CN117274825B - Crop yield evaluation method and system based on remote sensing technology - Google Patents
Crop yield evaluation method and system based on remote sensing technology Download PDFInfo
- Publication number
- CN117274825B CN117274825B CN202311562389.4A CN202311562389A CN117274825B CN 117274825 B CN117274825 B CN 117274825B CN 202311562389 A CN202311562389 A CN 202311562389A CN 117274825 B CN117274825 B CN 117274825B
- Authority
- CN
- China
- Prior art keywords
- data
- crop
- remote sensing
- model
- environmental
- 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.)
- Active
Links
- 238000011156 evaluation Methods 0.000 title claims abstract description 62
- 238000005516 engineering process Methods 0.000 title claims abstract description 29
- 230000007613 environmental effect Effects 0.000 claims abstract description 98
- 238000000034 method Methods 0.000 claims abstract description 93
- 238000003062 neural network model Methods 0.000 claims abstract description 17
- 238000013210 evaluation model Methods 0.000 claims abstract description 14
- 230000009546 growth abnormality Effects 0.000 claims abstract description 11
- 238000012937 correction Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 9
- 238000010276 construction Methods 0.000 claims description 8
- 238000001914 filtration Methods 0.000 claims description 8
- 230000005856 abnormality Effects 0.000 claims description 6
- 230000008030 elimination Effects 0.000 claims description 6
- 238000003379 elimination reaction Methods 0.000 claims description 6
- 238000005286 illumination Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of crop yield evaluation, and particularly discloses a crop yield evaluation method and system based on a remote sensing technology, wherein the method comprises the steps of regularly receiving remote sensing data and constructing a crop growth model based on the remote sensing data; acquiring historical environmental data of a crop area, and determining environmental influence parameters according to the historical environmental data; the environmental impact parameters are used for representing the impact degree of environmental data on the crop growth process; judging the growth abnormality degree of crops based on the environmental influence parameters and preset average influence parameters; and inputting the crop growth model into a trained yield evaluation model to obtain an evaluation result, and correcting the evaluation result according to the growth anomaly. According to the invention, the result of the neural network model is corrected by introducing the growth abnormality parameter, so that the accuracy is improved, the update frequency of the neural network model in the remote sensing data identification process is reduced, the cost is lower, and the accuracy is improved.
Description
Technical Field
The invention relates to the technical field of crop yield evaluation, in particular to a crop yield evaluation method and system based on a remote sensing technology.
Background
Under the background of the prior art, agricultural planting is an intelligent planting process, and planting work can be completed by replacing manual work through some automatic equipment, but the growth analysis process of crops also needs to be completed by manual work, which is a constraint factor of agricultural global intelligence, and in order to solve the problem, a plurality of intelligent recognition technologies are presented.
In the existing intelligent recognition technology, the mainstream mode is a recognition scheme based on a neural network model, and recognized data can be unmanned aerial vehicle data or remote sensing data; the acquisition difficulty of the remote sensing data is low, but the requirement of the identification process on the neural network model is high, so that the neural network model needs to be updated at a higher frequency, the cost is high, the process is complicated, and the technical problem to be solved by the technical scheme of the invention is how to provide a new way for improving the accuracy and reduce the cost.
Disclosure of Invention
The invention aims to provide a crop yield evaluation method and system based on a remote sensing technology, which are used for solving the problems in the background technology.
In order to achieve the above purpose, the present invention provides the following technical solutions:
a crop yield assessment method based on remote sensing technology, the method comprising:
the method comprises the steps of regularly receiving remote sensing data and constructing a crop growth model based on the remote sensing data;
acquiring historical environmental data of a crop area, and determining environmental influence parameters according to the historical environmental data; the environmental impact parameters are used for representing the impact degree of environmental data on the crop growth process;
judging the growth abnormality degree of crops based on the environmental influence parameters and preset average influence parameters;
and inputting the crop growth model into a trained yield evaluation model to obtain an evaluation result, and correcting the evaluation result according to the growth anomaly.
As a further scheme of the invention: the step of regularly receiving remote sensing data and constructing a crop growth model based on the remote sensing data comprises the following steps of:
the method comprises the steps of receiving remote sensing data at regular time and removing noise from the remote sensing data;
performing contour recognition on the remote sensing data after noise rejection, and positioning a crop area;
positioning a shadow area in remote sensing data, calculating the height according to the solar azimuth angle of the shadow area at the current moment, and constructing a crop model according to the height and the crop area;
and (5) counting a crop model according to the time sequence, and constructing a crop growth model.
As a further scheme of the invention: the step of regularly receiving remote sensing data and carrying out noise rejection on the remote sensing data comprises the following steps:
the method comprises the steps of regularly receiving remote sensing data, and decomposing the remote sensing data to obtain images under different scales; the different scales at least comprise RGB image scales;
noise elimination is carried out on the image under each scale, and an enhancement component is obtained;
weighting and fusing the enhancement components of each layer to obtain a final image;
the final image generation process comprises the following steps:
the method comprises the steps of carrying out a first treatment on the surface of the In the method, in the process of the invention,in order to obtain a final image of the image,the weight of the i-th layer scale, n is the total number of scales,for the enhancement component at the i-th scale,，-，-，for an image at the i-th scale,representing gaussian filtering of the image at the i-th scale based on the first convolution kernel,representing Gaussian filtering of the image at the ith scale based on a second convolution kernel;referred to as the reflected component,referred to as an illumination component;in the form of a logarithmic sign,is an index symbol; the first convolution kernel and the second convolution kernel are parameters input by the management party.
As a further scheme of the invention: the step of obtaining historical environmental data of the crop area and determining environmental impact parameters according to the historical environmental data comprises the following steps:
acquiring historical environmental data of a crop area, and establishing a data table; wherein, the rows of the data table correspond to data types, and the columns of the data table correspond to time;
counting historical environmental data based on the number of columns of the data table; the statistical rules are:
the method comprises the steps of carrying out a first treatment on the surface of the In the method, in the process of the invention,for judging the sentence, when the bracket is true, the output is 1, when the bracket is false, the output is zero, and when the output is 1, the statistical process is executed;for the presence of symbols, M is the total number of columns,for the data value of column j +1,z is a preset threshold value for the data value of the j-th column;
and determining environmental influence parameters of each period according to the historical environmental data obtained through statistics.
As a further scheme of the invention: the step of determining the degree of growth abnormality of the crop based on the environmental impact parameter and a preset average impact parameter includes:
reading environment influence parameters and corresponding average influence parameters of each time period;
comparing the environmental influence parameter with the average influence parameter according to the time corresponding relation, and calculating the matching degree;
calculating the growth anomaly degree of crops according to the matching degree;
the method comprises the steps of carrying out a first treatment on the surface of the In the method, in the process of the invention,in order for the degree of matching to be achieved,for the correction factor, N is the number of parameter components,for the ith component of the environmental impact parameter,is the i-th component of the average influencing parameter.
As a further scheme of the invention: the step of inputting the crop growth model into a trained yield evaluation model to obtain an evaluation result, and correcting the evaluation result according to the growth anomaly degree comprises the following steps:
selecting a crop model from the crop growth models according to a preset selection frequency;
inputting the crop model into a trained yield evaluation model to obtain an evaluation result; the crop model is a neural network model;
and calculating an average evaluation result according to the evaluation result corresponding to each crop model, and correcting the evaluation result according to the growth anomaly degree.
The technical scheme of the invention also provides a crop yield evaluation system based on the remote sensing technology, which comprises the following steps:
the model construction module is used for receiving remote sensing data at fixed time and constructing a crop growth model based on the remote sensing data;
the environment data processing module is used for acquiring historical environment data of the crop area and determining environment influence parameters according to the historical environment data; the environmental impact parameters are used for representing the impact degree of environmental data on the crop growth process;
the abnormality degree judging module is used for judging the growth abnormality degree of crops based on the environmental influence parameters and preset average influence parameters;
and the evaluation result correction module is used for inputting the crop growth model into the trained output evaluation model to obtain an evaluation result, and correcting the evaluation result according to the growth anomaly degree.
As a further scheme of the invention: the model construction module comprises:
the noise removing unit is used for receiving the remote sensing data at regular time and removing noise from the remote sensing data;
the profile recognition unit is used for carrying out profile recognition on the remote sensing data after noise rejection and positioning the crop area;
the construction execution unit is used for positioning a shadow area in the remote sensing data, calculating the height according to the sun azimuth angle of the shadow area at the current moment, and constructing a crop model according to the height and the crop area;
and the model statistics unit is used for counting the crop model according to the time sequence and constructing a crop growth model.
As a further scheme of the invention: the environmental data processing module includes:
the data table establishing unit is used for acquiring historical environment data of the crop area and establishing a data table; wherein, the rows of the data table correspond to data types, and the columns of the data table correspond to time;
the data table processing unit is used for counting historical environment data based on the number of columns of the data table; the statistical rules are:
the method comprises the steps of carrying out a first treatment on the surface of the In the method, in the process of the invention,for judging the sentence, when the bracket is true, the output is 1, when the bracket is false, the output is zero, and when the output is 1, the statistical process is executed;for the presence of symbols, M is the total number of columns,for the data value of column j +1,z is a preset threshold value for the data value of the j-th column;
and the data application unit is used for determining the environmental influence parameters of each period according to the historical environmental data obtained through statistics.
As a further scheme of the invention: the abnormality degree determination module includes:
reading environment influence parameters and corresponding average influence parameters of each time period;
comparing the environmental influence parameter with the average influence parameter according to the time corresponding relation, and calculating the matching degree;
calculating the growth anomaly degree of crops according to the matching degree;
the method comprises the steps of carrying out a first treatment on the surface of the In the method, in the process of the invention,in order for the degree of matching to be achieved,for the correction factor, N is the number of parameter components,for the ith component of the environmental impact parameter,is the i-th component of the average influencing parameter.
Compared with the prior art, the invention has the beneficial effects that: according to the invention, the result of the neural network model is corrected by introducing the growth abnormality parameter, so that the accuracy is improved, the update frequency of the neural network model in the remote sensing data identification process is reduced, the cost is lower, and the accuracy is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following description will briefly introduce the drawings that are needed in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the present invention.
FIG. 1 is a flow chart diagram of a crop yield assessment method based on remote sensing technology.
Fig. 2 is a first sub-flowchart of a crop yield assessment method based on remote sensing technology.
FIG. 3 is a second sub-flowchart of a crop yield assessment method based on remote sensing technology.
Fig. 4 is a third sub-flowchart of a crop yield assessment method based on remote sensing technology.
Fig. 5 is a fourth sub-flowchart of a crop yield assessment method based on remote sensing technology.
Fig. 6 is a block diagram of the composition and structure of a crop yield evaluation system based on remote sensing technology.
Detailed Description
In order to make the technical problems, technical schemes and beneficial effects to be solved more clear, the invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Fig. 1 is a flow chart of a crop yield evaluation method based on a remote sensing technology, and in an embodiment of the invention, a crop yield evaluation method based on a remote sensing technology includes:
step S100: the method comprises the steps of regularly receiving remote sensing data and constructing a crop growth model based on the remote sensing data;
the remote sensing data are obtained by a satellite server, the remote sensing data are analyzed, a crop growth model can be constructed, the crop growth model is an actual crop change model, the crop states are determined by the remote sensing data, the crop states are arranged based on time sequence, and the states of crops in the growth process can be obtained, so that the change process is obtained and is called as a crop growth model.
Step S200: acquiring historical environmental data of a crop area, and determining environmental influence parameters according to the historical environmental data; the environmental impact parameters are used for representing the impact degree of environmental data on the crop growth process;
acquiring historical environmental data of a crop area, wherein the historical environmental data comprise data such as illumination parameters, air temperature, air humidity and the like at each moment; by combining the constructed growth model with the historical environmental data, environmental influence parameters, namely, how the growth state of crops is under different environmental data can be determined; this relationship is based on the correspondence of actual data, that is, the correspondence of actual environmental conditions (historical environmental data) and actual growth conditions (crop growth model).
Step S300: judging the growth abnormality degree of crops based on the environmental influence parameters and preset average influence parameters;
for one crop, the influence of the environment on the growth process of the crop can be counted, the environment states of a plurality of crops (the magnitude of sampled data is extremely large) and the growth conditions of the crops are sampled through a big data technology, and the average corresponding relation can be obtained.
It should be noted that one of the environmental impact parameters is some partial derivative, for example, the growth condition is a multiple function of environmental data (including illumination parameters, air temperature and air humidity), and each of the partial derivatives is used as an environmental impact parameter to characterize the impact condition of the environmental data on the crop growth process.
Step S400: inputting a crop growth model into a trained yield evaluation model to obtain an evaluation result, and correcting the evaluation result according to the growth anomaly degree;
finally, the management side uses the evaluation result as an index to count the states of the corresponding crops at different moments, so that the corresponding relation between the same evaluation result and the states at different moments can be constructed, when the number of the objects is large, each moment has multiple states, a neural network model (the mapping relation of image-numerical value) can be trained according to the states and the evaluation result, and the states are input, so that the corresponding numerical value can be obtained; wherein the status is determined by a crop model determined from the remote sensing data at each moment in time, the value representing the result of the evaluation, such as crop yield.
In an example of the technical scheme of the invention, the evaluation result is corrected according to the generated crop growth anomaly degree, so that the method is more practical. The method has the advantages that the updating frequency of the neural network model can be reduced, and the principle is that for a common trained neural network model, the updating times after the training is little or no longer carried out (the party for training the neural network model and the party for using the neural network model are generally different), along with the influence of a planting mode or other factors, the trained neural network model can be gradually not applicable any more, and after the comparison process based on environmental influence parameters is introduced, the result of the neural network model can be corrected once, so that the accuracy of the result is improved, and the result reaches an updating threshold later.
Fig. 2 is a first sub-flowchart of a crop yield evaluation method based on a remote sensing technology, the step of receiving remote sensing data at fixed time and constructing a crop growth model based on the remote sensing data includes:
step S101: the method comprises the steps of receiving remote sensing data at regular time and removing noise from the remote sensing data;
step S102: performing contour recognition on the remote sensing data after noise rejection, and positioning a crop area;
step S103: positioning a shadow area in remote sensing data, calculating the height according to the solar azimuth angle of the shadow area at the current moment, and constructing a crop model according to the height and the crop area;
step S104: and (5) counting a crop model according to the time sequence, and constructing a crop growth model.
In one example of the technical scheme of the invention, the crop growth model is a set of crop models at different moments, the crop models are generated based on remote sensing data, the remote sensing data is processed by noise elimination firstly, then contour recognition is carried out on the remote sensing data after the noise elimination to obtain the contour of the crop, a simulated three-dimensional model can be constructed by introducing the height on the contour of the crop, the model is called a crop model, and the crop growth model is obtained by counting the crop models in time sequence.
The altitude calculation process is a theoretical calculation process based on a shadow area and a solar azimuth angle, and a certain deviation may exist between the altitude calculation process and an actual situation, but in the technical scheme of the application, the deviation has little influence on final evaluation, so the deviation can be ignored.
As a preferred embodiment of the present invention, the step of receiving remote sensing data at regular time and performing noise rejection on the remote sensing data includes:
the method comprises the steps of regularly receiving remote sensing data, and decomposing the remote sensing data to obtain images under different scales; the different scales at least comprise RGB image scales;
noise elimination is carried out on the image under each scale, and an enhancement component is obtained;
weighting and fusing the enhancement components of each layer to obtain a final image;
the final image generation process comprises the following steps:
the method comprises the steps of carrying out a first treatment on the surface of the In the method, in the process of the invention,in order to obtain a final image of the image,the weight of the i-th layer scale, n is the total number of scales,for the enhancement component at the i-th scale,，-，-，for an image at the i-th scale,representing gaussian filtering of the image at the i-th scale based on the first convolution kernel,representing Gaussian filtering of the image at the ith scale based on a second convolution kernel;referred to as the reflected component,referred to as an illumination component;in the form of a logarithmic sign,is an index symbol; the first convolution kernel and the second convolution kernel are parameters input by the management party.
The principle of the process of noise rejection is that the remote sensing data is decomposed to obtain images under different channels, the segmentation mode is autonomously determined by a manager, and under general conditions, the existing main stream segmentation mode, such as an RGB three-channel mode, is adopted to independently process the images under each channel and then fuse the images to obtain a final image.
FIG. 3 is a second sub-flowchart of a crop yield evaluation method based on remote sensing technology, wherein the steps of obtaining historical environmental data of a crop area and determining environmental impact parameters according to the historical environmental data include:
step S201: acquiring historical environmental data of a crop area, and establishing a data table; wherein, the rows of the data table correspond to data types, and the columns of the data table correspond to time;
step S202: counting historical environmental data based on the number of columns of the data table;
step S203: and determining environmental influence parameters of each period according to the historical environmental data obtained through statistics.
In an example of the technical scheme of the invention, environmental data is recorded in real time by a preset sensor and a storage device to obtain historical environmental data, the historical environmental data is a matrix (the data table) with continuously prolonged columns, when the columns reach a preset threshold value, the data is intercepted and processed, and the process is repeated to obtain environmental influence parameters of each period.
Wherein, the threshold is set by staff, the threshold is generally the number of columns, and the number of columns corresponds to the time;
with respect to the above statistical rule, there is also a common way to calculate the change situation of the data, and if the change of the data is accumulated to a certain extent, a statistical process is performed, that is, the statistical rule is:
the method comprises the steps of carrying out a first treatment on the surface of the In the method, in the process of the invention,for judging the sentence, when the bracket is true, the output is 1, when the bracket is false, the output is zero, and when the output is 1, the statistical process is executed;for the presence of symbols, M is the total number of columns,for the data value of column j +1,and Z is a preset threshold value for the data value of the j-th column.
FIG. 4 is a third sub-flowchart of a crop yield evaluation method based on remote sensing technology, wherein the step of determining the growth anomaly degree of the crop based on the environmental impact parameter and the preset average impact parameter comprises:
step S301: reading environment influence parameters and corresponding average influence parameters of each time period;
step S302: comparing the environmental influence parameter with the average influence parameter according to the time corresponding relation, and calculating the matching degree;
step S303: and calculating the growth anomaly degree of the crops according to the matching degree.
The average influence parameters are known data obtained based on a big data technology, the environment influence parameters of each period are read and compared with the corresponding average influence parameters, a matching degree can be calculated, and the higher the matching degree is, the smaller the growth anomaly degree of crops is; the calculation formula of the matching degree is as follows:
the method comprises the steps of carrying out a first treatment on the surface of the In the method, in the process of the invention,in order for the degree of matching to be achieved,for the correction factor, N is the number of parameter components,for the ith component of the environmental impact parameter,is the i-th component of the average influencing parameter.
On the basis, the growth anomaly degree can be obtained by combining a subtraction function related to the matching degree, and finally, the obtained growth anomaly degree is in direct proportion to the sum of the component differences.
FIG. 5 is a fourth sub-flowchart of a crop yield evaluation method based on remote sensing technology, wherein the step of inputting a crop growth model into a trained yield evaluation model to obtain an evaluation result, and correcting the evaluation result according to the growth anomaly degree comprises the following steps:
step S401: selecting a crop model from the crop growth models according to a preset selection frequency;
step S402: inputting the crop model into a trained yield evaluation model to obtain an evaluation result; the crop model is a neural network model;
step S403: and calculating an average evaluation result according to the evaluation result corresponding to each crop model, and correcting the evaluation result according to the growth anomaly degree.
Step S401 to step S403 are specific application processes, and the crop model is processed by the trained yield evaluation model, so that an evaluation result can be obtained, the evaluation result is corrected according to the growth anomaly degree of the crop calculated in the above content, and based on the above content, the evaluation result can be corrected based on the matching degree, that is, the higher the matching degree is, the larger the correction coefficient is, and the smaller the correction amplitude of the evaluation result is.
Fig. 6 is a block diagram of a composition structure of a crop yield evaluation system based on a remote sensing technology, in which in an embodiment of the present invention, a crop yield evaluation system based on a remote sensing technology, the system 10 includes:
the model construction module 11 is used for regularly receiving remote sensing data and constructing a crop growth model based on the remote sensing data;
an environmental data processing module 12, configured to obtain historical environmental data of a crop area, and determine environmental impact parameters according to the historical environmental data; the environmental impact parameters are used for representing the impact degree of environmental data on the crop growth process;
an abnormality determination module 13 for determining a growth abnormality of the crop based on the environmental impact parameter and a preset average impact parameter;
the evaluation result correction module 14 is configured to input the crop growth model into a trained yield evaluation model, obtain an evaluation result, and correct the evaluation result according to the growth anomaly degree.
Further, the model building module 11 includes:
the noise removing unit is used for receiving the remote sensing data at regular time and removing noise from the remote sensing data;
the profile recognition unit is used for carrying out profile recognition on the remote sensing data after noise rejection and positioning the crop area;
the construction execution unit is used for positioning a shadow area in the remote sensing data, calculating the height according to the sun azimuth angle of the shadow area at the current moment, and constructing a crop model according to the height and the crop area;
and the model statistics unit is used for counting the crop model according to the time sequence and constructing a crop growth model.
Specifically, the environmental data processing module 12 includes:
the data table establishing unit is used for acquiring historical environment data of the crop area and establishing a data table; wherein, the rows of the data table correspond to data types, and the columns of the data table correspond to time;
the data table processing unit is used for counting historical environment data based on the number of columns of the data table; the statistical rules are:the method comprises the steps of carrying out a first treatment on the surface of the AndIn,for judging the sentence, when the bracket is true, the output is 1, when the bracket is false, the output is zero, and when the output is 1, the statistical process is executed;for the presence of symbols, M is the total number of columns,for the data value of column j +1,z is a preset threshold value for the data value of the j-th column;
and the data application unit is used for determining the environmental influence parameters of each period according to the historical environmental data obtained through statistics.
Further, the abnormality degree determination module 13 includes:
reading environment influence parameters and corresponding average influence parameters of each time period;
comparing the environmental influence parameter with the average influence parameter according to the time corresponding relation, and calculating the matching degree;
calculating the growth anomaly degree of crops according to the matching degree;
the method comprises the steps of carrying out a first treatment on the surface of the In the method, in the process of the invention,in order for the degree of matching to be achieved,for the correction factor, N is the number of parameter components,for the ith component of the environmental impact parameter,is the i-th component of the average influencing parameter.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the invention.
Claims (3)
1. A method for evaluating crop yield based on a remote sensing technology, the method comprising:
the method comprises the steps of regularly receiving remote sensing data and constructing a crop growth model based on the remote sensing data;
acquiring historical environmental data of a crop area, and determining environmental influence parameters according to the historical environmental data; the environmental impact parameters are used for representing the impact degree of environmental data on the crop growth process;
judging the growth abnormality degree of crops based on the environmental influence parameters and preset average influence parameters;
inputting a crop growth model into a trained yield evaluation model to obtain an evaluation result, and correcting the evaluation result according to the growth anomaly degree;
the step of regularly receiving remote sensing data and constructing a crop growth model based on the remote sensing data comprises the following steps of:
the method comprises the steps of receiving remote sensing data at regular time and removing noise from the remote sensing data;
performing contour recognition on the remote sensing data after noise rejection, and positioning a crop area;
positioning a shadow area in remote sensing data, calculating the height according to the solar azimuth angle of the shadow area at the current moment, and constructing a crop model according to the height and the crop area;
counting crop models according to time sequence, and constructing crop growth models;
the step of regularly receiving remote sensing data and carrying out noise rejection on the remote sensing data comprises the following steps:
the method comprises the steps of regularly receiving remote sensing data, and decomposing the remote sensing data to obtain images under different scales; the different scales at least comprise RGB image scales;
noise elimination is carried out on the image under each scale, and an enhancement component is obtained;
weighting and fusing the enhancement components of each layer to obtain a final image;
the final image generation process comprises the following steps:
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->For the final image +.>Weight of the i-th layer scale, n is the total number of scales, +.>For the enhancement component at the ith scale, < +.>，/>-/>，/>-，/>For an image at the ith scale, +.>Representing a Gaussian filtering of the image at the ith scale based on the first convolution kernel, +.>Representing Gaussian filtering of the image at the ith scale based on a second convolution kernel; />Called reflection component>Referred to as an illumination component; />Is logarithmic sign>Is an index symbol; the first convolution kernel and the second convolution kernel are parameters input by a manager;
the step of obtaining historical environmental data of the crop area and determining environmental impact parameters according to the historical environmental data comprises the following steps:
acquiring historical environmental data of a crop area, and establishing a data table; wherein, the rows of the data table correspond to data types, and the columns of the data table correspond to time;
counting historical environmental data based on the number of columns of the data table; the statistical rules are:
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->For judging the sentence, when the bracket is true, the output is 1, when the bracket is false, the output is zero, and when the output is 1, the statistical process is executed; />For the presence of symbols, M is the total number of columns, +.>For the data value of column j+1, < >>Z is a preset threshold value for the data value of the j-th column;
determining environmental influence parameters of each period according to the historical environmental data obtained through statistics;
the step of determining the degree of growth abnormality of the crop based on the environmental impact parameter and a preset average impact parameter includes:
reading environment influence parameters and corresponding average influence parameters of each time period;
comparing the environmental influence parameter with the average influence parameter according to the time corresponding relation, and calculating the matching degree;
calculating the growth anomaly degree of crops according to the matching degree;
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->For matching degree (I)>For correction coefficients, N is the number of parameter components, ">Is the i-th component of the environmental impact parameter, < +.>Is the i-th component of the average influencing parameter.
2. The method for evaluating crop yield based on remote sensing technology according to claim 1, wherein the step of inputting the crop growth model into the trained yield evaluation model to obtain an evaluation result, and correcting the evaluation result according to the growth anomaly degree comprises:
selecting a crop model from the crop growth models according to a preset selection frequency;
inputting the crop model into a trained yield evaluation model to obtain an evaluation result; the crop model is a neural network model;
and calculating an average evaluation result according to the evaluation result corresponding to each crop model, and correcting the evaluation result according to the growth anomaly degree.
3. A crop yield assessment system based on remote sensing technology, the system comprising:
the model construction module is used for receiving remote sensing data at fixed time and constructing a crop growth model based on the remote sensing data;
the environment data processing module is used for acquiring historical environment data of the crop area and determining environment influence parameters according to the historical environment data; the environmental impact parameters are used for representing the impact degree of environmental data on the crop growth process;
the abnormality degree judging module is used for judging the growth abnormality degree of crops based on the environmental influence parameters and preset average influence parameters;
the evaluation result correction module is used for inputting the crop growth model into a trained output evaluation model to obtain an evaluation result, and correcting the evaluation result according to the growth anomaly degree;
the model construction module comprises:
the noise removing unit is used for receiving the remote sensing data at regular time and removing noise from the remote sensing data;
the profile recognition unit is used for carrying out profile recognition on the remote sensing data after noise rejection and positioning the crop area;
the construction execution unit is used for positioning a shadow area in the remote sensing data, calculating the height according to the sun azimuth angle of the shadow area at the current moment, and constructing a crop model according to the height and the crop area;
the model statistics unit is used for counting crop models according to the time sequence and constructing crop growth models;
the timing receiving remote sensing data, and the content for noise rejection of the remote sensing data comprises:
the method comprises the steps of regularly receiving remote sensing data, and decomposing the remote sensing data to obtain images under different scales; the different scales at least comprise RGB image scales;
noise elimination is carried out on the image under each scale, and an enhancement component is obtained;
weighting and fusing the enhancement components of each layer to obtain a final image;
the final image generation process comprises the following steps:
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->For the final image +.>Weight of the i-th layer scale, n is the total number of scales, +.>For the enhancement component at the ith scale, < +.>，/>-/>，/>-，/>For an image at the ith scale, +.>Representing a Gaussian filtering of the image at the ith scale based on the first convolution kernel, +.>Representing Gaussian filtering of the image at the ith scale based on a second convolution kernel; />Called reflection component>Referred to as an illumination component; />Is logarithmic sign>Is an index symbol; the first convolution kernel and the second convolution kernel are parameters input by a manager;
the environmental data processing module includes:
the data table establishing unit is used for acquiring historical environment data of the crop area and establishing a data table; wherein, the rows of the data table correspond to data types, and the columns of the data table correspond to time;
the data table processing unit is used for counting historical environment data based on the number of columns of the data table; the statistical rules are:
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->For judging the sentence, when the bracket is true, the output is 1, when the bracket is false, the output is zero, and when the output is 1, the statistical process is executed; />For the presence of symbols, M is the total number of columns, +.>For the data value of column j+1, < >>Z is a preset threshold value for the data value of the j-th column;
the data application unit is used for determining environmental influence parameters of each period according to the historical environmental data obtained through statistics;
the abnormality degree determination module includes:
reading environment influence parameters and corresponding average influence parameters of each time period;
comparing the environmental influence parameter with the average influence parameter according to the time corresponding relation, and calculating the matching degree;
calculating the growth anomaly degree of crops according to the matching degree;
the method comprises the steps of carrying out a first treatment on the surface of the In (1) the->For matching degree (I)>For correction coefficients, N is the number of parameter components, ">Is the i-th component of the environmental impact parameter, < +.>Is the i-th component of the average influencing parameter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311562389.4A CN117274825B (en) | 2023-11-22 | 2023-11-22 | Crop yield evaluation method and system based on remote sensing technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311562389.4A CN117274825B (en) | 2023-11-22 | 2023-11-22 | Crop yield evaluation method and system based on remote sensing technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117274825A CN117274825A (en) | 2023-12-22 |
| CN117274825B true CN117274825B (en) | 2024-02-23 |
Family
ID=89201306
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311562389.4A Active CN117274825B (en) | 2023-11-22 | 2023-11-22 | Crop yield evaluation method and system based on remote sensing technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117274825B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106918816A (en) * | 2017-03-21 | 2017-07-04 | 中国科学院遥感与数字地球研究所 | crop growth monitoring system and its data processing method and module |
| CN113223040A (en) * | 2021-05-17 | 2021-08-06 | 中国农业大学 | Remote sensing-based banana yield estimation method and device, electronic equipment and storage medium |
| CN113591631A (en) * | 2021-07-16 | 2021-11-02 | 杭州领见数字农业科技有限公司 | Crop yield estimation method based on multi-source data |
| CN115879653A (en) * | 2023-03-02 | 2023-03-31 | 北京四象爱数科技有限公司 | Multi-factor crop yield estimation method, equipment and medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170161560A1 (en) * | 2014-11-24 | 2017-06-08 | Prospera Technologies, Ltd. | System and method for harvest yield prediction |
-
2023
- 2023-11-22 CN CN202311562389.4A patent/CN117274825B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106918816A (en) * | 2017-03-21 | 2017-07-04 | 中国科学院遥感与数字地球研究所 | crop growth monitoring system and its data processing method and module |
| CN113223040A (en) * | 2021-05-17 | 2021-08-06 | 中国农业大学 | Remote sensing-based banana yield estimation method and device, electronic equipment and storage medium |
| CN113591631A (en) * | 2021-07-16 | 2021-11-02 | 杭州领见数字农业科技有限公司 | Crop yield estimation method based on multi-source data |
| CN115879653A (en) * | 2023-03-02 | 2023-03-31 | 北京四象爱数科技有限公司 | Multi-factor crop yield estimation method, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117274825A (en) | 2023-12-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111126662B (en) | Irrigation decision making method, device, server and medium based on big data | |
| CN109946762B (en) | Method and system for short-time rainfall forecast based on probability distribution | |
| CN111612055B (en) | Weather situation typing method, air pollution condition prediction method and device | |
| CN110579186B (en) | Crop growth monitoring method based on inversion of leaf area index by inverse Gaussian process | |
| CN113901384A (en) | Ground PM2.5 concentration modeling method considering global spatial autocorrelation and local heterogeneity | |
| CN115357847B (en) | Solar scale satellite-ground precipitation fusion method based on error decomposition | |
| CN116205310A (en) | Soil water content influence factor sensitive interval judging method based on interpretable integrated learning model | |
| CN114298615A (en) | Crop planting risk prevention method and device, storage medium and equipment | |
| CN117114913A (en) | Intelligent agricultural data acquisition system based on big data | |
| CN114818888A (en) | Soil composition data fusion method and system based on multi-channel Kalman filtering | |
| CN114066070A (en) | Weather forecasting method based on artificial intelligence and numerical calculation | |
| CN117274825B (en) | Crop yield evaluation method and system based on remote sensing technology | |
| CN116485174B (en) | Method and device for evaluating risk of ozone pollution on crop yield reduction | |
| CN117456367A (en) | Crop planting area identification method, device, equipment and storage medium | |
| CN117035174A (en) | Method and system for estimating biomass on single-woodland of casuarina equisetifolia | |
| CN115186230A (en) | Method and device for judging influence of typhoon on air quality | |
| CN117574096B (en) | Agriculture and forestry pest prediction method based on big data | |
| US11640642B2 (en) | Method and system for dynamically predicting deoxynivalenol content of wheat at harvest | |
| CN111489021A (en) | Beet yield prediction method based on particle swarm optimization BP neural network | |
| CN115861821B (en) | Ecological environment monitoring and protecting method based on multi-objective optimization | |
| CN118296971A (en) | Crop weathered phenotype prediction method, system, device and storage medium | |
| CN118114810A (en) | Soybean yield prediction method based on neural network and Kalman filtering | |
| CN117670130A (en) | Arid region ecological restoration method and device based on maximum entropy | |
| CN117371617A (en) | Photovoltaic power station supporting capacity and adjusting cost collaborative prediction method and system thereof | |
| CN118095645A (en) | Unmanned agricultural precise planting intelligent control method |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |