CN113343783A - Intelligent crop identification and growth prediction method and system - Google Patents
Intelligent crop identification and growth prediction method and system Download PDFInfo
- Publication number
- CN113343783A CN113343783A CN202110541378.2A CN202110541378A CN113343783A CN 113343783 A CN113343783 A CN 113343783A CN 202110541378 A CN202110541378 A CN 202110541378A CN 113343783 A CN113343783 A CN 113343783A
- Authority
- CN
- China
- Prior art keywords
- vegetation index
- crop
- remote sensing
- sensing image
- time sequence
- 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
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000007781 pre-processing Methods 0.000 claims abstract description 25
- 238000013145 classification model Methods 0.000 claims abstract description 22
- 230000015654 memory Effects 0.000 claims description 26
- 238000000605 extraction Methods 0.000 claims description 17
- 238000012937 correction Methods 0.000 claims description 15
- 239000006185 dispersion Substances 0.000 claims description 10
- 239000002689 soil Substances 0.000 claims description 10
- 230000007787 long-term memory Effects 0.000 claims description 8
- 230000006403 short-term memory Effects 0.000 claims description 7
- 230000003595 spectral effect Effects 0.000 claims description 7
- 230000008569 process Effects 0.000 description 14
- 238000010586 diagram Methods 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 10
- 238000012549 training Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 9
- 238000004590 computer program Methods 0.000 description 7
- 238000011160 research Methods 0.000 description 7
- 238000012545 processing Methods 0.000 description 6
- 238000002310 reflectometry Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 4
- 238000003860 storage Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000010521 absorption reaction Methods 0.000 description 2
- 238000012271 agricultural production Methods 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000003750 conditioning effect Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 230000001105 regulatory effect Effects 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 240000002624 Mespilus germanica Species 0.000 description 1
- 235000017784 Mespilus germanica Nutrition 0.000 description 1
- 235000000560 Mimusops elengi Nutrition 0.000 description 1
- CBENFWSGALASAD-UHFFFAOYSA-N Ozone Chemical compound [O-][O+]=O CBENFWSGALASAD-UHFFFAOYSA-N 0.000 description 1
- 235000007837 Vangueria infausta Nutrition 0.000 description 1
- 241000219094 Vitaceae Species 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 239000005427 atmospheric aerosol Substances 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 235000021021 grapes Nutrition 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000001747 pupil Anatomy 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
- G06F18/24137—Distances to cluster centroïds
- G06F18/2414—Smoothing the distance, e.g. radial basis function networks [RBFN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Molecular Biology (AREA)
- Agronomy & Crop Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Animal Husbandry (AREA)
- Marine Sciences & Fisheries (AREA)
- Mining & Mineral Resources (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method and a system for intelligently identifying crops and predicting growth vigor, wherein the method comprises the following steps: acquiring an original remote sensing image of crops and preprocessing the acquired original remote sensing image to obtain a preprocessed remote sensing image; calculating the vegetation index of each pixel point in the preprocessed remote sensing image, and acquiring vegetation index time sequence data; classifying the current crop type through a preset crop classification model according to the vegetation index time sequence data; and predicting the growth vigor of the current crops through a preset crop growth vigor prediction model according to the vegetation index time sequence data based on the classification result of the current crops. The method can realize the careful classification of the interior of the garden in the remote sensing image, distinguish crops more carefully and predict the growth vigor of the crops.
Description
Technical Field
The invention relates to the technical field of crossing of remote sensing image intelligent interpretation and intelligent agriculture, in particular to a crop intelligent identification and growth prediction method and system.
Background
The acquisition of the crop classification information has very important significance in various fields such as agricultural resource investigation, current land utilization state analysis, crop yield estimation, disaster assessment and the like. With the emission of the high-resolution remote sensing image satellite, the technology breaks through the optical remote sensing technology combining high resolution, multispectral and wide coverage and the like. Monitoring crop planting by using remote sensing satellite data has become an important link in the agricultural production process. Many different methods have been developed to improve the accuracy of crop identification based on remote sensing, and provide a lot of reference and auxiliary information for relevant departments to know the crop planting situation.
The intelligent interpretation of remote sensing images is a major branch of remote sensing image research, which is the process of classifying images or pixels thereon into different categories according to certain properties. The intelligent interpretation research of the remote sensing image not only has scientific theoretical significance, but also has practical significance. In the remote sensing science and technology, the intelligent interpretation of remote sensing images is one of the most basic research problems, and is the basis of research and application of other remote sensing technologies, and direct or indirect influence can be generated on the remote sensing images. In practical application, the images identified by classification can meet the requirements of observing the state of a target object and predicting the development and change trend of the target object, which cannot be seen by a user.
In agricultural resource investigation, the satellite remote sensing technology is utilized to timely and accurately acquire crop planting area information, and the method is a reliable basis for guiding agricultural production, adjusting agricultural structure and estimating and predicting the yield of regional crops by an agricultural department. Therefore, the method is very important for the intelligent interpretation research of the remote sensing images. However, the prior art cannot classify crops finely and predict the future growth of crops in the time dimension at present.
Disclosure of Invention
The invention provides a method and a system for intelligently identifying crops and predicting the growth vigor of the crops, and aims to solve the problem that the prior art cannot finely classify crops and predict the future growth vigor of the crops in a time dimension.
In order to solve the technical problems, the invention provides the following technical scheme:
in one aspect, the invention provides a crop intelligent identification and growth prediction method, which comprises the following steps:
acquiring an original remote sensing image of a crop to be predicted for growing, and preprocessing the original remote sensing image to obtain a preprocessed remote sensing image; wherein the pre-processing includes radiometric calibration and atmospheric correction;
calculating the vegetation index of each pixel point in the preprocessed remote sensing image, and acquiring vegetation index time sequence data;
classifying the current crop type through a preset crop classification model according to the vegetation index time sequence data; the crop classification model is a convolution long-time and short-time memory network, the input of the crop classification model is vegetation index time sequence data, and the output is a corresponding crop classification result;
predicting the growth vigor of the current crops through a preset crop growth vigor prediction model according to the vegetation index time sequence data based on the classification result of the current crops; the crop growth prediction model is a long-term and short-term memory network, the input of the crop growth prediction model is vegetation index time sequence data, and the output of the crop growth prediction model is vegetation index time sequence data in a future time period corresponding to the currently input vegetation index time sequence data.
Further, in the above method, the preprocessing the original remote sensing image includes:
establishing a quantitative relation between a digital value and a radiance value in a corresponding field of view through radiometric calibration so as to eliminate errors generated by a remote sensing image acquisition sensor;
and restoring the calibration value into the real information of the earth surface through atmospheric correction, and recovering the spectral information of the earth objects with high fidelity.
Optionally, in the above method, the vegetation index is a normalized difference vegetation index, a weighted difference vegetation index, or a soil conditioning vegetation index.
Preferably, in the above method, the vegetation index is a weight-difference vegetation index.
Further, in the above method, after calculating the vegetation index of each pixel point in the preprocessed remote sensing image, the method further includes:
and carrying out numerical value statistics on the vegetation indexes of all pixel points in the selected area in the preprocessed remote sensing image, judging the data dispersion degree according to the standard deviation, and judging the reliability of corresponding data according to the data dispersion degree.
On the other hand, the invention also provides a crop intelligent identification and growth prediction system, which comprises:
the remote sensing image preprocessing module is used for acquiring an original remote sensing image of a crop to be predicted for growing and preprocessing the original remote sensing image to obtain a preprocessed remote sensing image; wherein the pre-processing includes radiometric calibration and atmospheric correction;
the vegetation index feature extraction module is used for calculating the vegetation index of each pixel point in the remote sensing image after the remote sensing image is preprocessed by the remote sensing image preprocessing module and acquiring vegetation index time sequence data;
the crop intelligent classification module is used for classifying the current crop types through a preset crop classification model according to the vegetation index time sequence data extracted by the vegetation index feature extraction module; the crop classification model is a convolution long-time and short-time memory network, the input of the crop classification model is vegetation index time sequence data, and the output is a corresponding crop classification result;
the crop growth prediction module is used for predicting the growth of the current crops through a preset crop growth prediction model according to the vegetation index time sequence data extracted by the vegetation index feature extraction module based on the classification result of the crop intelligent classification module on the current crops; the crop growth prediction model is a long-term and short-term memory network, the input of the crop growth prediction model is vegetation index time sequence data, and the output of the crop growth prediction model is vegetation index time sequence data in a future time period corresponding to the currently input vegetation index time sequence data.
Further, in the above system, the remote sensing image preprocessing module is specifically configured to:
establishing a quantitative relation between a digital value and a radiance value in a corresponding field of view through radiometric calibration so as to eliminate errors generated by a remote sensing image acquisition sensor;
and restoring the calibration value into the real information of the earth surface through atmospheric correction, and recovering the spectral information of the earth objects with high fidelity.
Optionally, in the system, the vegetation index calculated by the vegetation index feature extraction module is a normalized differential vegetation index, a weighted difference vegetation index or a soil regulation vegetation index.
Preferably, in the above system, the vegetation index calculated by the vegetation index feature extraction module is a weighted difference vegetation index.
Further, in the above system, the vegetation index feature extraction module is further configured to:
and carrying out numerical value statistics on the vegetation indexes of all pixel points in the selected area in the preprocessed remote sensing image, judging the data dispersion degree according to the standard deviation, and judging the reliability of the data according to the data dispersion degree.
In yet another aspect, the present invention also provides an electronic device comprising a processor and a memory; wherein the memory has stored therein at least one instruction that is loaded and executed by the processor to implement the above-described method.
In yet another aspect, the present invention also provides a computer-readable storage medium having at least one instruction stored therein, the instruction being loaded and executed by a processor to implement the above method.
The technical scheme provided by the invention has the beneficial effects that at least:
the method can be widely applied to multiple relevant aspects such as crop fine classification identification, growth prediction, surface feature extraction and the like, provides necessary basis for research and application of relevant fields such as land planning, crop planting planning and the like by constructing an accurate crop fine classification model, and can assist development of crop detection of the relevant fields. The invention combines advanced intelligent scientific technology to identify the fine categories of crops like disciplinary cross new thinking, is beneficial to exploring and establishing complete theory and algorithm of fine category identification of crops under an agricultural-industry combination mode, is beneficial to improving standardization, flow engineering application and innovation construction level of accurate identification, classification and extraction of crops, has important theoretical significance for strengthening new technology, new state and new mode of new generation artificial intelligence in the agricultural field, and simultaneously has promotion effect on intelligent agricultural research and construction.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a flowchart illustrating an implementation of a method for intelligently identifying crops and predicting growth vigor according to an embodiment of the present invention;
FIG. 2 is a block diagram of an intelligent crop identification and growth prediction system according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a convolution long-term and short-term memory network;
fig. 4 is a schematic diagram illustrating comparison between the predicted result and the actual value according to the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
First embodiment
Aiming at the problem that the prior art cannot perform fine classification on crops and predict the future growth of the crops in the time dimension, the embodiment provides the intelligent crop identification and growth prediction method based on the multi-temporal remote sensing characteristic, and the method can be realized by electronic equipment which can be a terminal or a server. The execution flow of the method is shown in fig. 1, and mainly comprises the following steps:
s1, obtaining an original remote sensing image of the crop to be predicted for growing, and preprocessing the original remote sensing image to obtain a preprocessed remote sensing image; wherein the preprocessing comprises radiometric calibration and atmospheric correction;
s2, calculating the vegetation index of each pixel point in the preprocessed remote sensing image, and acquiring vegetation index time sequence data;
s3, classifying the current crop type through a preset crop classification model according to the vegetation index time sequence data; the crop classification model is a convolution long-time and short-time memory network, the input of the crop classification model is vegetation index time sequence data, and the output is a corresponding crop classification result;
s4, predicting the growth vigor of the current crops through a preset crop growth vigor prediction model according to vegetation index time sequence data based on the classification result of the current crops; the crop growth prediction model is a long-term and short-term memory network, the input of the crop growth prediction model is vegetation index time sequence data, and the output is vegetation index time sequence data in a future time period corresponding to the currently input vegetation index time sequence data.
Specifically, in this embodiment, the process of preprocessing the original remote sensing image in S1 is as follows:
s11, performing radiometric calibration on the original image; establishing a quantitative relation between a digital value and a radiance value in a corresponding field of view through radiometric calibration so as to eliminate errors generated by a remote sensing image acquisition sensor;
it should be noted that the purpose of radiometric calibration is to eliminate interference of the sensor itself, the atmosphere, the solar altitude, the terrain, and the like, and obtain data of the real reflectivity. Specifically, the radiometric calibration process comprises the steps of obtaining air, ground and atmospheric environment data, calculating the optical thickness of atmospheric aerosol, calculating the water and ozone content in atmosphere, analyzing and processing data such as ground substance spectrums of a calibration site and a training area, obtaining an aggregate parameter and time when the data of the calibration site are obtained, bringing the obtained and calculated parameters into an atmospheric radiation transmission model, obtaining the radiance when a sensor enters a pupil at a remote place, calculating a calibration coefficient, converting the recorded original DN value into the surface reflectivity, and converting the brightness gray value of an image into the absolute radiance so as to eliminate errors generated by the sensor.
S12, performing atmospheric correction on the image data after radiometric calibration; and restoring the calibration value into the real information of the earth surface through atmospheric correction, and recovering the spectral information of the earth objects with high fidelity.
It should be noted that the atmospheric correction means that the total radiance of the ground target finally measured by the sensor is not reflected by the real reflectivity of the ground, and includes the radiant quantity error caused by the atmospheric absorption, especially scattering effect. Atmospheric correction is a process of eliminating radiation errors caused by the atmospheric influence and inverting the real surface reflectivity of the ground object. The process converts the radiance or surface reflectivity into the actual surface reflectivity, thereby eliminating the errors caused by atmospheric scattering, absorption and reflection.
S13, setting an ROI (region of interest);
according to the land pattern spot data given by the Yinchuan remote measuring institute, crops needing to be identified are identified, such as: and marking areas such as medlar, grapes and the like.
The vegetation index calculation in S2 is to combine and calculate the satellite visible light and the near-infrared band according to the spectral characteristics of the vegetation to obtain a vegetation index value, and then generate a vegetation index time sequence according to the vegetation index sequence. The vegetation index value can be a simple, effective, and empirical measure of the condition of the surface vegetation.
The vegetation index types are: a Normalized Differential Vegetation Index (NDVI) that classifies vegetation from a range of water and soil images, calculated as: NDVI ═ (NIR-R)/(NIR + R); a Weighted Difference Vegetation Index (WDVI) which takes into account the image of the soil background value and introduces soil conditioning parameters, the calculation formula of which is: WDVI-NIR-g R; the soil regulating vegetation index (SAVI) is a vegetation index which can simply describe a soil-vegetation system, and is a new vegetation index obtained by introducing a soil regulating coefficient and eliminating the interference of soil background on a spectral signal on the basis of NDVI, and the calculation formula is as follows: SAVI ═ (1+ L) × (NIR-R)/(NIR + R + L).
The present embodiment uses the WDVI vegetation index. Calculating the WDVI value of each pixel point, then assigning values to each pixel point again to obtain a WDVI image, carrying out numerical statistics on the interior of the selected ROI to obtain the average value and the standard deviation in the region, judging the dispersion degree of data according to the standard deviation, and judging the reliability of the data according to the dispersion degree setting threshold. And superposing the images of the months in the same year, extracting WDVI time sequence data of the specific crops of the whole year, and making a time sequence curve according to the data.
The implementation process of classifying and identifying the crops in the step S3 is as follows:
constructing a convolution long-time and short-time memory network, inputting a training set into the constructed convolution long-time and short-time memory network, learning the vegetation index value of each time node and the change situation of the value in a certain time span, optimizing the parameters of the convolution long-time and short-time memory network, measuring the difference between an output result and a real result, learning again, iterating repeatedly until a judgment result meeting the loss function requirement is output, and obtaining a crop classification model; predicting the vegetation index value of the image of the next year by utilizing the crop classification model through the image of the previous year; and then the predicted WDVI vegetation index curve conforms to which specific crop is judged, so that the classification and identification of the corresponding crop are realized, and the fine classification of the crops in the garden is realized.
Specifically, the network structure of the convolutional long-short term memory network used in this embodiment is as shown in fig. 3, the network has a chain structure, the structure of each long-short term memory Unit (LSTM Unit) in the chain structure is repeated, the states of the units are sequentially transmitted in the chain structure, only a small amount of current interaction is performed with the network, and information easily flows between the LSTM units, so that the network structure has a better timing memory capability.
In this embodiment, the network structure of the convolution long-time memory network includes three gates:
forget gate, the first step of each cell of the LSTM is to decide what information to discard, the gate will read ht-1And xtOutputting a value between 0 and 1, 1 indicating complete retention, 0 indicating complete forgetting, and the value and the previous cell state ct-1An element product operation is performed.
An input gate that determines how much new information is allowed to be added to the cell state, two steps are required to implement the gate: firstly, a sigmoid layer of an input gate determines which information needs to be updated; one tanh layer generates one vector, namely: alternative cellular states for renewal
② renewal of cell status from ct-1Is updated to ct. C of old statet-1And ftMultiplying to determine the information which needs to be discarded; alternative states
And itThe new candidate is determined by multiplication.
An output gate that ultimately determines what value is output, two steps are required to implement the gate: firstly, operating a sigmoid layer to determine which part of the cell state is output; processing the cell state through a tanh layer to obtain a value between-1 and 1, multiplying the value by the output of the sigmoid gate, and finally outputting a part for determining the output.
The convolutional layer operation in this embodiment can make the network obtain a time sequence relationship, and simultaneously can extract spatial features, and by extracting the two, time features and spatial features can be obtained at the same time, that is: and (5) space-time relation characteristics.
In this embodiment, the intelligent decision process of the crop identification technology based on the convolution duration memory network is specifically as follows; wherein the network parameters include: the number of convolutional layers, the size of a convolutional core, the number of training times, the size of a batch, the length of a memory network number and the number of nodes in a hidden layer.
In the embodiment, parameters of the long-term memory network are optimized according to the steps, the difference between the output result and the real result is measured, the learning is performed again, iteration is repeated until the judgment result meeting the loss function requirement is output, so that the accurate intelligent decision for identifying the specific crops under the condition of small samples is realized, and the problem of missing of an intelligent and accurate decision method is solved.
The results obtained by multiple parameter adjustment experiments show that the relationship between the selection of the model parameters and the classification accuracy of the method for crop intelligent identification and growth prediction based on multi-temporal remote sensing characteristics is shown in table 1.
TABLE 1 relationship between parameter selection and classification accuracy for specific crop identification technology models
In table 1, epoch is the number of times of training using all samples in the training set; the batch size is the batch size, and that is, each training is trained by taking samples of the batch size in the training set.
The prediction of the growth vigor of the crops in the step S4 is specifically as follows: by adjusting the training set and the algorithm, a long-time memory network is used, the growth vigor of the specific crops in unknown months is predicted by using data of different months in the same year, and the predicted result and the real value are as shown in fig. 4. In this embodiment, the implementation process of predicting the growth vigor of crops is as follows: and training the model by using the previous part of data in the same year, and predicting the growth of crops in the latter part of data in the same year by using the trained model. The specific implementation process is as follows:
constructing a curve change characteristic generator, carrying out discrimination comparison on the preliminarily generated vegetation index value and the real vegetation index value, calculating a loss function of the vegetation index value, and feeding back the loss function to the generator; optimizing the weight of the generator according to the feedback result, further fitting the preliminarily generated vegetation index value and the change characteristic, judging again, and repeating iteration until a vegetation index curve meeting the requirements of the judger is generated; and predicting the future growth of the crops through the finally generated vegetation index curve. The parameters in the network include: the number of convolutional layers, the size of a convolutional core, the number of training times, the size of a batch, the length of a memory network number and the number of nodes in a hidden layer.
In summary, the crop intelligent identification and growth prediction method based on the multi-temporal remote sensing features is an intelligent technology for judging specific crops based on the temporal-spatial features, on the basis, after preprocessing such as data radiometric calibration and atmospheric correction, time and space information of the specific crops are extracted, further, based on the temporal-spatial information, the spatiotemporal relation features of the specific crops are obtained by utilizing a convolutional layer, and further, based on a long-short time memory network, a discrimination classification task for the specific crops is realized, the effectiveness of the discrimination classification task is verified, the precision description level of the specific crop fine classification is effectively improved, the precise crop fine classification is realized, and meanwhile, a novel, high-quality and high-efficiency intelligent auxiliary tool is provided for decision making of the specific crops.
Second embodiment
The embodiment provides a crop intelligent identification and growth prediction system based on multi-temporal remote sensing characteristics, and the structure of the system is shown in fig. 2, and the system comprises the following modules:
the remote sensing image preprocessing module is used for acquiring an original remote sensing image of a crop to be predicted for growing and preprocessing the original remote sensing image to obtain a preprocessed remote sensing image; wherein the pre-processing includes radiometric calibration and atmospheric correction;
the vegetation index feature extraction module is used for calculating the vegetation index of each pixel point in the remote sensing image after the remote sensing image is preprocessed by the remote sensing image preprocessing module and acquiring vegetation index time sequence data;
the crop intelligent classification module is used for classifying the current crop types through a preset crop classification model according to the vegetation index time sequence data extracted by the vegetation index feature extraction module; the crop classification model is a convolution long-time and short-time memory network, the input of the crop classification model is vegetation index time sequence data, and the output is a corresponding crop classification result;
the crop growth prediction module is used for predicting the growth of the current crops through a preset crop growth prediction model according to the vegetation index time sequence data extracted by the vegetation index feature extraction module based on the classification result of the crop intelligent classification module on the current crops; the crop growth prediction model is a long-term and short-term memory network, the input of the crop growth prediction model is vegetation index time sequence data, and the output of the crop growth prediction model is vegetation index time sequence data in a future time period corresponding to the currently input vegetation index time sequence data.
The crop intelligent identification and growth prediction system of the present embodiment corresponds to the crop intelligent identification and growth prediction method of the first embodiment; the functions realized by the functional modules in the intelligent crop identification and growth prediction system of the embodiment correspond to the flow steps in the intelligent crop identification and growth prediction method of the first embodiment one by one; therefore, it is not described herein.
Third embodiment
The present embodiment provides an electronic device, which includes a processor and a memory; wherein the memory has stored therein at least one instruction that is loaded and executed by the processor to implement the method of the first embodiment.
The electronic device may have a relatively large difference due to different configurations or performances, and may include one or more processors (CPUs) and one or more memories, where at least one instruction is stored in the memory, and the instruction is loaded by the processor and executes the method.
Fourth embodiment
The present embodiment provides a computer-readable storage medium, in which at least one instruction is stored, and the instruction is loaded and executed by a processor to implement the method of the first embodiment. The computer readable storage medium may be, among others, ROM, random access memory, CD-ROM, magnetic tape, floppy disk, optical data storage device, and the like. The instructions stored therein may be loaded by a processor in the terminal and perform the above-described method.
Furthermore, it should be noted that the present invention may be provided as a method, apparatus or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media having computer-usable program code embodied in the medium.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It should also be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.
Finally, it should be noted that while the above describes a preferred embodiment of the invention, it will be appreciated by those skilled in the art that, once the basic inventive concepts have been learned, numerous changes and modifications may be made without departing from the principles of the invention, which shall be deemed to be within the scope of the invention. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the embodiments of the invention.
Claims (10)
1. A crop intelligent identification and growth prediction method is characterized by comprising the following steps:
acquiring an original remote sensing image of a crop to be predicted for growing, and preprocessing the original remote sensing image to obtain a preprocessed remote sensing image; wherein the pre-processing includes radiometric calibration and atmospheric correction;
calculating the vegetation index of each pixel point in the preprocessed remote sensing image, and acquiring vegetation index time sequence data;
classifying the current crop type through a preset crop classification model according to the vegetation index time sequence data; the crop classification model is a convolution long-time and short-time memory network, the input of the crop classification model is vegetation index time sequence data, and the output is a corresponding crop classification result;
predicting the growth vigor of the current crops through a preset crop growth vigor prediction model according to the vegetation index time sequence data based on the classification result of the current crops; the crop growth prediction model is a long-term and short-term memory network, the input of the crop growth prediction model is vegetation index time sequence data, and the output of the crop growth prediction model is vegetation index time sequence data in a future time period corresponding to the currently input vegetation index time sequence data.
2. The intelligent crop identification and growth prediction method of claim 1 wherein the preprocessing of the raw remote sensing image comprises:
establishing a quantitative relation between a digital value and a radiance value in a corresponding field of view through radiometric calibration so as to eliminate errors generated by a remote sensing image acquisition sensor;
and restoring the calibration value into the real information of the earth surface through atmospheric correction, and recovering the spectral information of the earth objects with high fidelity.
3. The method of claim 1, wherein the vegetation index is a normalized difference vegetation index, a weighted difference vegetation index, or a soil adjusted vegetation index.
4. The method for intelligently identifying and predicting the growth of a crop as claimed in claim 1 wherein said vegetation index is a weight-difference vegetation index.
5. The method for intelligently identifying and predicting the growth of crops as claimed in claim 1, wherein after calculating the vegetation index of each pixel point in the preprocessed remote sensing image, the method further comprises:
and carrying out numerical value statistics on the vegetation indexes of all pixel points in the selected area in the preprocessed remote sensing image, judging the data dispersion degree according to the standard deviation, and judging the reliability of corresponding data according to the data dispersion degree.
6. The utility model provides a crops intelligent recognition and growth prediction system which characterized in that includes:
the remote sensing image preprocessing module is used for acquiring an original remote sensing image of a crop to be predicted for growing and preprocessing the original remote sensing image to obtain a preprocessed remote sensing image; wherein the pre-processing includes radiometric calibration and atmospheric correction;
the vegetation index feature extraction module is used for calculating the vegetation index of each pixel point in the remote sensing image after the remote sensing image is preprocessed by the remote sensing image preprocessing module and acquiring vegetation index time sequence data;
the crop intelligent classification module is used for classifying the current crop types through a preset crop classification model according to the vegetation index time sequence data extracted by the vegetation index feature extraction module; the crop classification model is a convolution long-time and short-time memory network, the input of the crop classification model is vegetation index time sequence data, and the output is a corresponding crop classification result;
the crop growth prediction module is used for predicting the growth of the current crops through a preset crop growth prediction model according to the vegetation index time sequence data extracted by the vegetation index feature extraction module based on the classification result of the crop intelligent classification module on the current crops; the crop growth prediction model is a long-term and short-term memory network, the input of the crop growth prediction model is vegetation index time sequence data, and the output of the crop growth prediction model is vegetation index time sequence data in a future time period corresponding to the currently input vegetation index time sequence data.
7. The intelligent crop identification and growth prediction system of claim 6, wherein the remote sensing image preprocessing module is specifically configured to:
establishing a quantitative relation between a digital value and a radiance value in a corresponding field of view through radiometric calibration so as to eliminate errors generated by a remote sensing image acquisition sensor;
and restoring the calibration value into the real information of the earth surface through atmospheric correction, and recovering the spectral information of the earth objects with high fidelity.
8. The crop intelligent identification and growth prediction system of claim 6 wherein the vegetation index calculated by the vegetation index feature extraction module is a normalized difference vegetation index, a weighted difference vegetation index or a soil adjusted vegetation index.
9. The crop intelligent identification and growth prediction system of claim 6 wherein the vegetation index calculated by the vegetation index feature extraction module is a weighted difference vegetation index.
10. The crop intelligent recognition and growth prediction system of claim 6 wherein the vegetation index feature extraction module is further configured to:
and carrying out numerical value statistics on the vegetation indexes of all pixel points in the selected area in the preprocessed remote sensing image, judging the data dispersion degree according to the standard deviation, and judging the reliability of the data according to the data dispersion degree.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110541378.2A CN113343783A (en) | 2021-05-18 | 2021-05-18 | Intelligent crop identification and growth prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110541378.2A CN113343783A (en) | 2021-05-18 | 2021-05-18 | Intelligent crop identification and growth prediction method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113343783A true CN113343783A (en) | 2021-09-03 |
Family
ID=77469177
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110541378.2A Pending CN113343783A (en) | 2021-05-18 | 2021-05-18 | Intelligent crop identification and growth prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113343783A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114239716A (en) * | 2021-12-15 | 2022-03-25 | 平安科技（深圳）有限公司 | Double-shaft interaction method, device, equipment and medium based on crop growth vigor |
| CN116069436A (en) * | 2023-03-17 | 2023-05-05 | 阿里巴巴达摩院(杭州)科技有限公司 | Remote sensing image processing method, electronic equipment and storage medium |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103760872A (en) * | 2014-01-27 | 2014-04-30 | 林兴志 | Alarming and intervening device for refined agricultural crop growth |
| CN109187356A (en) * | 2018-08-30 | 2019-01-11 | 南京都宁大数据科技有限公司 | Soybean growing way prediction technique based on satellite remote sensing date |
| CN109815910A (en) * | 2019-01-14 | 2019-05-28 | 广西师范学院 | Crop type recognition methods under MODEL OVER COMPLEX TOPOGRAPHY based on multi-source satellite data temporal aspect |
| CN110287944A (en) * | 2019-07-04 | 2019-09-27 | 吉林大学 | The crop pests monitoring method of multi-spectrum remote sensing image based on deep learning |
| CN110287869A (en) * | 2019-06-25 | 2019-09-27 | 吉林大学 | High-resolution remote sensing image Crop classification method based on deep learning |
| CN110545531A (en) * | 2019-09-20 | 2019-12-06 | 河南工业大学 | Crop growth monitoring method and system based on big data and cloud computing |
| CN111738329A (en) * | 2020-06-19 | 2020-10-02 | 中南大学 | Land use classification method for time series remote sensing images |
| CN111898503A (en) * | 2020-07-20 | 2020-11-06 | 中国农业科学院农业资源与农业区划研究所 | Crop identification method and system based on cloud coverage remote sensing image and deep learning |
| CN112712071A (en) * | 2021-03-26 | 2021-04-27 | 北京简巨科技有限公司 | Crop growth detection method and related equipment thereof |
| CN112749627A (en) * | 2020-12-09 | 2021-05-04 | 北京星衡科技有限公司 | Method and device for dynamically monitoring tobacco based on multi-source remote sensing image |
-
2021
- 2021-05-18 CN CN202110541378.2A patent/CN113343783A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103760872A (en) * | 2014-01-27 | 2014-04-30 | 林兴志 | Alarming and intervening device for refined agricultural crop growth |
| CN109187356A (en) * | 2018-08-30 | 2019-01-11 | 南京都宁大数据科技有限公司 | Soybean growing way prediction technique based on satellite remote sensing date |
| CN109815910A (en) * | 2019-01-14 | 2019-05-28 | 广西师范学院 | Crop type recognition methods under MODEL OVER COMPLEX TOPOGRAPHY based on multi-source satellite data temporal aspect |
| CN110287869A (en) * | 2019-06-25 | 2019-09-27 | 吉林大学 | High-resolution remote sensing image Crop classification method based on deep learning |
| CN110287944A (en) * | 2019-07-04 | 2019-09-27 | 吉林大学 | The crop pests monitoring method of multi-spectrum remote sensing image based on deep learning |
| CN110545531A (en) * | 2019-09-20 | 2019-12-06 | 河南工业大学 | Crop growth monitoring method and system based on big data and cloud computing |
| CN111738329A (en) * | 2020-06-19 | 2020-10-02 | 中南大学 | Land use classification method for time series remote sensing images |
| CN111898503A (en) * | 2020-07-20 | 2020-11-06 | 中国农业科学院农业资源与农业区划研究所 | Crop identification method and system based on cloud coverage remote sensing image and deep learning |
| CN112749627A (en) * | 2020-12-09 | 2021-05-04 | 北京星衡科技有限公司 | Method and device for dynamically monitoring tobacco based on multi-source remote sensing image |
| CN112712071A (en) * | 2021-03-26 | 2021-04-27 | 北京简巨科技有限公司 | Crop growth detection method and related equipment thereof |
Non-Patent Citations (2)
| Title |
|---|
| （菲）罗韦尔·阿蒂恩扎: "《Keras高级深度学习》", 31 March 2020, 北京：机械工业出版社, pages: 77 - 82 * |
| 龙泽昊: ""基于长短期记忆网络的冬小麦连续时序叶面积指数预测"", 《光谱学与光谱分析》, vol. 40, no. 3, pages 1 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114239716A (en) * | 2021-12-15 | 2022-03-25 | 平安科技（深圳）有限公司 | Double-shaft interaction method, device, equipment and medium based on crop growth vigor |
| CN116069436A (en) * | 2023-03-17 | 2023-05-05 | 阿里巴巴达摩院(杭州)科技有限公司 | Remote sensing image processing method, electronic equipment and storage medium |
| CN116069436B (en) * | 2023-03-17 | 2023-07-21 | 阿里巴巴达摩院(杭州)科技有限公司 | Remote sensing image processing method, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Barrera-Animas et al. | Rainfall prediction: A comparative analysis of modern machine learning algorithms for time-series forecasting | |
| Douna et al. | Towards a remote sensing data based evapotranspiration estimation in Northern Australia using a simple random forest approach | |
| Zhang et al. | Bayesian calibration of AquaCrop model for winter wheat by assimilating UAV multi-spectral images | |
| CN114529097B (en) | Multi-scale crop phenological period remote sensing dimensionality reduction prediction method | |
| CN113343783A (en) | Intelligent crop identification and growth prediction method and system | |
| CN117036088A (en) | Data acquisition and analysis method for identifying growth situation of greening plants by AI | |
| CN117114913A (en) | Intelligent agricultural data acquisition system based on big data | |
| Bessac et al. | Stochastic simulation of predictive space–time scenarios of wind speed using observations and physical model outputs | |
| CN115169728A (en) | Soil fertility prediction method based on simplified neural network | |
| Jiang et al. | Monitoring the coastal environment using remote sensing and GIS techniques | |
| Riihimaki et al. | Improving prediction of surface solar irradiance variability by integrating observed cloud characteristics and machine learning | |
| Parida et al. | Crop types discrimination and yield prediction using sentinel-2 data and aquacrop model in Hazaribagh District, Jharkhand | |
| CN117423003B (en) | Winter wheat seedling condition grading remote sensing monitoring method in overwintering period | |
| CN116863341B (en) | Crop classification and identification method and system based on time sequence satellite remote sensing image | |
| Kumawat et al. | Time-Variant Satellite Vegetation Classification Enabled by Hybrid Metaheuristic-Based Adaptive Time-Weighted Dynamic Time Warping | |
| Chtioui et al. | Moisture prediction from simple micrometeorological data | |
| CN117114513A (en) | Image-based crop pesticide and fertilizer use evaluation method, device, equipment and medium | |
| Jadhav et al. | Segmentation analysis using particle swarm optimization-self organizing map algorithm and classification of remote sensing data for agriculture | |
| Kanwal et al. | Yield forecasting and assessment of interannual wheat yield variability using machine learning approach in semiarid environment. | |
| CN116151454A (en) | Method and system for predicting yield of short-forest linalool essential oil by multispectral unmanned aerial vehicle | |
| Liang et al. | Bayesian MODIS NDVI back-prediction by intersensor calibration with AVHRR | |
| Mohan et al. | Temporal convolutional network based rice crop yield prediction using multispectral satellite data | |
| Hachicha et al. | Prediction of plant growth based on statistical methods and remote sensing data | |
| Chaudhary et al. | Deep Learning Approach for Forecasting Apple Yield Using Soil Parameters | |
| CN117907248B (en) | Remote sensing monitoring method and system for root system soil water content in key growth period of winter wheat |
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 |