Method, device and equipment for establishing plant nitrogen content monitoring model

Abstract

The invention provides a method, a device and equipment for establishing a plant nitrogen content monitoring model, and relates to the technical field of agriculture, the method comprises the following steps: obtaining a target image shot by a first spectrum camera and a ground grey board image shot by a second spectrum camera; the first spectrum camera is carried on the flight equipment; the target image is an image of a plant area; under the condition that the shooting time of the target image and the shooting time of the ground grey board image are successfully matched, irradiance data and radiance data are determined according to the target image and the ground grey board image; determining multi-band reflectivity and a plurality of vegetation indexes according to the irradiance data and the radiance data; and establishing a plant nitrogen content monitoring model based on a random forest regression model according to the plurality of vegetation indexes and the nitrogen content sampling values of the plants. According to the method provided by the embodiment of the invention, the splicing process is reduced, the computer resource demand is reduced, and based on a real-time radiometric calibration mode, higher precision can be obtained, and the monitoring performance of the nitrogen content is improved.

Description

Technical Field

[0001] The present invention relates to the field of agricultural technology, and in particular to a method, device and equipment for establishing a plant nitrogen content monitoring model. Background Art

[0002] Nitrogen is one of the nutrients for corn growth and development. The abundance or deficiency of nitrogen directly affects the growth and development process of corn, the quality of grains and the final yield. Accurate monitoring of the nitrogen nutritional status of corn helps to make accurate decisions on variable application of nitrogen fertilizer in the field, which is of great significance to improving the utilization rate of nitrogen fertilizer and reducing farmland environmental pollution.

[0003] In the related art, corn nitrogen detection is usually carried out by field sampling, laboratory chemical analysis and other methods, which is time-consuming, labor-intensive and has a lag. Therefore, how to efficiently and accurately monitor nitrogen content is a technical problem that technicians in this field need to solve urgently. Summary of the invention

[0004] The present invention provides a method, device and equipment for establishing a plant nitrogen content monitoring model, which achieves the effect of reducing the splicing process and lowering the demand for computer resources. Moreover, based on the real-time radiation calibration method, higher accuracy can be obtained, effectively improving the monitoring performance of plant nitrogen content.

[0005] The invention provides a method for establishing a plant nitrogen content monitoring model, comprising the following steps.

[0006] Acquire a target image taken by a first spectral camera and a ground gray plate image taken by a second spectral camera; the first spectral camera is mounted on a flying device; the target image is an image of the area where the plant is located; In the case where the shooting time of the target image and the shooting time of the ground gray plate image match successfully, determining the irradiance data and the radiance data according to the target image and the ground gray plate image; Determining multi-band reflectance and multiple vegetation indices based on the irradiance data and the radiance data; According to the multiple vegetation indices and the nitrogen content sampling values ​​of the plants, a plant nitrogen content monitoring model is established based on a random forest regression model.

[0007] According to a method for establishing a plant nitrogen content monitoring model provided by the present invention, when the time difference between the shooting time of the target image and the shooting time of the ground gray board image is less than a preset value, it is determined that the shooting time of the target image and the shooting time of the ground gray board image match successfully.

[0008] According to a method for establishing a plant nitrogen content monitoring model provided by the present invention, the radiance data is determined based on the following method:

[0009] Wherein, L represents the spectral radiance after radiation correction; is the number of pixel rows and columns in the ground gray plate image; is the preset coefficient; is the light sensitivity coefficient of the sensor; g is the sensor gain; is the brightness of the pixel; pBL is the background brightness of the sensor when the shutter is closed; Indicates time; , It is the line period correction factor of the electronic curtain shutter.

[0010] According to a method for establishing a plant nitrogen content monitoring model provided by the present invention, the multi-band reflectance and multiple vegetation indices are determined according to the irradiance data and the radiance data, including: Calculating the quotient of the radiance data and the irradiance data to obtain the multi-band reflectivity; A plurality of vegetation indices are determined according to the multi-band reflectance.

[0011] According to a method for establishing a plant nitrogen content monitoring model provided by the present invention, the method of determining a plurality of vegetation indices according to the multi-band reflectivity includes: Determining multiple types of vegetation indices according to the multi-band reflectance; According to the correlation between the multiple types of vegetation indices and the nitrogen content, multiple vegetation indices for establishing a plant nitrogen content monitoring model are determined from the multiple types of vegetation indices.

[0012] The present invention also provides a method for monitoring plant nitrogen content, comprising: Acquire a target image taken by a first spectral camera and a ground gray plate image taken by a second spectral camera; the first spectral camera is mounted on a flying device; the target image is an image of an area where plants are located; the time difference between the shooting time of the target image and the shooting time of the ground gray plate image is less than a preset value; Determining irradiance data and radiance data according to the target image and the ground gray plate image; Determining multi-band reflectance and multiple vegetation indices based on the irradiance data and the radiance data; The multi-band reflectance and the multiple vegetation indices are input into the plant nitrogen content monitoring model, and the nitrogen content monitoring result of the plant is output.

[0013] The present invention also provides a device for establishing a plant nitrogen content monitoring model, comprising: An acquisition module is used to acquire a target image taken by a first spectral camera and a ground gray plate image taken by a second spectral camera; the first spectral camera is mounted on the flight device; the target image is an image of the area where the plant is located; A first determination module is used to determine the irradiance data and the radiance data according to the target image and the ground gray plate image when the shooting time of the target image and the shooting time of the ground gray plate image successfully match; A second determination module is used to determine a multi-band reflectance and a plurality of vegetation indices according to the irradiance data and the radiance data; A module is established to establish a plant nitrogen content monitoring model based on a random forest regression model according to the multiple vegetation indices and the nitrogen content sampling values ​​of the plants.

[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method for establishing a plant nitrogen content monitoring model or the method for monitoring plant nitrogen content as described above is implemented.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for establishing a plant nitrogen content monitoring model or the method for monitoring plant nitrogen content as described above is implemented.

[0016] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-mentioned methods for establishing a plant nitrogen content monitoring model or a plant nitrogen content monitoring method.

[0017] The method, device and equipment for establishing a plant nitrogen content monitoring model provided by the present invention, when the shooting time of the target image and the shooting time of the ground gray plate image are successfully matched, determine the multi-band reflectivity and establish a plant nitrogen content monitoring model according to the irradiance data and the radiance data. That is, based on single image processing and real-time radiation calibration, the reflectance value is quickly extracted and the plant nitrogen content monitoring model is constructed, which not only avoids the problem that the splicing of multispectral images takes a long time and requires more computing resources in the existing method based on the calibration plate shot before and after the flight, but also effectively avoids the problem of poor accuracy of the monitoring results caused by the use of fixed radiance data, achieving the effect of reducing the splicing process and reducing the demand for computer resources, and based on the real-time radiation calibration method, higher accuracy can be obtained, effectively improving the monitoring performance of plant nitrogen content. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 This is one of the flow diagrams of the method for establishing a plant nitrogen content monitoring model provided by the present invention.

[0020] Figure 2 This is the second flow chart of the method for establishing a plant nitrogen content monitoring model provided by the present invention.

[0021] Figure 3 It is one of the schematic diagrams of the plant nitrogen content monitoring results provided by the present invention.

[0022] Figure 4 This is the second schematic diagram of the plant nitrogen content monitoring result provided by the present invention.

[0023] Figure 5 It is a schematic diagram of a device for establishing a plant nitrogen content monitoring model provided by the present invention.

[0024] Figure 6 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] Combine the following Figure 1-Figure 6 The invention describes the method, device and equipment for establishing a plant nitrogen content monitoring model.

[0027] In order to facilitate a clearer understanding of the technical solutions of the embodiments of the present application, some technical contents related to the embodiments of the present application are first introduced.

[0028] Conventional corn nitrogen detection usually adopts field sampling, laboratory chemical analysis and other methods, which is time-consuming, labor-intensive and has a lag. With the rapid development of remote sensing technology and drone technology, drone remote sensing monitoring of corn nitrogen nutrition status has gradually become a research hotspot in recent years. Using drones equipped with multispectral cameras to monitor nitrogen nutrition is one of the main methods at present. In the multispectral drone remote sensing monitoring of nitrogen process, the multispectral camera equipped with a drone is used to acquire image data of the target area, and then the acquired images are spliced ​​by image geometric correction using stitching software, and then the reflectance value is extracted after correction and normalization. The vegetation index is calculated based on the reflectance, and finally the nitrogen monitoring model is established using the vegetation index.

[0029] That is, the current mainstream multispectral drone remote sensing method for monitoring nitrogen requires processing of multispectral images. The stitching of multispectral images during the processing process often takes a lot of time and has high requirements for computer workstations. The areas where image data are collected are remote, and the original images can only be saved on the hard disk and brought back manually or transmitted to the workstation via the network. In order to ensure high-precision data, the stitching of multispectral images takes a long time. The more images are stitched and the higher the accuracy, the longer it takes. In other words, the existing method of stitching takes a long time and requires more computing resources, which is not conducive to the rapid extraction of data to establish a nitrogen monitoring model, affecting variable precision fertilization decisions and adversely affecting production. In addition, the method of shooting the calibration plate before and after the flight is easily affected by changes in light intensity, resulting in poor model accuracy.

[0030] Figure 1 This is one of the flow charts of the method for establishing a plant nitrogen content monitoring model provided by the present invention, such as Figure 1 As shown, the method includes the following: Step 101: Acquire a target image taken by a first spectral camera and a ground gray plate image taken by a second spectral camera; the first spectral camera is mounted on a flying device; the target image is an image of an area where plants are located.

[0031] Specifically, in the embodiment of the present application, the target image taken by the first spectral camera and the ground gray plate image taken by the second spectral camera are first obtained; wherein, the first spectral camera is mounted on the flying device; the target image is the image of the area where the plant is located. Optionally, a multi-spectral camera mounted on an unmanned aerial vehicle can be used to obtain the image of the target area, and at the same time, a multi-spectral device of the same model is set up on the ground, about 1m above the ground, to synchronously obtain the ground radiation correction gray plate image. Optionally, the two multi-spectral cameras have the same interval for taking pictures. The plant in the embodiment of the present application can be corn or other plants, and there is no specific limitation in the embodiment of the present application.

[0032] Step 102: When the shooting time of the target image and the shooting time of the ground gray plate image match successfully, determine the irradiance data and the radiance data according to the target image and the ground gray plate image.

[0033] Specifically, when the shooting time of the target image and the shooting time of the ground gray plate image match successfully, the irradiance data and the radiance data are determined according to the target image and the ground gray plate image, and real-time gray plate radiation calibration can be realized, which avoids the problem that the splicing of multispectral images takes a long time and requires more computing resources in the existing method based on shooting calibration plates before and after flight, and effectively avoids the problem of poor accuracy of monitoring results caused by using fixed radiance data, thereby achieving the effect of reducing the splicing process and reducing computer resource requirements. Moreover, based on the method of real-time radiation calibration of the present application, the nitrogen content of plants can be predicted more accurately compared with the existing method based on shooting calibration plates before and after flight. Optionally, the shooting time of the target image and the shooting time of the ground gray plate image match successfully, that is, the shooting time of the target image and the shooting time of the ground gray plate image are shot at the same time or a similar time, so that real-time gray plate radiation calibration can be realized, effectively avoiding the problem of poor accuracy of monitoring results caused by using fixed radiance data.

[0034] Step 103: Determine multi-band reflectance and multiple vegetation indices according to the irradiance data and the radiance data.

[0035] Specifically, when the shooting time of the target image and the shooting time of the ground gray plate image match successfully, the embodiment of the present application extracts irradiance data according to the target image and extracts radiance data according to the ground gray plate image, and after the real-time gray plate radiation calibration is realized, the multi-band reflectance can be quickly and accurately determined based on the irradiance data and radiance data, and then the vegetation index can be accurately determined based on the multi-band reflectance. Optionally, the vegetation index includes the normalized vegetation index NDVI and the nonlinear vegetation index NLI, as well as other types of vegetation indices.

[0036] Step 104: Establish a plant nitrogen content monitoring model based on a random forest regression model according to multiple vegetation indices and plant nitrogen content sampling values.

[0037] Specifically, after determining the multi-band reflectance and multiple vegetation indices, a plant nitrogen content monitoring model can be established based on the random forest regression model according to the multiple vegetation indices and the plant nitrogen content sampling values.

[0038] The method of the above embodiment, when the shooting time of the target image and the shooting time of the ground gray plate image are successfully matched, determines the multi-band reflectivity and establishes a plant nitrogen content monitoring model based on the irradiance data and the radiance data. That is, based on single image processing and real-time radiation calibration, the reflectance value is quickly extracted and the plant nitrogen content monitoring model is constructed, which not only avoids the problem that the splicing of multispectral images takes a long time and requires more computing resources in the existing method based on the calibration plate shot before and after the flight, but also effectively avoids the problem of poor accuracy of monitoring results caused by the use of fixed radiance data, achieving the effect of reducing the splicing process and reducing the demand for computer resources, and based on the real-time radiation calibration method, higher accuracy can be obtained, effectively improving the monitoring performance of plant nitrogen content.

[0039] In one embodiment, when the time difference between the shooting time of the target image and the shooting time of the ground gray board image is less than a preset value, it is determined that the shooting time of the target image and the shooting time of the ground gray board image match successfully.

[0040] Specifically, when the time difference between the shooting time of the target image and the shooting time of the ground gray card image is less than a preset value, it can be determined that the shooting time of the target image and the shooting time of the ground gray card image are taken at the same time or at a similar time, and then real-time radiation calibration can be achieved based on the target image and the ground gray card image, which effectively reduces the stitching process, reduces the requirements for computer resources, and improves the monitoring performance of plant nitrogen content.

[0041] For example, the irradiance value of the target image is extracted one by one in chronological order, the radiance value of the real-time corrected gray plate image is extracted, the corresponding gray plate radiance value is matched according to the shooting time proposed by the target image, and the 5-band reflectance value is calculated according to the reflectance value calculation formula.

[0042] The method of the above embodiment, when the time difference between the shooting time of the target image and the shooting time of the ground gray card image is less than a preset value, can realize real-time radiation calibration based on the target image and the ground gray card image, effectively reducing the stitching process and improving the monitoring performance of plant nitrogen content.

[0043] In one embodiment, the radiance data is determined based on the following:

[0044] Wherein, L represents the spectral radiance after radiation correction; is the number of pixel rows and columns in the ground gray plate image; is the preset coefficient; is the light sensitivity coefficient of the sensor; g is the sensor gain; is the brightness of the pixel; pBL is the background brightness of the sensor when the shutter is closed; Indicates time; , It is the line period correction factor of the electronic curtain shutter.

[0045] Specifically, in the embodiment of the present application, the radiance data is determined based on the following method:

[0046] Where (x, y) is the number of columns and rows of pixels on the image; L is the spectral radiation brightness after radiation correction; p is the brightness of the pixel; pBL is the background brightness of the sensor when the shutter is closed, and the value can be normalized according to the grayscale sampling depth (12bit or 16bit); g is the sensor gain, and the value is the camera ISO value / 100; a1 is the light sensitivity coefficient of the sensor; a2 and a3 are the line period correction coefficients of the electronic curtain shutter.

[0047]

[0048] Where r is the distance from the pixel to the center of the image , the polynomial coefficient ki (i = 0, ..., 5) is the image correction factor, which is recorded in the metadata tag. Optionally, the irradiance data can be determined based on an existing method, which will not be described in detail in the embodiments of the present application.

[0049] In the above embodiment, when the shooting time of the target image and the shooting time of the ground gray plate image are successfully matched, real-time radiation calibration can be achieved by determining the radiance data, which effectively reduces the stitching process and improves the monitoring performance of plant nitrogen content.

[0050] In one embodiment, determining multi-band reflectance and multiple vegetation indices based on irradiance data and radiance data includes: Calculate the quotient of the radiance data and the irradiance data to obtain the multi-band reflectance; Multiple vegetation indices are determined based on multi-band reflectance.

[0051] Specifically, in the embodiment of the present application, the multi-band reflectivity is determined based on the following method:

[0052] in, Indicates the band, Represents the pixel reflectance value in each band reflectance image, Indicates the radiance corresponding to each band (unit: w / (sr m2))、 Represents the irradiance of each band of the multispectral image of the plant canopy taken by drone within the target irradiance range during each growth period (in w / m2).

[0053] In one embodiment, multiple vegetation indices are determined based on the multi-band reflectance, including: Determine multiple types of vegetation indices based on multi-band reflectance; According to the correlation between the multiple types of vegetation indices and the nitrogen content, multiple vegetation indices for establishing a plant nitrogen content monitoring model are determined from the multiple types of vegetation indices.

[0054] Specifically, in order to improve the accuracy and performance of the plant nitrogen content monitoring model, in the embodiment of the present application, multiple types of vegetation indices are first determined based on the multi-band reflectivity; optionally, the multiple types of vegetation indices include the normalized vegetation index NDVI and the nonlinear vegetation index NLI, and may also include other types of vegetation indices, which are not specifically limited in the embodiment of the present application. Furthermore, in the embodiment of the present application, based on the correlation between the multiple types of vegetation indices and the nitrogen content, multiple vegetation indices for establishing the plant nitrogen content monitoring model are determined from the multiple types of vegetation indices, so that the plant nitrogen content monitoring model finally established can monitor the plant nitrogen content more accurately and improve the monitoring performance of the plant nitrogen content.

[0055] The method of the above embodiment determines multiple vegetation indices for establishing a plant nitrogen content monitoring model from multiple types of vegetation indices based on the correlation between multiple types of vegetation indices and nitrogen content, which can effectively improve the accuracy and performance of the plant nitrogen content monitoring model, so that the plant nitrogen content monitoring model finally established can monitor the plant nitrogen content more accurately.

[0056] The present application also provides a method for monitoring plant nitrogen content, comprising: A target image captured by a first spectral camera and a ground gray plate image captured by a second spectral camera are acquired; the first spectral camera is mounted on the flight device; the target image is an image of the area where the plant is located; the time difference between the shooting time of the target image and the shooting time of the ground gray plate image is less than a preset value; Determine the irradiance data and radiance data according to the target image and the ground gray plate image; Determine multi-band reflectance and multiple vegetation indices based on irradiance data and radiance data; Multi-band reflectance and multiple vegetation indices are input into the plant nitrogen content monitoring model to output the plant nitrogen content monitoring results.

[0057] Specifically, after establishing a plant nitrogen content monitoring model based on single image processing and real-time radiation calibration, the established plant nitrogen content monitoring model can also be used to quickly and accurately monitor the plant nitrogen content. For example, a target image taken by a first spectral camera and a ground gray plate image taken by a second spectral camera can be obtained; wherein the time difference between the shooting time of the target image and the shooting time of the ground gray plate image is less than a preset value. Then, based on the target image and the ground gray plate image, the irradiance data and the radiance data are determined, and based on the irradiance data and the radiance data, the multi-band reflectance and multiple vegetation indices are determined. Finally, the multi-band reflectance and multiple vegetation indices are input into the established plant nitrogen content monitoring model, and the plant nitrogen content monitoring results can be quickly and accurately output.

[0058] After the method of the above embodiment establishes a plant nitrogen content monitoring model based on single image processing and real-time radiation calibration, the established plant nitrogen content monitoring model can also be used to quickly and accurately monitor the plant nitrogen content, thereby effectively improving the monitoring performance of the plant nitrogen content.

[0059] For example, Figure 2 As shown, a flow chart of a method for establishing a plant nitrogen content monitoring model is provided in an embodiment of the present application, which is specifically as follows: (1) Use a multispectral camera mounted on an unmanned aerial vehicle to obtain images of the target area. At the same time, set up a multispectral device of the same model on the ground, about 1 m above the ground, to synchronously obtain ground radiation correction gray plate images; the two multispectral cameras take pictures at the same interval.

[0060] (2) Perform radiation correction on the original image. Extract the irradiance value of the target image one by one in chronological order, extract the radiance value of the real-time corrected gray plate image, match the corresponding gray plate radiance value according to the shooting time extracted from the target image, and calculate the reflectance value of the five bands according to the reflectance value calculation formula.

[0061] (3) The relevant vegetation index is calculated based on the reflectance values ​​of the five multispectral bands, and a monitoring model is constructed through correlation analysis.

[0062]

[0063] Where NDVI stands for Normalized difference vegetation index, SIPI stands for Structure Insensitive Pigment Index, PPR stands for Plant Pigment ratio, MTCI stands for MERIS Terrestrial Chlorophyll Index, MSR stands for Modified simple ratio, and NLI stands for Nonlinear vegetation index. Indicates the reflectance value of the near-red band, Indicates the red edge band reflectivity value, Represents the red band reflectance value, Represents the reflectance value of the blue band, Indicates the green band reflectance value.

[0064] Optionally, according to the correlation between multiple types of vegetation indices and nitrogen content, multiple vegetation indices for establishing a plant nitrogen content monitoring model are determined from multiple types of vegetation indices. The correlation coefficient calculation formula is as follows:

[0065] in, and represent the observed values ​​of two variables, and Represent the average values ​​of two variables, n is the sample size, and ∑ means the sum of all observations.

[0066] Optionally, a correlation coefficient table of plant nitrogen accumulation and vegetation index is shown in Table 1. A correlation coefficient table of leaf nitrogen accumulation and vegetation index is shown in Table 2.

[0067] Table 1

[0068] Table 2

[0069] Optionally, a plant nitrogen content monitoring model can be established based on a random forest regression model according to multiple vegetation indices and plant nitrogen content sampling values. The plant nitrogen content sampling values ​​are obtained by: Through destructive sampling in the field, all parts of the corn plant are dried using drying equipment, and the weight of each organ is weighed. The total nitrogen content of each organ is obtained through the Kjeldahl method in the laboratory. The nitrogen content of each organ and other related physiological indicators are calculated based on the total nitrogen content and the weight of each organ.

[0070] The calculation formula for plant nitrogen accumulation is as follows:

[0071] PNA means the nitrogen absorption of plants is the total nitrogen absorbed by plants per unit area. Indicates the weight of an organ per unit area. represents the percentage content of each organ, and i represents the name of each organ of the plant.

[0072] The calculation formula of leaf nitrogen accumulation is as follows:

[0073] LNA means leaf nitrogen accumulation, which is the total nitrogen absorbed by leaves per unit area. Indicates the weight per unit area of ​​the blade. Indicates the percentage of the whole leaf.

[0074] For example, this application uses a random forest algorithm to construct a monitoring model for nitrogen accumulation in corn plants based on single image processing and real-time radiation calibration, which reduces the splicing process and the requirements for computer resources. Through this method, an ordinary high-configuration laptop can complete the extraction and calculation of reflectivity, which meets the requirements of field sampling and the monitoring of reflectivity value and nitrogen content. Figure 3 It can be seen that the model has high accuracy, with a determination coefficient R2 of 0.74; the leaf nitrogen accumulation monitoring model constructed by Figure 4 It can be seen that the model determination coefficient R2 is 0.74. Optionally, the calculation formula of the determination coefficient R2 is:

[0075] Where n is the number of samples, is the observed value, is the predicted value of the random forest regression model, is the average observed value.

[0076] In summary, the method implemented in the present application, on the one hand, reduces the requirements for computing resources for image stitching, and on the other hand, can also effectively improve the accuracy of corn nitrogen monitoring.

[0077] The following is a description of the device for establishing a plant nitrogen content monitoring model provided by the present invention. The device for establishing a plant nitrogen content monitoring model described below and the method for establishing a plant nitrogen content monitoring model described above can be referred to in correspondence with each other. Figure 5 As shown, including: The acquisition module 510 is used to acquire a target image taken by a first spectral camera and a ground gray plate image taken by a second spectral camera; the first spectral camera is mounted on the flying device; the target image is an image of the area where the plant is located; A first determination module 520 is used to determine irradiance data and radiance data according to the target image and the ground gray plate image when the shooting time of the target image and the shooting time of the ground gray plate image match successfully; A second determination module 530 is used to determine a multi-band reflectance and a plurality of vegetation indices according to the irradiance data and the radiance data; The module 540 is established to establish a plant nitrogen content monitoring model based on a random forest regression model according to multiple vegetation indices and plant nitrogen content sampling values.

[0078] Optionally, the first determining module 520 is further configured to: When the time difference between the shooting time of the target image and the shooting time of the ground gray board image is less than a preset value, it is determined that the shooting time of the target image and the shooting time of the ground gray board image match successfully.

[0079] Optionally, the second determining module 530 is further configured to: The radiance data is determined based on:

[0080] Wherein, L represents the spectral radiance after radiation correction; is the number of pixel rows and columns in the ground gray plate image; is the preset coefficient; is the light sensitivity coefficient of the sensor; g is the sensor gain; is the brightness of the pixel; pBL is the background brightness of the sensor when the shutter is closed; Indicates time; , It is the line period correction factor of the electronic curtain shutter.

[0081] Optionally, the second determining module 530 is specifically configured to: Calculate the quotient of the radiance data and the irradiance data to obtain the multi-band reflectance; Multiple vegetation indices are determined based on multi-band reflectance.

[0082] Optionally, the second determining module 530 is specifically configured to: Determine multiple types of vegetation indices based on multi-band reflectance; According to the correlation between the multiple types of vegetation indices and the nitrogen content, multiple vegetation indices for establishing a plant nitrogen content monitoring model are determined from the multiple types of vegetation indices.

[0083] Figure 6 An example of a physical structure diagram of an electronic device is provided, and the electronic device may include: a processor 610, a communication interface 620, a memory 630, and a communication bus 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other through the communication bus 640. The processor 610 may call the logic instructions in the memory 630 to execute a method for establishing a plant nitrogen content monitoring model, the method comprising: obtaining a target image taken by a first spectral camera and a ground gray plate image taken by a second spectral camera; the first spectral camera is mounted on the flight device; the target image is an image of the area where the plant is located; when the shooting time of the target image and the shooting time of the ground gray plate image are successfully matched, determining irradiance data and radiance data according to the target image and the ground gray plate image; determining multi-band reflectance and multiple vegetation indices according to the irradiance data and the radiance data; and establishing a plant nitrogen content monitoring model based on a random forest regression model according to multiple vegetation indices and the nitrogen content sampling values ​​of the plant.

[0084] In addition, the logic instructions in the above-mentioned memory 630 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0085] On the other hand, the present invention also provides a computer program product, which includes a computer program, and the computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the method for establishing a plant nitrogen content monitoring model provided by the above-mentioned methods, and the method includes: obtaining a target image taken by a first spectral camera and a ground gray board image taken by a second spectral camera; the first spectral camera is mounted on a flying device; the target image is an image of the area where the plant is located; when the shooting time of the target image and the shooting time of the ground gray board image match successfully, determining irradiance data and radiance data according to the target image and the ground gray board image; determining multi-band reflectance and multiple vegetation indices according to the irradiance data and the radiance data; and establishing a plant nitrogen content monitoring model based on a random forest regression model according to multiple vegetation indices and plant nitrogen content sampling values.

[0086] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the method for establishing a plant nitrogen content monitoring model provided by the above-mentioned methods, the method comprising: acquiring a target image taken by a first spectral camera and a ground gray plate image taken by a second spectral camera; the first spectral camera is mounted on a flying device; the target image is an image of the area where the plants are located; when the shooting time of the target image and the shooting time of the ground gray plate image match successfully, determining irradiance data and radiance data based on the target image and the ground gray plate image; determining multi-band reflectance and multiple vegetation indices based on the irradiance data and the radiance data; and establishing a plant nitrogen content monitoring model based on a random forest regression model based on multiple vegetation indices and the nitrogen content sampling values ​​of the plants.

[0087] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0088] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0089] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for establishing a plant nitrogen content monitoring model, characterized in that: include: Acquire a target image taken by a first spectral camera and a ground gray plate image taken by a second spectral camera; The first spectral camera is carried on the flight device; The target image is an image of the area where the plant is located; In the case where the shooting time of the target image and the shooting time of the ground gray plate image match successfully, determining the irradiance data and the radiance data according to the target image and the ground gray plate image; Determining multi-band reflectance and multiple vegetation indices based on the irradiance data and the radiance data; According to the multiple vegetation indices and the nitrogen content sampling values ​​of the plants, a plant nitrogen content monitoring model is established based on a random forest regression model.

2. The method for establishing a plant nitrogen content monitoring model according to claim 1, characterized in that: The method further comprises: When the time difference between the shooting time of the target image and the shooting time of the ground gray board image is less than a preset value, it is determined that the shooting time of the target image and the shooting time of the ground gray board image are successfully matched.

3. The method for establishing a plant nitrogen content monitoring model according to claim 1, characterized in that: The method further comprises: The radiance data is determined based on the following method: Wherein, L represents the spectral radiance after radiation correction; is the number of pixel rows and columns in the ground gray plate image; is the preset coefficient; is the light sensitivity coefficient of the sensor; g is the sensor gain; is the brightness of the pixel; pBL is the background brightness of the sensor when the shutter is closed; Indicates time; , It is the line period correction factor of the electronic curtain shutter.

4. The method for establishing a plant nitrogen content monitoring model according to any one of claims 1 to 3, characterized in that: Determining multi-band reflectance and multiple vegetation indices according to the irradiance data and the radiance data includes: Calculating the quotient of the radiance data and the irradiance data to obtain multi-band reflectivity; A plurality of vegetation indices are determined according to the multi-band reflectance.

5. The method for establishing a plant nitrogen content monitoring model according to claim 4, characterized in that: Determining a plurality of vegetation indices according to the multi-band reflectance includes: Determining multiple types of vegetation indices according to the multi-band reflectance; According to the correlation between the multiple types of vegetation indices and nitrogen content, multiple vegetation indices for establishing a plant nitrogen content monitoring model are determined from the multiple types of vegetation indices.

6. A method for monitoring nitrogen content in plants, characterized in that: The plant nitrogen content monitoring model as claimed in any one of claims 1 to 5 comprises: Acquire a target image taken by a first spectral camera and a ground gray plate image taken by a second spectral camera; the first spectral camera is mounted on a flying device; the target image is an image of an area where plants are located; the time difference between the shooting time of the target image and the shooting time of the ground gray plate image is less than a preset value; Determining irradiance data and radiance data according to the target image and the ground gray plate image; Determining multi-band reflectance and multiple vegetation indices based on the irradiance data and the radiance data; The multi-band reflectivity and the multiple vegetation indices are input into a plant nitrogen content monitoring model, and a plant nitrogen content monitoring result is output.

7. A device for establishing a plant nitrogen content monitoring model, characterized in that: include: An acquisition module, used to acquire a target image taken by a first spectral camera and a ground gray plate image taken by a second spectral camera; The first spectral camera is mounted on the flying device; the target image is an image of the area where the plant is located; A first determination module is used to determine the irradiance data and the radiance data according to the target image and the ground gray plate image when the shooting time of the target image and the shooting time of the ground gray plate image successfully match; A second determination module is used to determine a multi-band reflectance and a plurality of vegetation indices according to the irradiance data and the radiance data; A module is established to establish a plant nitrogen content monitoring model based on a random forest regression model according to the multiple vegetation indices and the nitrogen content sampling values ​​of the plants.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for establishing a plant nitrogen content monitoring model as described in any one of claims 1 to 5 or the method for monitoring plant nitrogen content as described in claim 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for establishing a plant nitrogen content monitoring model according to any one of claims 1 to 5 or the method for monitoring plant nitrogen content according to claim 6 is implemented.

Images