CN113435345A - Growth stage determination method and device, agricultural system, equipment and storage medium - Google Patents
Growth stage determination method and device, agricultural system, equipment and storage medium Download PDFInfo
- Publication number
- CN113435345A CN113435345A CN202110728351.4A CN202110728351A CN113435345A CN 113435345 A CN113435345 A CN 113435345A CN 202110728351 A CN202110728351 A CN 202110728351A CN 113435345 A CN113435345 A CN 113435345A
- Authority
- CN
- China
- Prior art keywords
- target
- growth stage
- image
- pixel distribution
- distribution information
- 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
- 230000012010 growth Effects 0.000 title claims abstract description 199
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000003860 storage Methods 0.000 title claims abstract description 14
- 238000009826 distribution Methods 0.000 claims abstract description 109
- 238000012549 training Methods 0.000 claims abstract description 39
- 238000012545 processing Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 6
- 230000003698 anagen phase Effects 0.000 claims description 2
- 238000012271 agricultural production Methods 0.000 abstract description 2
- 241000209094 Oryza Species 0.000 description 14
- 235000007164 Oryza sativa Nutrition 0.000 description 14
- 235000009566 rice Nutrition 0.000 description 14
- 238000010586 diagram Methods 0.000 description 11
- 241000196324 Embryophyta Species 0.000 description 7
- 238000004891 communication Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 206010019345 Heat stroke Diseases 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 240000007594 Oryza sativa Species 0.000 description 1
- 241000209140 Triticum Species 0.000 description 1
- 235000021307 Triticum Nutrition 0.000 description 1
- 230000009418 agronomic effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 239000003337 fertilizer Substances 0.000 description 1
- 238000003306 harvesting Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- 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
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Animal Husbandry (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Agronomy & Crop Science (AREA)
- Evolutionary Biology (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)
- Image Analysis (AREA)
Abstract
The embodiment of the application provides a growth stage determining method and device, an agricultural system, equipment and a storage medium, and relates to the technical field of agricultural production. The method comprises the following steps: obtaining a target image of a target area; and determining a target growth stage of the target crop in the target area based on the pixel distribution information in the target image through the trained recognition model, wherein the recognition model is obtained by training according to the sample pixel distribution information in the sample image corresponding to the target crop and the growth stage corresponding to the sample image. Therefore, the growth stage of the target crop can be automatically and accurately determined according to the pixel distribution information in the image of the area where the target crop is located, manual on-site observation is not needed, a large amount of manpower can be saved, and meanwhile the accuracy of the determined growth stage can be guaranteed.
Description
Technical Field
The application relates to the technical field of agricultural production, in particular to a growth stage determining method and device, an agricultural system, equipment and a storage medium.
Background
Under the condition of determining the growth stage of the crops, the crops can be subjected to accurate and quantitative fertilizer and water management according to the growth stage so as to realize high-yield cultivation. Therefore, determining the growth stage of a crop is the first step in achieving high-yield cultivation. However, the current growth stage of the crop is mainly determined by manual field observation, which consumes a lot of manpower, and the obtained result is inaccurate in case of poor agricultural basic knowledge because the observer directly determines the growth stage according to the agricultural basic knowledge.
Disclosure of Invention
The embodiment of the application provides a growth stage determining method and device, an agricultural system, equipment and a storage medium, which can accurately determine the growth stage of a target crop according to pixel distribution information in an image of an area where the target crop is located without manual field observation.
The embodiment of the application can be realized as follows:
in a first aspect, an embodiment of the present application provides a growth stage determining method, including:
obtaining a target image of a target area;
and determining a target growth stage of the target crop in the target area based on the pixel distribution information in the target image through a trained recognition model, wherein the recognition model is obtained by training according to the sample pixel distribution information in the sample image corresponding to the target crop and the growth stage corresponding to the sample image.
In a second aspect, an embodiment of the present application provides a growth stage determining apparatus, including:
the acquisition module is used for acquiring a target image of a target area;
and the determining module is used for determining the target growth stage of the target crop in the target area based on the pixel distribution information in the target image through a trained recognition model, wherein the recognition model is obtained by training according to the sample pixel distribution information in the sample image corresponding to the target crop and the growth stage corresponding to the sample image.
In a third aspect, an embodiment of the present application provides an intelligent agricultural system, which includes an obtaining module, a processing module, and a display module; the acquisition module is used for acquiring a target image of a target area; the processing module is used for processing the target image through the growth stage determining method provided by the embodiment of the application to obtain a target growth stage of the target crop in the target area; the display module is used for displaying the target area and the target growth stage of the target crop.
In a fourth aspect, an embodiment of the present application provides an electronic device, including a processor and a memory, where the memory stores machine executable instructions capable of being executed by the processor, and the processor can execute the machine executable instructions to implement the growth stage determination method described in the foregoing embodiment.
In a fifth aspect, embodiments of the present application provide a readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the growth stage determination method as described in the foregoing embodiments.
According to the growth stage determining method and device, the agricultural system, the agricultural equipment and the storage medium, the target growth stage of the target crop in the target area is determined through the trained recognition model based on the obtained pixel distribution information in the target image of the target area where the target crop is located. Therefore, the growth stage of the target crop can be automatically and accurately determined without manual field observation, and a large amount of labor can be saved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.
Fig. 1 is a block diagram of an electronic device according to an embodiment of the present disclosure;
fig. 2 is a schematic flowchart of a growth stage determining method according to an embodiment of the present disclosure;
FIG. 3 is a schematic illustration of an ortho image provided by an embodiment of the present application;
FIG. 4 is a flowchart illustrating one of the sub-steps included in step S130 of FIG. 2;
FIG. 5 is a second schematic flowchart of the sub-steps included in step S130 in FIG. 2;
fig. 6 is a schematic block diagram of a growth stage determining apparatus according to an embodiment of the present disclosure;
fig. 7 is a second schematic block diagram of a growth stage determining apparatus according to an embodiment of the present application.
Icon: 100-an electronic device; 110-a memory; 120-a processor; 130-a communication unit; 200-growth phase determining means; 210-a training module; 220-an acquisition module; 230 — a determination module.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.
It is to be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus 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 apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
At present, the current growth stage of the crops is mainly determined by manual field observation. However, this approach has the following disadvantages: the growth stage identification needs a certain agronomic basis, the culture degree is low, and the growth stage of crops is difficult to accurately identify by individual farmers who do not go through system learning, so that high-yield cultivation cannot be effectively carried out based on the growth stage, and high yield and high harvest are achieved; because of the field observation, a large amount of manpower is consumed, for example, a large area of farmland needs a large amount of manpower to determine the growth stage of crops; the weather condition may cause a heatstroke to the observer, for example, when the growth stage of rice is observed in the field in summer, the field observation at a high temperature is likely to cause a heatstroke.
In order to solve the above problems, embodiments of the present application provide a method, an apparatus, an agricultural system, an apparatus, and a storage medium for determining a growth stage, which are used for automatically and accurately determining the growth stage of a target crop according to pixel distribution information in an image of an area where the target crop is located, without manual field observation.
Some embodiments of the present application will be described in detail below with reference to the accompanying drawings. The embodiments described below and the features of the embodiments can be combined with each other without conflict.
Referring to fig. 1, fig. 1 is a block diagram of an electronic device 100 according to an embodiment of the present disclosure. The electronic device 100 may be, but is not limited to, a computer or a server. The electronic device 100 may include a memory 110, a processor 120, and a communication unit 130. The elements of the memory 110, the processor 120 and the communication unit 130 are electrically connected to each other directly or indirectly to realize data transmission or interaction. For example, the components may be electrically connected to each other via one or more communication buses or signal lines.
The memory 110 is used to store programs or data. The Memory 110 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like.
The processor 120 is used to read/write data or programs stored in the memory 110 and perform corresponding functions. For example, the memory 110 stores the growth stage determining device 200, and the growth stage determining device 200 includes at least one software function module which can be stored in the memory 110 in the form of software or firmware (firmware). The processor 120 executes various functional applications and data processing by running software programs and modules stored in the memory 110, such as the growth stage determination apparatus 200 in the embodiment of the present application, so as to implement the growth stage determination method in the embodiment of the present application.
The communication unit 130 is used for establishing a communication connection between the electronic apparatus 100 and another communication terminal via a network, and for transceiving data via the network.
It should be understood that the structure shown in fig. 1 is only a schematic structural diagram of the electronic device 100, and the electronic device 100 may also include more or fewer components than shown in fig. 1, or have a different configuration than shown in fig. 1. The components shown in fig. 1 may be implemented in hardware, software, or a combination thereof.
Referring to fig. 2, fig. 2 is a flowchart illustrating a growth stage determining method according to an embodiment of the present disclosure. The method may be applied to the electronic device 100 described above. The specific flow of the growth stage determination method is explained in detail below.
Step S120, a target image of the target area is obtained.
And S130, determining a target growth stage of the target crop in the target area based on the pixel distribution information in the target image through the trained recognition model.
In this embodiment, the recognition model is obtained by training according to sample pixel distribution information in a sample image corresponding to the target crop and a growth stage corresponding to the sample image. For example, if the target crop to be identified is rice, the crop targeted by the data used in the training process of the identification model is also rice, that is, the class is the same.
It can be understood that, in order to ensure the accuracy of the recognition result, the target crop targeted for training the recognition model and the target crop to be recognized may be the same variety of crops in the same category, for example, all the crops are rice of variety 1; or different varieties of crops with the same growth stage in the whole life cycle, for example, the rice of variety 1 and the rice of variety 2, but the rice of variety 1 and the rice of variety 2 have the same growth stage in the whole life cycle. It should be noted that the foregoing is merely an example, and no specific limitation is imposed on this in the embodiment of the present application, and the specific variety of the crop corresponding to the information used in training the recognition model may be determined according to actual requirements, as long as the growth stage of the crop can be accurately obtained based on the recognition model.
The identification model may be obtained by training the device performing the growth stage determination in advance, or may be obtained by training other devices and then sending the trained devices to the device performing the growth stage determination.
The target image of the target area where the target crop is located can be obtained in any manner. And then determining the growth stage of the target crop planted in the target area by using the identification model and the pixel distribution information in the target image.
In a planting area, under the condition that the planting density of crops is constant, various conditions of the crops are changed along with the continuous growth of the crops. For example, as the crop grows, the leaf area index of the crop increases, and the vegetation coverage in the planting area gradually increases; the number of leaves of an individual crop is increased or decreased; the area of the leaf of the crop individual is increased; increased crop intensity, etc. The image of the planting area may show the information, i.e. the pixel distribution information of the image of the planting area embodies the information of the crops within the planting area. For example, as the crop grows, the vegetation coverage in the planting area gradually increases, and the change in the overall appearance of the planting area can be directly reflected in the image. Therefore, the vegetation growth change in the planting area can be represented by a series of images. The pixel distribution information in the image may include vegetation coverage, the number of leaves of the crop, a pixel distribution histogram (including distribution probability or occurrence frequency of pixel values in the image), each pixel position and its corresponding pixel value, and the like. Therefore, the growth stage of the target crop can be accurately determined according to the pixel distribution information of the target area where the target crop is located.
It should be noted that, because the identification model in the embodiment of the present application is obtained by training based on the pixel distribution information of the planting area image, and the pixel distribution information can ensure the change of the overall appearance of the planting area, it can be understood that: the identification model is used for identifying and obtaining the corresponding crop growth stage based on the image for representing the whole appearance change condition of the planting area, so that the effect of improving the accuracy can be generated, and compared with a mode of realizing each crop growth stage one by one based on a sufficient number of single plants, the method for determining the growth stage has the advantages of higher identification efficiency, simpler algorithm and the like.
According to the embodiment of the application, the target growth stage of the target crop in the target area is determined through the trained recognition model based on the obtained pixel distribution information in the target image of the target area where the target crop is located. Therefore, the growth stage of the target crop can be automatically and accurately determined without manual field observation, a large amount of labor can be saved, and the condition that the heatstroke and the like of a person are caused by manual field observation in the process of determining the growth stage can be avoided.
In this embodiment, the target image of the target area may be obtained according to the received input operation, for example, an image manually selected by a person to be used as the target image. And receiving an image sent by other equipment, and taking the image as the target image. It is understood that the target image may also be obtained by a method, which is not specifically limited herein.
Optionally, as a possible implementation manner, a region image of the target region may be obtained, and the region image may be used as the target image. Alternatively, the region image may be obtained by: and carrying a visible light camera by adopting an unmanned aerial vehicle, and acquiring a visible light image of a target area.
The area images adopted in the training stage and the application stage of the recognition model can be acquired by the unmanned aerial vehicle based on the same shooting height and shooting angle, and also can be acquired by the unmanned aerial vehicle based on different shooting heights and shooting angles, and the acquired area images can show the vegetation growth change condition of the target area no matter what shooting height and shooting angle are used for acquiring. In an alternative embodiment, if there is a case that the shooting heights and shooting angles of the area images adopted by the recognition model in the training stage and the application stage are required to be the same, the device for acquiring the area images may be controlled to shoot the area images at the shooting heights and shooting angles consistent with those of the area images used in training the recognition model in the application stage of the recognition model.
Wherein the region image may be an image of the target region obtained at an arbitrary angle.
In an implementation manner of this embodiment, the area image may be an orthoimage of the target area. An orthographic image is an aerial photograph that has been geometrically corrected (e.g., to have a uniform scale), and is equivalent to an orthographic projected aerial photograph. As shown in fig. 3, in the orthographic image, various conditions (e.g., vegetation coverage, leaf size) of the crop in the target area are reflected more accurately and intuitively. Therefore, by using the orthographic image of the target area as the target image, the accuracy of the determined growth stage can be further ensured.
The target crop may be any crop. Alternatively, the target crop may be a graminaceous plant, such as rice, wheat, and the like. When the target crop is a gramineous plant, the growth stage can be expressed by leaf age. Correspondingly, an orthographic image of the target area in which the gramineous plant is planted may be taken as the target image.
In the case of obtaining the target image, a target growth stage of the target crop in the target area may be obtained based on pixel distribution information in the target image through the recognition model.
The pixel distribution information in the target image may include global pixel distribution information, where the global pixel distribution information is used to represent the distribution of the whole pixels of the target image from the perspective of the whole image. The pixel distribution information may also include local pixel distribution information, where the local pixel distribution information is used to indicate a pixel distribution condition of a local image from a local image perspective, and an area of the local image is smaller than an area of the entire image. For example, the local pixel distribution information is used to represent the pixel distribution of one or more target crop individuals, that is, the local image may include only one target crop individual, or may include a plurality of target crop individuals.
As an optional implementation manner, the pixel distribution information in the target image includes global pixel distribution information and local pixel distribution information, so that the target growth stage of the target crop in the target image is identified based on the global feature and the local feature of the target image, and the identification accuracy is improved.
Optionally, as an optional implementation manner, the target image may be input into the recognition model, and the output of the recognition model is taken as the target growth stage. That is, the input of the recognition model is an image and the output is a growth stage. Thereby, a rapid determination of the growth stage is facilitated.
Optionally, as an optional implementation manner, pixel distribution information in the target image may be extracted first, and then the pixel distribution information is input into the recognition model, and the output of the recognition model is taken as the target growth stage. That is, the output of the recognition model is the pixel distribution information, and the output is the growth stage. Therefore, the recognition model does not need to perform feature extraction operation, and the growth stage is convenient to determine quickly. It is understood that, correspondingly, in this embodiment, the recognition model may be trained according to the sample pixel distribution information and the corresponding growth stage.
Alternatively, the target growth stage may also be obtained by the steps shown in fig. 4. Referring to fig. 4, fig. 4 is a flowchart illustrating sub-steps included in step S130 in fig. 2. Step S130 may include sub-step S131 and sub-step S132.
And a substep S131, obtaining the weight corresponding to each growth stage of the pixel distribution information in the target image through the identification model.
In this embodiment, the target image may be input to the recognition model, so as to obtain the weight corresponding to each growth stage according to the pixel distribution in the target image, that is, the output of the recognition model is the weight corresponding to each growth stage according to the pixel distribution in the target image.
And a substep S132 of determining the target growth stage according to each growth stage and the weight corresponding to each growth stage.
The target growth stage may be expressed by a specific growth stage, for example, in the case where the target crop is rice, the target growth stage may be expressed by a seedling stage, an effective tillering stage, a tillering prosperity stage, or the like. The target growth stage can also be represented by numerical values, for example, the growth stage of the gramineous plant can be represented by specific leaf age values, such as 1 leaf age, 2 leaf age, … …, 20 leaf age and the like, and the leaf age is the main stem and leaf number of the gramineous plant.
Optionally, the weights corresponding to the growth stages may be compared to determine the maximum weight, and the growth stage corresponding to the maximum weight is used as the target growth stage of the target crop.
Optionally, in a case that the growth stages are represented by numerical values, the target growth stage may be obtained by weighted average processing according to each growth stage and the weight corresponding to each growth stage. Thereby, the accuracy of the target growth phase can be guaranteed.
For example, if the target crop is rice, according to the pixel distribution information in the target image, the obtained growth stages and the weights corresponding to the growth stages through the recognition model are as follows: 1-6 leaf age, 0; 7 leaves old, 0.2; 8 leaves old, 0.7; 9 leaves old, 0.1; 0, 10-20 leaves old; in this case, the target growth stage can be determined to be 7.9 leaf ages by a weighted average process. Compared with the mode of directly determining the target growth stage as an integral leaf age, the predicted leaf age is closer to the actual leaf age.
Under the condition that the range of the target area is small, the condition of the target crop in the target area can be clearly presented by one target image, and at the moment, the target growth stage of the target crop in the target area can be obtained based on the identification model according to the target image.
In the case where the range of the target area is large, the condition of the target crop in the target area may not be clearly presented by one target image. In this case, the target growth stage of the target crop can be obtained in the manner shown in fig. 5.
Referring to fig. 5, fig. 5 is a second schematic flowchart illustrating the sub-steps included in step S130 in fig. 2. Step S130 may include that the method may further comprise sub-step S134 and sub-step S135.
And a substep S134, obtaining, through the recognition model, target growth stages corresponding to the at least one target sub-region respectively based on pixel distribution information in the image of the at least one target sub-region in the target region.
And a substep S135, determining a target growth stage corresponding to the target region according to the target growth stage corresponding to each of the at least one target sub-region.
Wherein the target region may comprise a plurality of target sub-regions. In a case where a target growth stage of a target crop planted in a target area is to be determined, when the target area is relatively large, an image of at least one target sub-area among a plurality of target sub-areas included in the target area may be obtained. And then, obtaining a target growth stage corresponding to each of the at least one target sub-region through the identification model according to the obtained pixel distribution information in the image of the at least one target sub-region.
For example, the target area includes target sub-areas 1, 2, and 3, images of the target sub-areas 1, 2, and 3 may be obtained, and then a target growth stage of a target crop in the target sub-area 1 is obtained based on pixel distribution information in the image of the target sub-area 1 through the recognition model; in the same way, the target growth stage corresponding to the target subregion 2 and the target growth stage corresponding to the target subregion 3 are obtained.
Then, the target growth stage of the target crop in the target area can be determined according to the target growth stage of the target crop in the at least one target sub-area.
Alternatively, in the case where a target growth stage of a target crop in only one target sub-area is obtained, the target growth stage may be directly used as the target growth stage of the target crop in the target area.
Optionally, in a case where the at least one target sub-area is a plurality of target sub-areas, that is, in a case where a target growth stage of a target crop in the plurality of target sub-areas is obtained, a target growth stage corresponding to a mode may be used as the target growth stage of the target crop in the target area.
When the growth stage is represented by a numerical value and the target growth stages of the target crops in the target sub-regions are obtained, the average value of the target growth stages corresponding to the target sub-regions may be used as the target growth stage of the target crops in the target region. Therefore, the target growth stage corresponding to the target area can be obtained by sampling different sampling points, the identification accuracy is high, and the overall error caused by local difference can be reduced.
Before step S130, the method may further include step S111 and step S112.
Step S111, a sample set is obtained.
Step S112, training an initial model through a sample set until the deviation between a predicted growth stage obtained by processing the initial model based on the sample pixel distribution information of the current sample image and a growth stage corresponding to the current sample image is smaller than a preset deviation threshold value, so as to obtain the identification model.
It should be noted that the above process of training the initial model to obtain the recognition model may not be a necessary process of the growth stage determination method in the embodiment of the present application, and it is understood that, before the growth stage determination method in the embodiment of the present application is executed, the recognition model may be obtained based on the initial model training, and then in the growth stage determination method in the embodiment of the present application, as long as the recognition model is used to determine the growth stage of the crop, the training of the recognition model is not required.
In this embodiment, the sample set may include samples corresponding to different growth stages of the target crop during the whole growth period. Each sample in the sample set may include a sample image of the target crop and a growth stage corresponding to the sample image. Alternatively, the growth stage included in the sample may be determined by manual field observation or may be obtained by other means, and is not particularly limited herein.
In the case where the sample set is obtained, an initial model may be trained using the sample set to obtain the recognition model. Wherein the initial model may be a convolutional neural network.
During training, the sample pixel distribution information in the sample image may include global pixel distribution information of the sample image, where the global pixel distribution information is determined by the overall pixel distribution in the sample image, that is, the sample pixel distribution information includes global pixel distribution information used to characterize the overall pixel distribution of the sample image.
Alternatively, an initial model may be used to extract sample pixel distribution information in the sample image, and then the initial model may be used to predict a growth stage corresponding to the sample image based on the sample pixel distribution information as a prediction result. And then, adjusting parameters in the initial model according to the prediction result and the growth stage corresponding to the sample image so as to update the initial model. Then, based on the updated initial model, repeating the above process (i.e. the process of updating the initial model according to the sample image) until the initial model meeting the requirements is obtained, and at this time, taking the initial model meeting the requirements as the identification model; the initial model meeting the requirements can represent that the deviation between the prediction result output by the initial model and the actual growth stage of the corresponding sample image is smaller than a preset deviation threshold value; the preset deviation threshold may be set according to the actual prediction accuracy requirement, and is not limited herein.
The sample pixel distribution information in the sample image may further include local pixel distribution information of the sample image. The local pixel distribution information is used to indicate the pixel distribution of the local image from the viewpoint of the local image. For example, the local pixel distribution information of the sample image is used for representing the pixel distribution condition of one or more target crop individuals in a local area in the sample image. It can be understood that the growth stages corresponding to the global pixel distribution information and the local pixel distribution information in a sample image are the same.
The sample pixel distribution information in the sample image extracted by using the initial model may include global pixel distribution information of the sample image, where the global pixel distribution information is determined by an overall pixel distribution condition in the sample image.
Optionally, the extracted sample pixel distribution information in the sample image may further include local pixel distribution information of the sample image. The local pixel distribution information is used to indicate the pixel distribution of the local image from the viewpoint of the local image. For example, the local pixel distribution information of the sample image is used for representing the pixel distribution condition of one or more target crop individuals in a local area in the sample image. It can be understood that the growth stages corresponding to the global pixel distribution information and the local pixel distribution information in a sample image are the same. Therefore, training can be performed based on the global features and the local features so as to improve the identification accuracy of the identification model obtained by the training mode.
The device that executes steps S120 to S130 and the device that executes steps S111 to S112 may be the same device or different devices.
The process of obtaining the identification model is briefly illustrated below by taking the target crop as rice as an example.
The target image is represented by an orthoimage. An unmanned aerial vehicle can be used for carrying a visible light camera, visible light images of the field are collected at a certain angle and height at multiple points regularly, and the visible light images are orthographic images; meanwhile, the leaf age of the field is collected manually. Images collected at different periods in the whole growth period of the rice are used as training sets, and corresponding leaf ages of the images are used as labels.
And then, training the convolutional neural network by using the training set and the labels in a machine learning mode, and establishing a relation model between the image and the leaf age, wherein the trained relation model is the recognition model.
In this relational model, leaf ages can be classified into 20 types in advance, namely 1 leaf age, 2 leaf age, 3 leaf age, … …, and 20 (leaf age). The 20 types basically comprise all possible main stem total leaf ages of the current common rice variety types, and the capping leaf age for rice growth is 20.
And inputting the acquired image and the leaf age corresponding to the acquired image into the relational model. Since the leaf ages recorded in the actual growth are usually counted one decimal place, such as 2.8 leaf ages, and the classification is performed according to integers, the relational model can perform classification on the recorded leaf ages by adopting a rounding method, such as classifying the 2.8 leaf ages into 3 leaf ages, and classifying the 5.2 leaf ages into 5 leaf ages. Of course, it is understood that the relationship model may be used to classify leaf age in other ways. Or classifying the collected leaf age manually or in other ways, and inputting the classified leaf age and the image into the relational model.
The relational model may include a convolution layer, where the convolution layer is used to perform feature extraction on an image, that is, to extract pixel distribution information in the image, and pixels of the pixel distribution information in the image may include global pixel distribution information and local pixel distribution information of the image. The relational model may determine a weight corresponding to each leaf age for the feature based on the extracted feature. Next, the predicted leaf age is determined based on each leaf age and the weight corresponding to each leaf age. For example, the predicted leaf age is calculated by a weighted average method according to each leaf age and the weight corresponding to each leaf age. Then, based on the predicted leaf age and the corresponding leaf age as a label, a loss value is determined. Parameters in the relational model can be adjusted based on the loss values, then the training process is repeated based on the adjusted model until the training stopping condition is reached, and the relational model obtained when the training is stopped is used as a trained recognition model. The training stopping condition may be set according to an actual situation, for example, the training is determined to be stopped when the calculated loss value is smaller than a preset loss value, or the training is stopped when the number of times of training reaches a preset number of times.
After an image is input into the recognition model, the recognition model can recognize the weight of each characteristic value in the image corresponding to each characteristic set of 20 leaf ages, and then a leaf age result is obtained through weighted averaging based on each leaf age and the corresponding weight, wherein the leaf age result represents the recognized leaf age.
In order to perform the corresponding steps in the above embodiments and various possible manners, an implementation manner of the growth stage determining apparatus 200 is given below, and optionally, the growth stage determining apparatus 200 may adopt the device structure of the electronic device 100 shown in fig. 1. Further, referring to fig. 6, fig. 6 is a block diagram illustrating a growth stage determining apparatus 200 according to an embodiment of the present disclosure. It should be noted that the basic principle and the technical effects of the growth stage determining apparatus 200 provided in the present embodiment are the same as those of the above embodiments, and for the sake of brief description, no part of the present embodiment is mentioned, and reference may be made to the corresponding contents in the above embodiments. The growth stage determining apparatus 200 may include: an acquisition module 220 and a determination module 230.
The obtaining module 220 is configured to obtain a target image of a target area.
The determining module 230 is configured to determine, through a trained recognition model, a target growth stage of the target crop in the target area based on the pixel distribution information in the target image, where the recognition model is obtained by training according to the sample pixel distribution information in the sample image corresponding to the target crop and the growth stage corresponding to the sample image.
Optionally, in this embodiment, the determining module 230 is specifically configured to: obtaining the weight of the pixel distribution information in the target image corresponding to each growth stage through the identification model; and determining the target growth stage according to each growth stage and the corresponding weight of each growth stage.
Optionally, in this embodiment, each growth stage is represented by a numerical value, and the determining module 230 is specifically configured to: and obtaining the target growth stage through weighted average processing according to each growth stage and the corresponding weight of each growth stage.
Optionally, in this embodiment, the obtaining module 220 is specifically configured to: obtaining an ortho-image of the target area, or obtaining an ortho-image of the target area planted with a gramineous plant.
Optionally, in this embodiment, the target region includes a plurality of target sub-regions, the target image of the target region includes an image of at least one target sub-region, and the determining module 230 is specifically configured to: obtaining, by the identification model, a target growth stage corresponding to each of at least one target sub-region based on pixel distribution information in an image of the at least one target sub-region within the target region; and determining a target growth stage corresponding to the target area according to the target growth stage corresponding to each of the at least one target sub-area.
Referring to fig. 7, fig. 7 is a second block diagram of a growth stage determining apparatus 200 according to an embodiment of the present disclosure. Optionally, the growth stage determining apparatus 200 may further include a training module 210.
The training module 210 is configured to: obtaining a sample set, wherein each sample in the sample set comprises a sample image corresponding to the target crop and a growth stage corresponding to the sample image; training an initial model through a sample set until the deviation between a predicted growth stage obtained by processing the initial model based on sample pixel distribution information of a current sample image and a growth stage corresponding to the current sample image is smaller than a preset deviation threshold value so as to obtain the identification model; wherein the sample pixel distribution information comprises: global pixel distribution information for characterizing an overall pixel distribution of the sample image.
Optionally, in this embodiment, the sample pixel distribution information further includes local pixel distribution information, where the local pixel distribution information represents a pixel distribution condition of at least one target crop individual.
Alternatively, the modules may be stored in the memory 110 shown in fig. 1 in the form of software or Firmware (Firmware) or may be fixed in an Operating System (OS) of the electronic device 100, and may be executed by the processor 120 in fig. 1. Meanwhile, data, codes of programs, and the like required to execute the above-described modules may be stored in the memory 110.
The embodiment of the application also provides an intelligent agricultural system, which comprises an acquisition module, a processing module and a display module; the acquisition module is used for acquiring a target image of a target area; the processing module is used for processing the target image by the growth stage determining method provided by any embodiment of the application to obtain a target growth stage of the target crop in the target area; the display module is used for displaying the target area and the target growth stage of the target crop.
It should be noted that the intelligent agricultural system may provide a user interface for a user, and the user may operate in the user interface to trigger the intelligent agricultural system to acquire a target image of a target area specified by the user, and determine a growth stage of crops in the target area based on the target image. The intelligent agricultural system comprises an acquisition module, a display module and a display module, wherein the acquisition module of the intelligent agricultural system can acquire a target image in any one of the following modes:
in the first mode, under the condition that the target image is stored, the target image can be directly obtained to the cloud or obtained from a data storage of the intelligent agricultural system;
in the second mode, when the target image needs to be acquired in real time, an image capturing instruction can be sent to the unmanned aerial vehicle or the image acquisition device configured in the target area to control the unmanned aerial vehicle or the image acquisition device to capture and return the image.
The embodiments of the present application further provide a readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the growth stage determining method provided in any embodiment of the present application.
In summary, the embodiments of the present application provide a method and an apparatus for determining a growth stage, an agricultural system, an apparatus, and a storage medium, which determine a target growth stage of a target crop in a target area based on obtained pixel distribution information in a target image of the target area where the target crop is located through a trained recognition model. Therefore, the growth stage of the target crop can be automatically and accurately determined without manual field observation, and a large amount of labor can be saved.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative, and for example, the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (11)
1. A method for determining a growth stage, comprising:
obtaining a target image of a target area;
and determining a target growth stage of the target crop in the target area based on the pixel distribution information in the target image through a trained recognition model, wherein the recognition model is obtained by training according to the sample pixel distribution information in the sample image corresponding to the target crop and the growth stage corresponding to the sample image.
2. The method of claim 1, wherein determining the target growth stage of the target crop in the target area based on the pixel distribution information in the target image through the trained recognition model comprises:
obtaining the weight of the pixel distribution information in the target image corresponding to each growth stage through the identification model;
and determining the target growth stage according to each growth stage and the corresponding weight of each growth stage.
3. The method of claim 2, wherein each growth stage is represented by a numerical value, and wherein determining the target growth stage according to each growth stage and the corresponding weight of each growth stage comprises:
and obtaining the target growth stage through weighted average processing according to each growth stage and the corresponding weight of each growth stage.
4. The method of claim 1, wherein obtaining the target image of the target area comprises:
obtaining an ortho-image of the target area, or obtaining an ortho-image of the target area planted with a gramineous plant.
5. The method according to any one of claims 1-4, wherein the target area comprises a plurality of target sub-areas, the target image comprises an image of at least one target sub-area, and the determining the target growth stage of the target crop in the target area based on the pixel distribution information in the target image through the trained recognition model comprises:
obtaining, by the identification model, a target growth stage corresponding to each of at least one target sub-region based on pixel distribution information in an image of the at least one target sub-region within the target region;
and determining a target growth stage corresponding to the target area according to the target growth stage corresponding to each of the at least one target sub-area.
6. The method according to any one of claims 1 to 4, wherein the training process of the recognition model comprises:
obtaining a sample set, wherein each sample in the sample set comprises a sample image corresponding to the target crop and a growth stage corresponding to the sample image;
training an initial model through a sample set until the deviation between a predicted growth stage obtained by processing the initial model based on sample pixel distribution information of a current sample image and a growth stage corresponding to the current sample image is smaller than a preset deviation threshold value so as to obtain the identification model;
wherein the sample pixel distribution information comprises: global pixel distribution information for characterizing an overall pixel distribution of the sample image.
7. The method of claim 6, wherein the sample pixel distribution information further comprises local pixel distribution information, wherein the local pixel distribution information represents a pixel distribution of at least one target crop individual.
8. A growth stage determining apparatus, comprising:
the acquisition module is used for acquiring a target image of a target area;
and the determining module is used for determining the target growth stage of the target crop in the target area based on the pixel distribution information in the target image through a trained recognition model, wherein the recognition model is obtained by training according to the sample pixel distribution information in the sample image corresponding to the target crop and the growth stage corresponding to the sample image.
9. An intelligent agricultural system is characterized by comprising an acquisition module, a processing module and a display module; the acquisition module is used for acquiring a target image of a target area; the processing module is used for processing the target image through the growth stage determination method of any one of claims 1 to 7 to obtain a target growth stage of the target crop in the target area; the display module is used for displaying the target area and the target growth stage of the target crop.
10. An electronic device comprising a processor and a memory, the memory storing machine executable instructions executable by the processor to implement the growth stage determination method of any one of claims 1-7.
11. A readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the growth phase determination method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110728351.4A CN113435345A (en) | 2021-06-29 | 2021-06-29 | Growth stage determination method and device, agricultural system, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110728351.4A CN113435345A (en) | 2021-06-29 | 2021-06-29 | Growth stage determination method and device, agricultural system, equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113435345A true CN113435345A (en) | 2021-09-24 |
Family
ID=77757708
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110728351.4A Pending CN113435345A (en) | 2021-06-29 | 2021-06-29 | Growth stage determination method and device, agricultural system, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113435345A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115413550A (en) * | 2022-11-07 | 2022-12-02 | 中化现代农业有限公司 | Beet plant protection method and beet plant protection equipment |
| CN116863403A (en) * | 2023-07-11 | 2023-10-10 | 仲恺农业工程学院 | Crop big data environment monitoring method and device and electronic equipment |
-
2021
- 2021-06-29 CN CN202110728351.4A patent/CN113435345A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115413550A (en) * | 2022-11-07 | 2022-12-02 | 中化现代农业有限公司 | Beet plant protection method and beet plant protection equipment |
| CN115413550B (en) * | 2022-11-07 | 2023-03-14 | 中化现代农业有限公司 | Beet plant protection method and beet plant protection equipment |
| CN116863403A (en) * | 2023-07-11 | 2023-10-10 | 仲恺农业工程学院 | Crop big data environment monitoring method and device and electronic equipment |
| CN116863403B (en) * | 2023-07-11 | 2024-01-02 | 仲恺农业工程学院 | Crop big data environment monitoring method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11580731B2 (en) | Systems, devices, and methods for in-field diagnosis of growth stage and crop yield estimation in a plant area | |
| EP3482630B1 (en) | Method, system and computer program for performing a pest forecast | |
| CN111767802B (en) | Method and device for detecting abnormal state of object | |
| KR102200314B1 (en) | system for monitoring agricultural produce using drone | |
| WO2016028612A1 (en) | Continual crop development profiling using dynamical extended range weather forecasting with routine remotely-sensed validation imagery | |
| CN113435345A (en) | Growth stage determination method and device, agricultural system, equipment and storage medium | |
| CN108140118A (en) | For identifying the method and information system of at least one plant plant being planted on arable land | |
| CN116012720B (en) | Crop growth condition monitoring method, device and medium based on high-resolution remote sensing image | |
| CN111985724B (en) | Crop yield estimation method, device, equipment and storage medium | |
| CN113469112B (en) | Crop growth condition image identification method and system | |
| CN115620151B (en) | Method and device for identifying phenological period, electronic equipment and storage medium | |
| CN114359745A (en) | Information processing method, storage medium, and electronic device | |
| CN115601585A (en) | Agricultural pest and disease diagnosis method and device based on picture analysis | |
| JP2021073860A (en) | Crop growth estimation device, crop growth estimation system, crop growth estimation method and program | |
| US20190012749A1 (en) | Dynamic cost function calculation for agricultural users | |
| CN116579521B (en) | Yield prediction time window determining method, device, equipment and readable storage medium | |
| Ambildhuke et al. | IoT based Portable Weather Station for Irrigation Management using Real-Time Parameters | |
| CN112800845A (en) | Leaf age identification method and device, electronic equipment and readable storage medium | |
| CN112036313A (en) | Tobacco planting area detection method, device and equipment and readable storage medium | |
| Singla et al. | Spatiotemporal analysis of LANDSAT Data for Crop Yield Prediction. | |
| CN116399401A (en) | Agricultural planting system and method based on artificial intelligence | |
| CN114663652A (en) | Image processing method, image processing apparatus, management system, electronic device, and storage medium | |
| CN117236519B (en) | Water and fertilizer regulation and control method and device, electronic equipment and storage medium | |
| CN114358630A (en) | Crop cultivation method and device, electronic equipment and storage medium | |
| CN115660236B (en) | Crop weather period prediction method, device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |