquisition module, a prediction module and a prediction module, wherein the acquisition module is used for acquiring canopy coverage, remote sensing image data and meteorological data of a to-be-predicted crop area, and carrying out vegetation index extraction processing on the remote sensing image data to obtain a vegetation index;
the prediction module is used for inputting the canopy coverage, the vegetation index and the meteorological data into a phenological period prediction model to obtain a phenological period prediction result output by the phenological period prediction model;
the phenological period prediction results comprise phenological period results and phenological period interval days, the phenological period results are used for representing the current phenological period where the target crops in the crop area to be predicted are located, and the phenological period interval days are used for representing the interval days between the current prediction time and the next phenological period of the current phenological period.
The present invention also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for predicting a waiting period as described in any of the above methods when executing the program.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method of predicting a phenological period as described in any one of the above.
According to the phenological period prediction method, the device, the electronic equipment and the storage medium, canopy coverage, remote sensing image data and meteorological data of a to-be-predicted crop area are obtained, and the canopy coverage, the vegetation index extracted based on the remote sensing image data and the meteorological data are input into a phenological period prediction model, so that not only is temperature data considered, but also more comprehensive meteorological data is considered, and the phenological period prediction is carried out by combining the canopy coverage and the vegetation index, and the phenological period prediction accuracy is improved; meanwhile, compared with meteorological data, the remote sensing image data has higher resolution, so that the prediction of the phenological period at the block level can be carried out, and the accuracy of the phenological period prediction is further improved; in addition, the phenological period prediction result output by the phenological period prediction model comprises a phenological period result and phenological period interval days, so that the current phenological period where the target crop in the crop area to be predicted is located can be predicted, the interval days between the current prediction time and the next phenological period of the current phenological period can be predicted, and the phenological period prediction accuracy is further improved.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of a phenological period prediction method according to the present invention;
FIG. 2 is a second schematic flow chart of the method for predicting a phenological period according to the present invention;
fig. 3 is a third schematic flow chart of the phenological period prediction method provided in the present invention;
FIG. 4 is a schematic structural diagram of a phenological period prediction device provided in the present invention;
fig. 5 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. 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.
The phenological period refers to the response of the growth, development and activity of crops and the change of organisms to the phenological period, and is called the phenological period when the response is generated. Specifically, the phenological stage can be divided according to the external morphological changes of the plant. The phenological stage can be divided into seedling stage, tillering stage, flowering stage, mature stage, etc. Different crop types have different stage development characteristics and morphogenetic processes. The growth and development of crops are related to various factors such as temperature, illumination, moisture, soil, the crops and the like.
The phenological period of the crops is predicted, so that the crops can be guided to be accurately planted, pests can be prevented and controlled, risks can be avoided, and production resources can be reasonably allocated, so that the yield and the quality of the crops are improved. Therefore, it is important to accurately predict the phenological period of the crop, which has a great influence on agricultural production, and on the basis of this, accurate prediction of the phenological period of the crop is required.
Traditional phenological period prediction is mostly realized by a growth degree daily method and synchronization of a specific phenological period in a corresponding stem apical development stage. Specifically, the phenological period in the early and late stages of fertility, which is less affected by the temperature photoreaction, is mainly predicted by the degree of growth day, and the phenological period in the reproductive development stage is mainly predicted according to the synchronicity of the phenological period in the stem apex development stage. However, the method is mainly completed manually, manual observation is greatly influenced by a main observer and seriously depends on the experience of personnel, so that the accuracy of the forecast of the phenological period is low, the efficiency of manual observation is low, and the efficiency of the forecast of the phenological period is greatly reduced.
At present, the phenological period of crops is predicted by an accumulated temperature method, namely, the phenological period of the crops is judged according to the current accumulated temperature of the crops. For example, after the sowing period, when the accumulated temperature of the crops exceeds a certain temperature and the water content of the soil reaches a certain value, the germination period is reached, after the emergence of the seedlings, the tillering period is reached through a certain accumulated temperature, and the stage that the crops reach different climatic stages is sequentially judged according to the accumulated temperature of the crops. However, the temperature accumulation method only considers temperature data, the predicted phenological period result is not accurate, and the phenological period result at the current time can only be predicted, so that the phenological period prediction accuracy is not high.
In view of the above problems, the present invention proposes the following embodiments. Fig. 1 is a schematic flow chart of a phenological period prediction method provided in the present invention, as shown in fig. 1, the phenological period prediction method includes:
and 110, acquiring canopy coverage, remote sensing image data and meteorological data of a to-be-predicted crop area, and performing vegetation index extraction processing on the remote sensing image data to obtain a vegetation index.
Here, the crop area to be predicted is an area where crops are planted, and the crops thereof need to be subjected to phenological period prediction. The crop area to be predicted usually comprises one crop, but also can comprise a plurality of crops, and if the crop area to be predicted comprises a plurality of crops, the area to be predicted can be firstly semantically divided so as to respectively predict the phenological period of each crop.
Here, canopy Coverage (CC) is used to characterize the growth and development of the crop, so that a better phenological prediction can be made based on Canopy coverage.
In one embodiment, the canopy coverage can be determined by a model of the crop growth mechanism. Specifically, the sowing time of the target crop in the crop area to be predicted is input into the crop growth mechanism model, and the canopy coverage output by the crop growth mechanism model is obtained, namely the canopy coverage in the crop growth process is simulated based on the crop growth mechanism model, the canopy coverage of the crops at different growth stages is different, in other words, the canopy coverage of the crops at different phenological stages is different, and then phenological stage prediction can be carried out based on the canopy coverage. The crop growth mechanism model can be set according to actual needs, for example, the aquacrop mechanism model.
In one embodiment, the canopy coverage includes day-to-day canopy coverage, thereby increasing the amount of data in canopy coverage and increasing the accuracy of the phenological period prediction.
In one embodiment, a historical data acquisition period is determined based on the current predicted time and a preset time length; canopy coverage of the area to be predicted during historical data acquisition is obtained. The preset time can be set according to actual needs, for example, 15 days; the preset time lengths of different crops can be different, the preset time lengths of the same crop of different varieties can be different, and the preset time lengths of the same crop in different regions can be different. For example, the preset time duration is 15 days, and the historical data acquisition period is the previous 15 days of the current prediction time, that is, the canopy coverage of the to-be-predicted crop area in the previous 15 days of the current prediction time is acquired, in other words, the canopy coverage includes the canopy coverage of 15 days.
Here, the remote sensing image data may be acquired by a remote sensing satellite, and the type of the remote sensing satellite is not particularly limited in the embodiment of the present invention, for example, a sentry satellite, a Planet satellite, and the like.
In one embodiment, the remote sensing image data is sentinel image data, so that the high-resolution characteristic of the sentinel image data is utilized, the block-level phenological period prediction can be performed, and the phenological period prediction accuracy is improved.
In another embodiment, the remote sensing image data is Planet image data, so that the high-frequency characteristic of the Planet image data is utilized to ensure that the remote sensing image data can completely cover each growth period node of the crops, and the accuracy of the phenological period prediction is improved. Especially for areas with more rainwater, months with more rainwater, or areas with more cloud layers and months with more cloud layers, the forecast of the phenological period can be carried out by ensuring the remote sensing image data with 2-3 periods per week.
In an embodiment, a sampling frequency may be set, and then remote sensing image data of an object region to be predicted is obtained based on the sampling frequency. For example, if the sampling frequency is once a day, the time phase interval of the remote sensing image data is 1 day, and the remote sensing image data is acquired day by day. It can be understood that the sampling frequency is set to be high frequency, data can be continuously monitored at high frequency, remote sensing image data can be ensured to completely cover each growth period node of crops, and the accuracy of the forecast of the phenological period is further improved.
In one embodiment, in order to reduce the data volume of the remote sensing image data, the first-stage remote sensing image data can be selected for 5 days in a time period with good weather, so that the data volume is reduced, and the prediction efficiency of the weather stage is improved.
In one embodiment, a historical data acquisition period is determined based on the current predicted time and a preset time length; and acquiring remote sensing image data of the object area to be predicted during the acquisition of the historical data. The preset time can be set according to actual needs, for example, 15 days; the preset time lengths of different crops can be different, the preset time lengths of the same crop of different varieties can be different, and the preset time lengths of the same crop in different regions can be different. For example, the preset time length is 15 days, and the historical data acquisition period is 15 days before the current prediction time, that is, the remote sensing image data of the object region to be predicted 15 days before the current prediction time is acquired, in other words, the remote sensing image data includes 15 days.
Here, the vegetation index may include, but is not limited to, at least one of: NDVI (Normalized Difference Vegetation Index), NDWI (Normalized Difference Water Index), EVI (Enhanced Vegetation Index), NIR (Near Infrared), b/r (ratio of reflectance of blue band to reflectance of red band), and the like. For example, the NDVI is obtained by performing atmospheric correction, radiometric calibration, band calculation, and the like on the remote sensing image data.
In one embodiment, the vegetation index includes NDVI, which improves the accuracy of the phenological phase prediction since NDVI can well characterize the growth and development of the crop.
Here, the acquisition source of the weather data may include historical weather data and weather forecast data. The meteorological data may include, but is not limited to, at least one of: average temperature, maximum temperature, minimum temperature, average humidity, duration of sunshine, cumulative precipitation, maximum wind speed, and the like.
In one embodiment, considering that temperature has a greater impact on crop growth, the meteorological data may include: average temperature, maximum temperature, minimum temperature.
In one embodiment, the meteorological data may include, in view of the greater effect of humidity on crop growth: average humidity.
In one embodiment, the meteorological data may include, in view of the greater effect of illumination on crop growth: the duration of the sun exposure.
In one embodiment, the meteorological data may include, in view of the greater effect of precipitation on crop growth: accumulating the precipitation.
In one embodiment, the meteorological data may include, in view of the greater effect of wind speed on crop growth: the maximum wind speed.
In some embodiments, the meteorological data is normalized, so that the meteorological data after being normalized is input into the phenological period prediction model, and therefore accuracy of phenological period prediction is improved.
In one embodiment, a historical data acquisition period is determined based on the current predicted time and a preset duration; meteorological data of an area to be predicted during historical data acquisition is acquired. The preset time can be set according to actual needs, for example, 15 days; the preset time lengths of different crops can be different, the preset time lengths of the same crop of different varieties can be different, and the preset time lengths of the same crop in different regions can be different. For example, the preset time period is 15 days, and the historical data acquisition period is 15 days before the current prediction time, that is, the meteorological data of the object area to be predicted is acquired 15 days before the current prediction time, in other words, the meteorological data includes the meteorological data of 15 days.
It should be noted that the factors affecting the growth of the crops include not only the weather data but also the factors of the crops themselves, such as the variety of the crops, the characteristics of the crops, the actual growth of the crops, and the like. Therefore, the embodiment of the invention also obtains the canopy coverage and the remote sensing image data, and combines the canopy coverage and the remote sensing image data together to predict the phenological period, thereby improving the accuracy of phenological period prediction.
And 120, inputting the canopy coverage, the vegetation index and the meteorological data into a phenological period prediction model to obtain a phenological period prediction result output by the phenological period prediction model.
The phenological period prediction results comprise phenological period results and phenological period interval days, the phenological period results are used for representing the current phenological period where the target crops in the crop area to be predicted are located, and the phenological period interval days are used for representing the interval days between the current prediction time and the next phenological period of the current phenological period.
Here, the phenological period prediction model is trained based on the sample canopy coverage, the sample vegetation index, the sample meteorological data, and the sample phenological period prediction result corresponding to each sample data. The sample vegetation index is determined and obtained based on the sample remote sensing image data. The sample phenological period prediction result comprises a sample phenological period result and a sample phenological period interval days.
In an embodiment, since the phenological period prediction model is a multiple-input multiple-output model, the phenological period prediction model may be an LSTM (Long short-term memory) model, so that the context relationship is fully considered, and the phenological period prediction accuracy is further improved.
In one embodiment, the sample canopy coverage includes day-by-day sample canopy coverage, thereby improving the data volume of the sample canopy coverage, further improving the model training effect, and finally improving the accuracy of the phenological period prediction.
The sample remote sensing image data can be acquired through a remote sensing satellite, and the type of the remote sensing satellite is not particularly limited in the embodiment of the invention, for example, a sentry satellite, a Planet satellite and the like.
In one embodiment, the sample remote sensing image data is sentinel image data, so that the high-resolution characteristic of the sentinel image data is utilized, the model training at the ground level can be performed, the forecast of the physical climate period at the ground level can be performed, and the accuracy of the forecast of the physical climate period is finally improved.
In another embodiment, the sample remote sensing image data is Planet image data, so that the high-frequency characteristic of the Planet image data is utilized, the sample remote sensing image data can completely cover each growth period node of the crop, the model training effect is improved, and the accuracy of the phenological period prediction is improved.
In one embodiment, a sampling frequency may be set, and the sample remote sensing image data may be obtained based on the sampling frequency. For example, if the sampling frequency is once a day, the time interval of the sample remote sensing image data is 1 day, and then the sample remote sensing image data day by day is obtained. It can be understood that the sampling frequency is set to be high frequency, data can be continuously monitored at high frequency, and each growth period node of a crop can be completely covered by sample remote sensing image data, so that the model training effect is improved, and the accuracy of the phenological period prediction is improved.
In one embodiment, in order to reduce the data volume of the sample remote sensing image data, the first-period sample remote sensing image data can be selected for 5 days in a time period with good weather, so that the data volume is reduced, and the model training efficiency is improved.
Wherein the sample vegetation index may include, but is not limited to, at least one of: sample NDVI, sample NDWI, sample EVI, sample NIR, sample b/r, etc., and the type of vegetation index is not particularly limited in the embodiments of the present invention.
The acquisition sources of the sample meteorological data can comprise historical meteorological data and weather forecast data. The sample meteorological data may include, but is not limited to, at least one of: the average temperature of the sample, the maximum temperature of the sample, the minimum temperature of the sample, the average humidity of the sample, the sunshine duration of the sample, the accumulated precipitation of the sample, the maximum wind speed of the sample and the like.
In some embodiments, the sample meteorological data is normalized for model training based on the normalized sample meteorological data, so that the model training effect is improved, and the accuracy of the prediction of the phenological period is improved.
It will be appreciated that the prediction of the phenological period at the site level may provide accurate guidance, for example to provide a basis for accurate planting. The interval days between the current prediction time and the next phenological period of the current phenological period are obtained through prediction, so that more accurate guidance can be provided, for example, accurate planting basis is provided.
According to the phenological period prediction method provided by the embodiment of the invention, canopy coverage, remote sensing image data and meteorological data of a to-be-predicted region are obtained, and the canopy coverage, the vegetation index extracted based on the remote sensing image data and the meteorological data are input into a phenological period prediction model, so that not only temperature data but also more comprehensive meteorological data are considered, and phenological period prediction is carried out by combining the canopy coverage and the vegetation index, and the accuracy of phenological period prediction is further improved; meanwhile, compared with meteorological data, the remote sensing image data has higher resolution, so that the block-level phenological period prediction can be carried out, and the phenological period prediction accuracy is further improved; in addition, the phenological period prediction result output by the phenological period prediction model comprises a phenological period result and phenological period interval days, so that the current phenological period where the target crop in the crop area to be predicted is located can be predicted, the interval days between the current prediction time and the next phenological period of the current phenological period can be predicted, and the phenological period prediction accuracy is further improved.
Based on the above embodiment, fig. 2 is a second flowchart of the method for predicting a phenological period provided by the present invention, as shown in fig. 2, the step 120 includes:
and 121, inputting the canopy coverage, the vegetation index and the meteorological data into a feature extraction layer of a phenological period prediction model to obtain a feature vector output by the feature extraction layer.
Here, the specific structure of the feature extraction layer may be set according to actual needs, and this is not specifically limited in the embodiment of the present invention.
In an embodiment, since the feature extraction layer is a multi-input feature extraction layer, the feature extraction layer may be an LSTM feature extraction layer, so that the context is fully considered, a feature vector considering the context is obtained, and the accuracy of the prediction of the phenological period is further improved.
And step 122, inputting the feature vector into a first phenological period prediction layer of the phenological period prediction model to obtain a phenological period result output by the first phenological period prediction layer.
Here, the first phenological period prediction layer is configured to predict phenological period results based on the feature vectors. The specific structure of the first phenological period prediction layer can be set according to actual needs.
In some embodiments, step 122 includes:
and inputting the feature vector into a phenological period classification layer of the phenological period prediction model to obtain a phenological period result output by the phenological period classification layer.
Here, the phenological period classification layer is configured to perform phenological period classification prediction based on the feature vector to obtain a prediction probability of each phenological period type, and determine a phenological period corresponding to the maximum prediction probability as a phenological period result. The specific structure of the phenological period classification layer can be set according to actual needs.
The phenological period classification layer is obtained based on a first loss function training, and the first loss function can be set according to actual needs, for example, a cross entropy loss function. For ease of understanding, this first loss function is as follows:
in the formula, K represents the number of species in a phenological period,
the first kind of the phenological period is shown,
denotes the first
The true value of the phenological period of the species,
denotes the first
The predicted value of the species phenological period.
It can be understood that the accuracy of the phenological period prediction can be further improved by predicting the phenological period result through the phenological period classification layer.
And 123, inputting the feature vector to a second candidate period prediction layer of the candidate period prediction model to obtain the candidate period interval days output by the second candidate period prediction layer.
Here, the second candidate prediction layer is configured to predict the number of candidate interval days based on the feature vector. The specific structure of the second candidate prediction layer can be set according to actual needs.
In some embodiments, the step 123 comprises:
and inputting the characteristic vector into a phenological period regression layer of the phenological period prediction model to obtain phenological period interval days output by the phenological period regression layer.
Here, the phenological period regression layer is configured to perform regression prediction based on the feature vector to obtain a regression result, and then determine phenological period interval days based on the regression result. The concrete structure of the phenological period regression layer can be set according to actual needs.
The phenological period regression layer is obtained based on a second loss function, which may be set according to actual needs, for example, as follows:
in the formula (I), the compound is shown in the specification,
the number of samples is represented as a function of,
the number of samples is indicated to be the first,
is shown as
The true value of the number of samples,
is shown as
Regression results for individual samples.
It can be understood that, the accuracy of the phenological period prediction can be further improved by predicting the phenological period interval days through the phenological period regression layer.
According to the phenological period prediction method provided by the embodiment of the invention, the phenological period result is predicted through the first phenological period prediction layer, and the phenological period interval days are predicted through the second phenological period prediction layer, so that a feature extraction layer is shared, namely, not only the phenological period result can be predicted through a model, but also the phenological period interval days can be predicted, and the phenological period prediction accuracy is further improved.
Based on any of the above embodiments, fig. 3 is a third schematic flow chart of the phenological period prediction method provided by the present invention, as shown in fig. 3, the canopy coverage is obtained based on the following steps:
step 310, determining the sowing time of the target crop, the canopy growth coefficient of the target crop and the initial canopy size of the target crop at the sowing time.
Here, the sowing time is used to represent the sowing date of the target crop. The canopy growth coefficients (increased soil coverage or degree of growth day to day) vary from crop to crop.
And 320, determining the canopy coverage of the crop area to be predicted based on the seeding time, the canopy growth coefficient and the initial canopy size.
Specifically, determining canopy coverage of the crop area to be predicted based on the seeding time, the current prediction time, the canopy growth coefficient and the initial canopy size.
In some embodiments, canopy coverage may be determined based on a calculation formula.
Further, the calculation formulas of the canopy coverage at different stages are different. In one embodiment, the calculation of the canopy coverage is divided into two stages, the first stage being where the canopy coverage is less than or equal to half the maximum canopy coverage and the second stage being where the canopy coverage is greater than half the maximum canopy coverage.
In one embodiment, the calculation formula for the first stage is as follows:
where CC denotes the canopy coverage at the current predicted time,
the method comprises the steps of representing the size of an initial canopy, representing the time length from the seeding time to the current prediction time, and representing a canopy growth coefficient by CGC.
In one embodiment, the calculation formula for the second stage is as follows:
where CC denotes the canopy coverage at the current predicted time,
the maximum canopy coverage is shown as,
the method comprises the steps of representing the size of an initial canopy, representing the time length from the seeding time to the current prediction time, and representing a canopy growth coefficient by CGC.
It should be noted that the canopy coverage day by day can be obtained based on the above formula, so that the data volume of the canopy coverage is increased, and the accuracy of the phenological period prediction is further improved.
In one embodiment, the canopy coverage may be determined by a model of the crop growth mechanism. Specifically, the sowing time, the canopy growth coefficient and the initial canopy size are input into the crop growth mechanism model, and canopy coverage output by the crop growth mechanism model is obtained. The crop growth mechanism model can be set according to actual needs, for example, the aquacrop mechanism model.
Furthermore, the canopy coverage output by the crop growth mechanism model can be corrected according to the actually measured canopy coverage, so that the accuracy of the crop growth mechanism model is improved, the accuracy of the canopy coverage is further improved, and the accuracy of the phenological period prediction is further improved finally.
The phenological period prediction method provided by the embodiment of the invention determines the sowing time of the target crop, the canopy growth coefficient of the target crop and the initial canopy size of the target crop at the sowing time; the canopy coverage of a crop area to be predicted is determined based on the seeding time, the canopy growth coefficient and the initial canopy size, support is provided for obtaining the canopy coverage, the canopy coverage is combined to predict the phenological period, and therefore accuracy of phenological period prediction is improved.
Based on any one of the above embodiments, in the method, the remote sensing image data is obtained based on the following steps:
obtaining remote sensing data of the object area to be predicted;
and screening the remote sensing data based on the cloud coverage rate and the preset cloud coverage rate of the remote sensing data to obtain the remote sensing image data of the object area to be predicted.
Here, the remote sensing data may be acquired by a remote sensing satellite, and the type of the remote sensing satellite is not particularly limited in the embodiment of the present invention, for example, a sentry satellite, a Planet satellite, and the like.
In one embodiment, the remote sensing data is sentinel data, so that the high-resolution characteristic of the sentinel data is utilized, the block-level phenological period prediction can be carried out, and the phenological period prediction accuracy is improved.
In another embodiment, the remote sensing data is Planet data, so that the high-frequency characteristic of the Planet data is utilized, the remote sensing data can be ensured to completely cover each growth period node of the crops, and the accuracy of the forecast of the phenological period is improved. Especially for areas with more rainwater, months with more rainwater, or areas with more cloud cover, months with more cloud cover, the forecast of the phenological period can be carried out by ensuring the remote sensing data of 2-3 periods per week.
In one embodiment, a sampling frequency may be set, and remote sensing data of an object area to be predicted may be obtained based on the sampling frequency. For example, if the sampling frequency is once a day, the time interval of the remote sensing data is 1 day, and then the remote sensing data day by day is acquired. It can be understood that the sampling frequency is set to be high frequency, data can be continuously monitored at high frequency, remote sensing image data can completely cover all growth period nodes of crops, and therefore the accuracy of phenological period prediction is improved.
In one embodiment, in order to reduce the data volume of the remote sensing data, the first-stage remote sensing data can be selected in 5 days in a time period with good weather, so that the data volume is reduced, and the prediction efficiency of the weather period is improved.
In one embodiment, a historical data acquisition period is determined based on the current predicted time and a preset time length; and acquiring remote sensing data of the area to be predicted during the acquisition of the historical data. The preset time can be set according to actual needs, for example, 15 days; the preset time lengths of different crops can be different, the preset time lengths of the same crop of different varieties can be different, and the preset time lengths of the same crop in different regions can be different. For example, the preset time length is 15 days, and the historical data acquisition period is 15 days before the current prediction time, that is, the remote sensing data of the object area to be predicted 15 days before the current prediction time is acquired, in other words, the remote sensing data includes 15 days.
Here, the preset cloud coverage may be set according to actual needs, for example, 5%.
Here, the remote-sensing image data includes remote-sensing image data with a cloud coverage rate smaller than a preset cloud coverage rate, so that the remote-sensing image data does not include remote-sensing image data with a cloud layer as much as possible, the influence of the cloud layer on the prediction of the phenological period is reduced, and the accuracy of the phenological period prediction is further improved. Specifically, the remote sensing data with the cloud coverage rate larger than or equal to the preset cloud coverage rate in the remote sensing data are removed, and the remote sensing data with the cloud coverage rate smaller than the preset cloud coverage rate are reserved.
The phenological period prediction method provided by the embodiment of the invention screens the remote sensing data based on the cloud coverage rate and the preset cloud coverage rate of the remote sensing data, so that the remote sensing image data does not include the remote sensing image data with a cloud layer as far as possible, the influence of the cloud layer on vegetation index extraction is further reduced, the influence of the cloud layer on phenological period prediction is further reduced, and the phenological period prediction accuracy is further improved.
Based on any one of the above embodiments, in the method, the remote sensing image data is obtained based on the following steps:
obtaining remote sensing data of the to-be-predicted crop area;
inputting the remote sensing data into a cloud layer identification model to obtain a cloud layer area output by the cloud layer identification model, and screening the remote sensing data based on the cloud layer area to obtain remote sensing image data of the object area to be predicted.
Here, the remote sensing data may be acquired by a remote sensing satellite, and the type of the remote sensing satellite is not particularly limited in the embodiment of the present invention, for example, a sentry satellite, a Planet satellite, and the like.
In one embodiment, the remote sensing data is sentinel data, so that the high-resolution characteristic of the sentinel data is utilized, the block-level phenological period prediction can be carried out, and the phenological period prediction accuracy is improved.
In another embodiment, the remote sensing data is Planet data, so that the high-frequency characteristic of the Planet data is utilized, the remote sensing data can be ensured to completely cover each growth period node of the crops, and the accuracy of the forecast of the phenological period is improved. Especially for areas with more rainwater, months with more rainwater, or areas with more cloud layers and months with more cloud layers, the forecast of the phenological period can be carried out by ensuring the remote sensing data of 2-3 periods per week.
In one embodiment, a sampling frequency may be set, and remote sensing data of an object region to be predicted may be obtained based on the sampling frequency. For example, if the sampling frequency is once a day, the time interval of the remote sensing data is 1 day, and then the remote sensing data day by day is obtained. It can be understood that the sampling frequency is set to be high frequency, data can be continuously monitored at high frequency, remote sensing image data can completely cover all growth period nodes of crops, and therefore accuracy of phenological period prediction is improved.
In one embodiment, in order to reduce the data volume of the remote sensing data, the first-stage remote sensing data can be selected in 5 days in a time period with good weather, so that the data volume is reduced, and the prediction efficiency of the weather period is improved.
In one embodiment, a historical data acquisition period is determined based on the current predicted time and a preset duration; and acquiring remote sensing data of the object region to be predicted during the acquisition of the historical data. The preset time can be set according to actual needs, for example, 15 days; the preset time lengths of different crops can be different, the preset time lengths of the same crop of different varieties can be different, and the preset time lengths of the same crop in different regions can be different. For example, the preset time length is 15 days, and the historical data acquisition period is 15 days before the current prediction time, that is, the remote sensing data of the object area to be predicted 15 days before the current prediction time is acquired, in other words, the remote sensing data includes 15 days.
The cloud layer identification model is used for carrying out cloud layer identification on the remote sensing data to obtain a cloud layer area in the crop area to be predicted. The cloud layer identification model is obtained by training based on sample remote sensing data and a sample cloud layer area corresponding to the sample remote sensing data. And the sample cloud layer area is obtained by labeling the sample remote sensing data.
In one embodiment, the sample telemetry data includes telemetry data for different time periods. In another embodiment, the sample telemetry data includes telemetry data for different plots. In another embodiment, the sample telemetry data includes telemetry data for different time periods, and the sample telemetry data includes telemetry data for different plots.
According to the method for predicting the phenological period, provided by the embodiment of the invention, the remote sensing data is screened based on the cloud layer area output by the cloud layer identification model, so that the remote sensing image data without the cloud layer is obtained, and further, the influence of the cloud layer on vegetation index extraction is reduced, so that the influence of the cloud layer on phenological period prediction is reduced, and the accuracy of phenological period prediction is further improved.
In any of the above embodiments, the meteorological data includes at least one of an average temperature, a maximum temperature, a minimum temperature, an average humidity, a duration of sunshine, a cumulative precipitation, and a maximum wind speed.
In one embodiment, the average temperature is daily average temperature, the highest temperature is daily highest temperature, the lowest temperature is daily lowest temperature, the average humidity is daily average humidity, the cumulative precipitation is daily cumulative precipitation, and the maximum wind speed is daily maximum wind speed, so that day-by-day meteorological data are obtained, the data volume of the meteorological data is increased, and the accuracy of the forecast of the phenological period is improved.
According to the phenological period prediction method provided by the embodiment of the invention, more comprehensive meteorological data are considered, phenological period prediction is carried out, and the phenological period prediction accuracy is further improved.
Based on any of the above embodiments, in this method, in step 110, obtaining canopy coverage, remote sensing image data, and meteorological data of the area to be predicted includes:
determining a historical data acquisition period based on the current prediction time and a preset time length;
and acquiring canopy coverage, remote sensing image data and meteorological data of the object area to be predicted during the acquisition of the historical data.
Here, the current prediction time is the current time at which the phenological period prediction is performed.
Here, the preset time period may be set according to actual needs, for example, 15 days. The preset time lengths of different crops can be different, the preset time lengths of the same crop of different varieties can be different, and the preset time lengths of the same crop in different regions can be different.
For convenience of understanding, for example, the preset time duration is 15 days, the historical data acquisition period is the previous 15 days of the current prediction time, and the canopy coverage, the remote sensing image data and the meteorological data of the object region to be predicted in the previous 15 days of the current prediction time are further acquired, in other words, the canopy coverage, the remote sensing image data and the meteorological data of the object region to be predicted in the previous 15 days of the current prediction time are included.
According to the phenological period prediction method provided by the embodiment of the invention, based on the current prediction time and the preset duration, the historical data acquisition period is determined, so that canopy coverage, remote sensing image data and meteorological data of the object region to be predicted in the historical data acquisition period are acquired, and therefore, the data input to the phenological period prediction model is ensured to be the data most relevant to the current prediction time, and the phenological period prediction accuracy is further improved.
The following describes the device for predicting a phenological period according to the present invention, and the device for predicting a phenological period described below and the method for predicting a phenological period described above may be referred to in correspondence with each other.
Fig. 4 is a schematic structural diagram of a phenological period prediction apparatus provided in the present invention, as shown in fig. 4, the phenological period prediction apparatus includes:
the acquiring module 410 is used for acquiring canopy coverage, remote sensing image data and meteorological data of a to-be-predicted crop area, and performing vegetation index extraction processing on the remote sensing image data to obtain a vegetation index;
the prediction module 420 is configured to input the canopy coverage, the vegetation index and the meteorological data into a phenological period prediction model to obtain a phenological period prediction result output by the phenological period prediction model;
the phenological period prediction result comprises a phenological period result and phenological period interval days, the phenological period result is used for representing the current phenological period where the target crop in the crop area to be predicted is located, and the phenological period interval days are used for representing the interval days between the current prediction time and the next phenological period of the current phenological period.
According to the phenological period prediction device provided by the embodiment of the invention, canopy coverage, remote sensing image data and meteorological data of a to-be-predicted region are obtained, and the canopy coverage, the vegetation index extracted based on the remote sensing image data and the meteorological data are input into a phenological period prediction model, so that not only temperature data but also more comprehensive meteorological data are considered, and phenological period prediction is carried out by combining the canopy coverage and the vegetation index, and the accuracy of phenological period prediction is further improved; meanwhile, compared with meteorological data, the remote sensing image data has higher resolution, so that the prediction of the phenological period at the block level can be carried out, and the accuracy of the phenological period prediction is further improved; in addition, the phenological period prediction result output by the phenological period prediction model comprises a phenological period result and phenological period interval days, so that the current phenological period of the target crop in the crop area to be predicted can be predicted, the interval days between the current prediction time and the next phenological period of the current phenological period can be predicted, and the phenological period prediction accuracy is further improved.
Based on any of the above embodiments, the prediction module 420 includes:
the feature extraction unit is used for inputting the canopy coverage, the vegetation index and the meteorological data into a feature extraction layer of a phenological period prediction model to obtain a feature vector output by the feature extraction layer;
the first prediction unit is used for inputting the feature vector to a first phenological period prediction layer of the phenological period prediction model to obtain a phenological period result output by the first phenological period prediction layer;
and the second prediction unit is used for inputting the feature vector to a second candidate prediction layer of the candidate prediction model to obtain the candidate interval days output by the second candidate prediction layer.
According to any of the above embodiments, the first prediction unit is further configured to:
inputting the feature vector into a phenological period classification layer of the phenological period prediction model to obtain a phenological period result output by the phenological period classification layer;
the second prediction unit is further to:
and inputting the characteristic vector into a phenological period regression layer of the phenological period prediction model to obtain phenological period interval days output by the phenological period regression layer.
Based on any embodiment, the apparatus further comprises a canopy acquisition module, configured to:
determining a sowing time of the target crop, a canopy growth coefficient of the target crop, and an initial canopy size of the target crop at the sowing time;
and determining the canopy coverage of the crop area to be predicted based on the sowing time, the canopy growth coefficient and the initial canopy size.
Based on any one of the above embodiments, the apparatus further comprises a remote sensing acquisition module, the remote sensing acquisition module is configured to:
obtaining remote sensing data of the object area to be predicted;
screening the remote sensing data based on the cloud coverage rate and the preset cloud coverage rate of the remote sensing data to obtain remote sensing image data of the to-be-predicted object area; or the like, or, alternatively,
and inputting the remote sensing data into a cloud layer identification model to obtain a cloud layer region output by the cloud layer identification model, and screening the remote sensing data based on the cloud layer region to obtain the remote sensing image data of the to-be-predicted object region.
In any of the above embodiments, the meteorological data includes at least one of an average temperature, a maximum temperature, a minimum temperature, an average humidity, a duration of sunshine, a cumulative precipitation, and a maximum wind speed.
Based on any of the above embodiments, the obtaining module 410 includes:
the period determining unit is used for determining a historical data acquisition period based on the current prediction time and the preset time length;
and the data acquisition unit is used for acquiring canopy coverage, remote sensing image data and meteorological data of the object area to be predicted during the acquisition of the historical data.
Fig. 5 illustrates a physical structure diagram of an electronic device, which may include, as shown in fig. 5: a processor (processor) 510, a communication Interface (Communications Interface) 520, a memory (memory) 530 and a communication bus 540, wherein the processor 510, the communication Interface 520 and the memory 530 communicate with each other via the communication bus 540. Processor 510 may invoke logic instructions in memory 530 to perform a method for predicting a period of phenology, the method comprising: acquiring canopy coverage, remote sensing image data and meteorological data of a crop area to be predicted, and performing vegetation index extraction processing on the remote sensing image data to obtain a vegetation index; inputting the canopy coverage, the vegetation index and the meteorological data into a phenological period prediction model to obtain a phenological period prediction result output by the phenological period prediction model; the phenological period prediction result comprises a phenological period result and phenological period interval days, the phenological period result is used for representing the current phenological period where the target crop in the crop area to be predicted is located, and the phenological period interval days are used for representing the interval days between the current prediction time and the next phenological period of the current phenological period.
In addition, the logic instructions in the memory 530 may be implemented in the form of software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as a stand-alone product. 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 removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk, and various media capable of storing program codes.
In yet another aspect, the present invention also provides a non-transitory computer-readable storage medium having stored thereon a computer program, which when executed by a processor, implements a method for predicting a phenological period provided by the above methods, the method comprising: acquiring canopy coverage, remote sensing image data and meteorological data of a crop area to be predicted, and performing vegetation index extraction processing on the remote sensing image data to obtain a vegetation index; inputting the canopy coverage, the vegetation index and the meteorological data into a phenological period prediction model to obtain a phenological period prediction result output by the phenological period prediction model; the phenological period prediction result comprises a phenological period result and phenological period interval days, the phenological period result is used for representing the current phenological period where the target crop in the crop area to be predicted is located, and the phenological period interval days are used for representing the interval days between the current prediction time and the next phenological period of the current phenological period.
The above-described embodiments of the apparatus are merely illustrative, and 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 modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. Based on the understanding, the above technical solutions substantially or otherwise contributing to the prior art may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method according to the various embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for predicting a phenological period, comprising:
acquiring canopy coverage, remote sensing image data and meteorological data of a to-be-predicted crop area, and performing vegetation index extraction processing on the remote sensing image data to obtain a vegetation index;
inputting the canopy coverage, the vegetation index and the meteorological data into a phenological period prediction model to obtain a phenological period prediction result output by the phenological period prediction model;
the phenological period prediction result comprises a phenological period result and phenological period interval days, the phenological period result is used for representing the current phenological period where the target crop in the crop area to be predicted is located, and the phenological period interval days are used for representing the interval days between the current prediction time and the next phenological period of the current phenological period.
2. The phenological period prediction method according to claim 1, wherein the step of inputting the canopy coverage, the vegetation index and the meteorological data into a phenological period prediction model to obtain a phenological period prediction result output by the phenological period prediction model includes:
inputting the canopy coverage, the vegetation index and the meteorological data into a feature extraction layer of a phenological period prediction model to obtain a feature vector output by the feature extraction layer;
inputting the feature vector into a first phenological period prediction layer of the phenological period prediction model to obtain a phenological period result output by the first phenological period prediction layer;
and inputting the feature vector into a second waiting period prediction layer of the waiting period prediction model to obtain the number of days of the waiting period interval output by the second waiting period prediction layer.
3. The method according to claim 2, wherein the inputting the feature vector into a first phenological period prediction layer of the phenological period prediction model to obtain a phenological period result output by the first phenological period prediction layer includes:
inputting the feature vector into a phenological period classification layer of the phenological period prediction model to obtain a phenological period result output by the phenological period classification layer;
the step of inputting the feature vector into a second waiting period prediction layer of the waiting period prediction model to obtain the number of days of the waiting period interval output by the second waiting period prediction layer comprises the following steps:
and inputting the characteristic vector into a phenological period regression layer of the phenological period prediction model to obtain phenological period interval days output by the phenological period regression layer.
4. The phenological period prediction method according to claim 1, wherein the canopy coverage is obtained based on the steps of:
determining the sowing time of the target crop, the canopy growth coefficient of the target crop and the initial canopy size of the target crop at the sowing time;
determining canopy coverage of the crop area to be predicted based on the seeding time, the canopy growth coefficient and the initial canopy size.
5. The method of predicting a phenological period of claim 1, wherein the remote sensing image data is obtained based on the following steps:
obtaining remote sensing data of the object area to be predicted;
screening the remote sensing data based on the cloud coverage rate and the preset cloud coverage rate of the remote sensing data to obtain remote sensing image data of the to-be-predicted object area; or the like, or, alternatively,
inputting the remote sensing data into a cloud layer identification model to obtain a cloud layer area output by the cloud layer identification model, and screening the remote sensing data based on the cloud layer area to obtain remote sensing image data of the object area to be predicted.
6. The method of claim 1, wherein the meteorological data comprises at least one of an average temperature, a maximum temperature, a minimum temperature, an average humidity, a duration of sunshine, a cumulative precipitation, and a maximum wind speed.
7. The phenological period prediction method according to any one of claims 1 to 6, wherein the obtaining canopy coverage, remote sensing image data and meteorological data of the area to be predicted comprises:
determining a historical data acquisition period based on the current prediction time and a preset time length;
and acquiring canopy coverage, remote sensing image data and meteorological data of the object area to be predicted during the acquisition of the historical data.
8. A phenological period prediction device, characterized by comprising:
the system comprises an acquisition module, a prediction module and a prediction module, wherein the acquisition module is used for acquiring canopy coverage, remote sensing image data and meteorological data of a to-be-predicted crop area, and carrying out vegetation index extraction processing on the remote sensing image data to obtain a vegetation index;
the prediction module is used for inputting the canopy coverage, the vegetation index and the meteorological data into a phenological period prediction model to obtain a phenological period prediction result output by the phenological period prediction model;
the phenological period prediction result comprises a phenological period result and phenological period interval days, the phenological period result is used for representing the current phenological period where the target crop in the crop area to be predicted is located, and the phenological period interval days are used for representing the interval days between the current prediction time and the next phenological period of the current phenological period.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the method of predicting a waiting period as claimed in any one of claims 1 to 7.
10. A non-transitory computer-readable storage medium, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the method for predicting a phenological period according to any of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211576214.4A CN115577866A (en) | 2022-12-09 | 2022-12-09 | Method and device for predicting waiting period, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211576214.4A CN115577866A (en) | 2022-12-09 | 2022-12-09 | Method and device for predicting waiting period, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115577866A true CN115577866A (en) | 2023-01-06 |
Family
ID=84590665
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211576214.4A Pending CN115577866A (en) | 2022-12-09 | 2022-12-09 | Method and device for predicting waiting period, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115577866A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116052141A (en) * | 2023-03-30 | 2023-05-02 | 北京市农林科学院智能装备技术研究中心 | Crop growth period identification method, device, equipment and medium |
| CN116151445A (en) * | 2023-01-31 | 2023-05-23 | 南京信息工程大学 | Initial flowering phase prediction method and system based on LSTM neural network model |
| CN117236519A (en) * | 2023-11-10 | 2023-12-15 | 中化现代农业有限公司 | Water and fertilizer regulation and control method and device, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106845428A (en) * | 2017-01-26 | 2017-06-13 | 中国科学院遥感与数字地球研究所 | A kind of crop yield remote sensing estimation method and system |
| CN110243406A (en) * | 2019-06-21 | 2019-09-17 | 武汉思众空间信息科技有限公司 | Crop Estimation Method, device, electronic equipment and storage medium |
| WO2021007352A1 (en) * | 2019-07-08 | 2021-01-14 | Indigo Ag, Inc. | Crop yield forecasting models |
| CN114529097A (en) * | 2022-02-26 | 2022-05-24 | 黑龙江八一农垦大学 | Multi-scale crop phenological period remote sensing dimensionality reduction prediction method |
-
2022
- 2022-12-09 CN CN202211576214.4A patent/CN115577866A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106845428A (en) * | 2017-01-26 | 2017-06-13 | 中国科学院遥感与数字地球研究所 | A kind of crop yield remote sensing estimation method and system |
| CN110243406A (en) * | 2019-06-21 | 2019-09-17 | 武汉思众空间信息科技有限公司 | Crop Estimation Method, device, electronic equipment and storage medium |
| WO2021007352A1 (en) * | 2019-07-08 | 2021-01-14 | Indigo Ag, Inc. | Crop yield forecasting models |
| CN114529097A (en) * | 2022-02-26 | 2022-05-24 | 黑龙江八一农垦大学 | Multi-scale crop phenological period remote sensing dimensionality reduction prediction method |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116151445A (en) * | 2023-01-31 | 2023-05-23 | 南京信息工程大学 | Initial flowering phase prediction method and system based on LSTM neural network model |
| CN116052141A (en) * | 2023-03-30 | 2023-05-02 | 北京市农林科学院智能装备技术研究中心 | Crop growth period identification method, device, equipment and medium |
| CN116052141B (en) * | 2023-03-30 | 2023-06-27 | 北京市农林科学院智能装备技术研究中心 | Crop growth period identification method, device, equipment and medium |
| CN117236519A (en) * | 2023-11-10 | 2023-12-15 | 中化现代农业有限公司 | Water and fertilizer regulation and control method and device, electronic equipment and storage medium |
| CN117236519B (en) * | 2023-11-10 | 2024-04-09 | 中化现代农业有限公司 | Water and fertilizer regulation and control method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115577866A (en) | Method and device for predicting waiting period, electronic equipment and storage medium | |
| CN111008733B (en) | Crop growth control method and system | |
| Robertson et al. | Predicting growth and development of pigeonpea: a simulation model | |
| CN114743100B (en) | Agricultural product growth condition monitoring method and system | |
| CN112070297A (en) | Weather index prediction method, device, equipment and storage medium for farming activities | |
| CN116543316B (en) | Method for identifying turf in paddy field by utilizing multi-time-phase high-resolution satellite image | |
| CN112819227A (en) | County scale winter wheat yield per unit prediction method and system | |
| CN114662742A (en) | Dynamic decision method and device for field irrigation and fertilization system | |
| CN112836623A (en) | Facility tomato farming decision auxiliary method and device | |
| CN112580491A (en) | Method and device for determining growth stage of crop and nonvolatile storage device | |
| CN113269716A (en) | Multi-source remote sensing data-based bergamot pear maturity remote sensing prediction method | |
| CN116681169A (en) | Method for evaluating influence of extreme climate on crop yield | |
| Dinh et al. | Statistical analysis of the weather impact on Robusta Coffee Yield in Vietnam | |
| Zhang et al. | A crop variety yield prediction system based on variety yield data compensation | |
| CA2761682C (en) | Real-time process for targeting trait phenotyping of plant breeding experiments | |
| CN116579521B (en) | Yield prediction time window determining method, device, equipment and readable storage medium | |
| CN116482041B (en) | Rice heading period nondestructive rapid identification method and system based on reflection spectrum | |
| CN115620151B (en) | Method and device for identifying phenological period, electronic equipment and storage medium | |
| CN116050679B (en) | Irrigation decision-making method and device and electronic equipment | |
| CN114219227A (en) | Method and system for precise fertilization decision and plot level display | |
| CN114723118A (en) | Insect pest early warning system based on Internet of things | |
| CN114842338A (en) | Wheat stripe rust prediction method and device based on coupling of remote sensing and meteorological data | |
| 丸山篤志 et al. | Modeling of phenological development in winter wheat to estimate the timing of heading and maturity based on daily mean air temperature and photoperiod | |
| CN111160151A (en) | Sugarcane continuous cropping range extraction method, device and equipment based on radar time sequence image | |
| CN115660236B (en) | Crop weather period prediction method, device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |