Crop fertilizer decision-making method, device and storage medium based on multispectral image

Abstract

The present invention provides a method, device and storage medium for crop fertilizer decision-making based on multispectral images. The method comprises: receiving an image to be processed sent by a multispectral drone via a local area network; the image to be processed is a multi-band image of a field collected by the multispectral drone; determining NDVI according to the near-infrared band reflectivity and red band reflectivity of the image to be processed, and determining RECI according to the near-infrared band reflectivity and red-edge band reflectivity of the image to be processed; performing growth classification on the crops in the image to be processed based on RECI and NDVI to obtain multiple regions with independent boundaries; determining the fertilizer requirement of the centroid point of each region, and obtaining a fertilization prescription map based on the fertilizer requirement of the centroid points of all regions. The method, device and storage medium for crop fertilizer decision-making based on multispectral images provided by the present invention realize the automatic acquisition of multispectral images, and realize the rapid determination of fertilizer deficiency information of the entire field through pixel-level fertilizer deficiency analysis, thereby improving the efficiency of crop fertilizer decision-making.

Description

Technical Field

[0001] The present invention relates to the technical field of crop fertilization, and in particular to a crop fertilizer decision-making method, device and storage medium based on multispectral images. Background Art

[0002] Improving the level of food production management is an important way to ensure food security, and water and fertilizer are the most critical production factors in the growth of food crops. The traditional model of fertilization and irrigation based on experience can guarantee food production to a certain extent, but there are problems with extensive fertilization and pesticide application. Excessive fertilization not only wastes resources, but also causes soil compaction and poor air permeability, thus hindering the growth of crops. Therefore, there is an urgent need for a means to accurately sense the health status of crops and prevent and control them in a timely manner to improve production efficiency and reduce production costs.

[0003] Currently, spectral-based crop growth sensing technology can be used to monitor the growth status of crops. Data obtained by drone digital, multispectral, hyperspectral, thermal infrared or radar sensors can improve the accuracy of crop nutrient and structure diagnosis.

[0004] However, existing methods for crop nutrient diagnosis / crop fertilizer decision-making based on spectral sensors still have the problems of consuming a lot of manpower and computing resources and having low diagnostic efficiency. Summary of the invention

[0005] The present invention provides a crop fertilizer decision-making method and device based on multispectral images, which are used to solve the technical problem of low efficiency of crop nutrient diagnosis in the prior art.

[0006] In a first aspect, the present invention provides a crop fertilizer decision-making method based on multispectral images, comprising the following steps.

[0007] Receiving an image to be processed sent by a multispectral UAV through a local area network; the image to be processed is a multispectral image of a field collected by the multispectral UAV; Determine the Normalized Difference Vegetation Index (NDVI) according to the near-infrared band reflectance and the red band reflectance of the image to be processed, and determine the Red Edge Chlorophyll Vegetation Index (RECI) according to the near-infrared band reflectance and the red edge band reflectance of the image to be processed; Based on the RECI and the NDVI, the crops in the image to be processed are graded for growth potential, so as to obtain a plurality of regions with independent boundaries, and the crops in each region are of the same growth potential grade; The fertilizer requirement of the centroid point of each area is determined, and a fertilizer prescription map is obtained based on the fertilizer requirement of the centroid points of all areas.

[0008] In some embodiments, determining the fertilizer requirement of the centroid of each area includes: Based on the NDVI of the centroid of each area, the fertilizer requirement of the centroid is determined according to the nitrogen fertilizer optimization algorithm.

[0009] In some embodiments, obtaining a fertilization prescription map based on the fertilizer requirement of the centroid of all regions includes: Convert the pixel coordinates of the centroid of each area into longitude and latitude coordinates in the geocentric coordinate system; Based on the fertilizer requirement of the centroid of each area and the latitude and longitude coordinates, using Kriging interpolation method to obtain the crop fertilizer deficiency information of the field; A fertilization prescription map of target resolution is obtained according to the crop fertilizer deficiency information.

[0010] In some embodiments, the step of grading the growth potential of crops in the image to be processed based on the RECI and the NDVI includes: Eliminating bare soil in the image to be processed based on the RECI; Based on the NDVI, the crops in the image to be processed after the bare soil is removed are graded for their growth potential.

[0011] In some embodiments, the method further comprises: Dark current correction and radiation correction are performed on the brightness values ​​of the remote sensing image pixels in different bands of the image to be processed to obtain the reflectivity of the image in different bands in the image to be processed.

[0012] In some embodiments, the method further comprises: According to different flight speeds and image shooting intervals, unjoined images to be processed are extracted at equal time intervals from the received images to be processed; Obtaining the shooting posture and positioning information of the extracted image to be processed when it was shot; Based on the shooting posture and the positioning information, the to-be-processed images of different bands are corrected and registered.

[0013] In some embodiments, the multi-spectral drone replaces batteries via a robotic arm of the drone airport.

[0014] In a second aspect, the present invention provides a crop fertilizer decision-making device based on multispectral images, comprising the following modules: A receiving module, used for receiving an image to be processed sent by a multispectral UAV through a local area network; the image to be processed is a multispectral image of a field collected by the multispectral UAV; A determination module, used to determine the normalized vegetation index NDVI according to the near-infrared band reflectance and the red band reflectance of the image to be processed, and to determine the red-edge chlorophyll vegetation index RECI according to the near-infrared band reflectance and the red-edge band reflectance of the image to be processed; A first acquisition module is used to grade the growth potential of the crops in the image to be processed based on the RECI and the NDVI to obtain a plurality of regions with independent boundaries, and the crops in each region are of the same growth grade; The second acquisition module is used to determine the fertilizer requirement of the centroid point of each area, and obtain a fertilization prescription map based on the fertilizer requirement of the centroid points of all areas.

[0015] In a third aspect, the present invention 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 crop fertilizer decision-making method based on multispectral images as described in the first aspect above is implemented.

[0016] In a fourth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the crop fertilizer decision-making method based on multispectral images as described in the first aspect above.

[0017] In a fifth aspect, the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the crop fertilizer decision-making method based on multispectral images as described in the first aspect above.

[0018] The crop fertilizer decision-making method, device and storage medium based on multispectral images provided by the present invention are as follows: an edge computing module of an unmanned aerial vehicle field receives an image to be processed sent by a multispectral unmanned aerial vehicle through a local area network, wherein the image to be processed is a multi-band image of a field collected by the multispectral unmanned aerial vehicle; then, NDVI is determined according to the near-infrared band reflectivity and the red band reflectivity of the image to be processed, and RECI is determined according to the near-infrared band reflectivity and the red edge band reflectivity of the image to be processed; based on RECI and NDVI, the growth potential of crops in the image to be processed is graded to obtain a plurality of regions with independent boundaries, and the crops in each region are of the same growth potential grade; the centroid point of each region is determined and the fertilizer requirement of the point is calculated, and a fertilization prescription map is obtained based on the fertilizer requirement of all centroid points, thereby realizing automatic acquisition of multispectral images, and quickly determining the fertilizer deficiency information of the entire field through pixel-level fertilizer deficiency analysis, thereby improving the efficiency of crop fertilizer decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] 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.

[0020] Figure 1It is a flow chart of the crop fertilizer decision-making method based on multispectral images provided by the present invention.

[0021] Figure 2 It is a structural schematic diagram of the crop fertilizer decision-making system based on multispectral images provided by the present invention.

[0022] Figure 3 It is a schematic diagram of a process for generating a fertilization prescription map provided by the present invention.

[0023] Figure 4 It is a structural schematic diagram of a crop fertilizer decision-making device based on multispectral images provided by the present invention.

[0024] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0025] At present, spectral-based crop growth perception technology has made significant progress. The data obtained by drone digital, multi-spectral, hyperspectral, thermal infrared, radar and other sensors can improve the accuracy and efficiency of crop nutrient and structure diagnosis. At the same time, agricultural plant protection drones have been developed for many years, and there are already many relatively mature manufacturers and technologies. For example, the MS600 multi-spectral sensor is combined with the DJI M350RTK platform to achieve efficient acquisition of crop nutrient data, and another example is the RededgeMX multi-spectral sensor. These technologies provide strong support for the rapid perception of crop growth information, but can only obtain data, not generate accurate prescription information; DJI's P4M and Mavic 3M multispectral drones, XAG's M500 series drones and P150 series plant protection drones can all quickly acquire images and stitch them together to generate fertilization prescriptions, and then use DJI's T series plant protection drones and XAG's P150 series plant protection drones to achieve more accurate plant protection and fertilization operations. They can also achieve crop nutrient perception and precise plant protection and fertilization operations, but the technical means used are to manually copy data after the drone flight is completed to stitch together a large number of spectral images to generate orthophotos, and then use computers with strong computing power to calculate the spectral index to obtain the prescription map. The data processing process not only consumes a lot of human resources and computing resources, but also has a long data processing cycle and a low degree of automation, which is not conducive to cost savings.

[0026] The drone airport can adapt to the field working environment and realize data acquisition under unmanned duty, such as DJI Airport 1 and 2. However, its adapted sensors can only obtain digital and thermal infrared data, which is not enough for crop nutrient diagnosis. At the same time, its prescription generation process still needs to first stitch and generate orthophotos, and then generate prescription maps, which is time-consuming.

[0027] Although there are drones and sensor platforms that can obtain crop growth information, these systems have problems such as long data processing cycles, low levels of automation, and independent and scattered operation processes, and cannot meet the needs of rapid diagnosis and unattended operations.

[0028] In addition, the current crop nutrient diagnosis method relies on the splicing of multispectral data to generate prescription maps. This process requires a lot of manpower and computing resources and has a long cycle, which is not conducive to the promotion and application of the technology. Moreover, the current nutrient diagnosis and plant protection and fertilization operations are independent and cannot be fully automated. They are also difficult to operate and are not conducive to large-scale promotion and application.

[0029] Based on the above technical problems, the present invention proposes a crop fertilizer decision-making method based on multispectral images, wherein the edge computing module of the drone field receives the image to be processed sent by the multispectral drone through the local area network, and the image to be processed is a multi-band image of the field collected by the multispectral drone; then the NDVI is determined according to the near-infrared band reflectivity and the red band reflectivity of the image to be processed, and the RECI is determined according to the near-infrared band reflectivity and the red edge band reflectivity of the image to be processed; based on RECI and NDVI, the growth potential of the crops in the image to be processed is graded to obtain multiple areas with independent boundaries, and the crops in each area are of the same growth grade; the fertilizer requirement of the centroid point of each area is determined, and a fertilization prescription map is obtained based on the fertilizer requirement at all centroid points, thereby realizing the automatic acquisition and analysis of multispectral images, and quickly determining the fertilizer deficiency information of the entire field through pixel-level fertilizer deficiency analysis, thereby improving the efficiency of crop fertilizer decision-making, improving the flexibility of application, and facilitating promotion and utilization.

[0030] 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.

[0031] Figure 1 is a flow chart of the crop fertilizer decision-making method based on multispectral images provided by the present invention, such as Figure 1 As shown, the present invention provides a crop fertilizer decision method based on multispectral images, and its execution subject can be an edge computing module of a drone airport. The method includes the following steps: Step 101: receiving an image to be processed sent by a multispectral drone via a local area network; the image to be processed is a multispectral image of a field collected by the multispectral drone.

[0032] Specifically, in view of the unmanned operation mode in the field, a drone airport capable of 24-hour operation in the field is selected. The drone airport provides drone take-off and landing platform, charging, data downloading, prescription generation, wireless communication, meteorological environment monitoring, dust and rain protection and other functions.

[0033] Figure 2 Schematic diagram of the structure of the crop fertilizer decision-making system based on multispectral images provided by the present invention. Figure 2 As shown in the figure, multispectral UAV assisted takeoff and landing relies on the centimeter-level real-time dynamic carrier phase differential technology (Real-time kinematic, RTK) positioning service module on the UAV airport and the UAV, and on the other hand, uses the field of view image sensor of the UAV to achieve precise positioning. The airport charging uses a wired method to ensure stable power supply. The multispectral UAV and the UAV airport are connected through a local area network (such as a 2.4GHz local area network). The data download method is completed through the local area network built between the UAV and the airport. The prescription generation is completed by the edge computing module integrated in the airport, which integrates multiple algorithms; the wireless communication uses a 4G communication module.

[0034] The edge computing module is a miniature computer with 4G communication capability and Windows operating system installed.

[0035] The edge computing module receives the images to be processed sent by the multispectral UAV through the local area network. Specifically, the multispectral UAV collects multispectral images of the field and sends them to the UAV field through the local area network. The UAV field receives the multispectral images sent by the multispectral UAV through the local area network, and then transmits them to the edge computing module through the serial port transmission module. The edge computing module receives the multispectral images, that is, the images to be processed.

[0036] Step 102: determine the Normalized Difference Vegetation Index NDVI according to the near-infrared band reflectance and the red band reflectance of the image to be processed, and determine the Red Edge Chlorophyll Vegetation Index RECI according to the near-infrared band reflectance and the red edge band reflectance of the image to be processed.

[0037] Specifically, the multispectral camera carried by the multispectral drone contains multiple near-infrared channels, because the near-infrared channel can provide useful information for diagnosing crop nutrient information, while ordinary digital cameras and thermal infrared provide limited information. For example, the selected channels include: green (G): 560nm±16nm; red (R): 650nm±16nm; red edge (RE): 730nm±16nm; near infrared (NIR): 860nm±26nm and other four spectral channels and a 20-megapixel RGB channel.

[0038] After the edge computing module receives the image to be processed sent by the multispectral drone through the local area network, it calculates the reflectivity of the image in different bands in the image to be processed, including the near-infrared band reflectivity, the red edge band reflectivity, and the red band reflectivity.

[0039] The Normalized Difference Vegetation Index (NDVI) is calculated based on the near-infrared band reflectance and the red band reflectance in the image to be processed. The calculation formula is as follows: NDVI = (NIR-R) / (NIR+R) Among them, NDVI is the normalized difference vegetation index; NIR is the near-infrared band reflectance; R is the red band reflectance.

[0040] The Red-Edge ChlorophyII Vegetation Index (RECI) is determined based on the near-infrared band reflectance and red-edge band reflectance of the image to be processed. The calculation formula is as follows: RECI=(NIR / RE)-1 Among them, RECI is the red edge chlorophyll vegetation index; NIR is the near infrared band reflectance; RE is the red edge band reflectance.

[0041] Step 103: Based on the RECI and the NDVI, the crops in the image to be processed are graded according to their growth potential, so as to obtain a plurality of regions with independent boundaries, wherein the crops in each region are of the same growth potential grade.

[0042] Specifically, the growth condition of crops in the processed image is graded based on the obtained RECI and NDVI. For example, by setting a plurality of standard values, the crops are divided into a plurality of grades to intuitively indicate the quality of the growth condition of the crops.

[0043] After the growth potential grading is completed, the crops are divided into different areas according to the growth potential levels. The crops divided into the same area have the same growth potential level, and finally multiple areas with independent boundaries are obtained.

[0044] For example, the crops in the image to be processed are divided into three levels according to their growth potential, namely, good growth, average growth, and poor growth. Thus, the crops in the image to be processed are divided into one or more good growth level areas, one or more average growth level areas, and one or more poor growth level areas.

[0045] Step 104: determine the fertilizer requirement of the centroid of each area, and obtain a fertilizer prescription map based on the fertilizer requirement of the centroid of all areas.

[0046] Specifically, the fertilizer requirement of each area centroid is determined, and the fertilizer requirement of all area centroids is combined to further obtain a fertilizer prescription map corresponding to the entire plot.

[0047] like Figure 2 As shown, after obtaining the fertilization prescription map, the edge computing module transmits it to the 4G communication module of the drone farm. The drone farm uploads the fertilization prescription map to the cloud through the 4G communication module, and automatically synchronizes the fertilization prescription map to the plant protection fertilization drone remotely, controlling the fertilization drone to fertilize according to the prescription, thereby realizing variable precision operation.

[0048] The crop fertilizer decision-making method based on multispectral images provided in the embodiment of the present application can realize full automation of data acquisition, processing, analysis and plant protection fertilization operations in an unattended state. The edge computing module automatically generates a fertilization prescription map, eliminating the complex process of copying data in the traditional mode and calculating the prescription on a computer, thereby improving operation efficiency. When analyzing the image specifically, the fertilizer deficiency information of the entire field is quickly determined through regional fertilizer deficiency analysis, thereby improving the efficiency of crop fertilizer decision-making and increasing the flexibility of application.

[0049] In some embodiments, the method further comprises: According to different flight speeds and image shooting intervals, unjoined images to be processed are extracted at equal time intervals from the received images to be processed; Obtaining the shooting posture and positioning information of the extracted image to be processed when it was shot; Based on the shooting posture and the positioning information, the to-be-processed images of different bands are corrected and registered.

[0050] Specifically, after receiving the image to be processed sent by the multispectral UAV through the local area network, the image to be processed needs to be preprocessed, including image sampling, image correction and other processing.

[0051] Figure 3 It is a schematic diagram of the process of generating a fertilization prescription map provided by the present invention, such as Figure 3 As shown in the figure, after the edge computer obtains the original image (i.e., the image to be processed), it extracts the unstitched images at equal time intervals from the received original image according to different flight speeds and image shooting intervals. Then, the metadata information of the image is used to obtain the attitude and global positioning system (GPS) information of the drone when shooting, and the images of different bands are corrected and registered according to the resolution calculation formula provided by the camera carried by the drone to obtain the pre-processed image.

[0052] In some embodiments, the method further comprises: Perform dark current correction and radiometric correction on the brightness values of remote sensing image pixels in different bands of the image to be processed, and obtain the reflectance of the images in different bands of the image to be processed.

[0053] Specifically, perform dark current correction and radiometric correction on the brightness values (DigitalNumber, DN), i.e., spectral indices, of remote sensing image pixels in different bands of the image to be processed, convert the original DN values into reflectance, and obtain the reflectance of the images in different bands.

[0054] As Figure 3 shown, for the three-band image, including the red-band image, the red-edge band image, and the near-infrared band image, calculate the spectral indices of each image, and perform dark current correction and image radiometric correction to obtain the reflectance of the three-band image, so as to calculate RECI and NDVI based on the reflectance of the red-band, red-edge band, and near-infrared band images.

[0055] In some embodiments, the method for grading the growth of crops in the image to be processed based on the RECI and the NDVI includes: Exclude the bare soil in the image to be processed based on the RECI; Grade the growth of crops in the image to be processed after excluding the bare soil based on the NDVI.

[0056] Specifically, when grading the growth of crops in the image to be processed, first exclude the bare soil in the image to be processed based on the RECI. After excluding the bare soil, the remaining in the image to be processed are crops, and then grade the growth of the crops based on the NDVI.

[0057] For example, set the category of pixels with RECI < 0.35 as bare soil, set the growth level of pixels with NDVI > 0.85 as good, the growth level of pixels with 0.65 < NDVI < 0.85 as average, and the growth level of pixels with NDVI < 0.65 as poor. Exclude the pixels with RECI < 0.35 in the image to be processed, and the remaining pixels with RECI > 0.35 are the category of crops. Grade the growth of the pixels of the category of crops, and divide them into the above three growth levels based on the NDVI.

[0058] The method for making crop fertilizer decisions based on multi-spectral images provided by the embodiments of the present application judges whether the image pixels in the image to be processed are bare soil based on the RECI, accurately separates the crop part in the image to be processed, and judges the growth of the crops based on the NDVI to realize crop grading, so as to refine the fertilization decision.

[0059] In some embodiments, the determining the fertilizer requirement of the centroid point of each region includes: Based on the NDVI of the centroid of each area, the fertilizer requirement of the centroid is determined according to the nitrogen fertilizer optimization algorithm.

[0060] The crop fertilizer decision method based on multispectral images provided in the embodiment of the present application can use the centroid of a region as the representative point of the region to determine the fertilizer requirement of the centroid of each region on the basis of obtaining different growth grade regions by continuing to classify crops according to their growth potential, and the operation is simple and efficient. In addition, based on NDVI and the Nitrogen Fertilization Optimization Algorithm (NFOA), it can effectively diagnose the nutrient status of crops and improve the accuracy of crop fertilizer decision making.

[0061] In some embodiments, obtaining a fertilization prescription map based on the fertilizer requirement of the centroid of each area includes: Convert the pixel coordinates of the centroid of each area into longitude and latitude coordinates in the geocentric coordinate system; Based on the fertilizer requirement of the centroid of each area and the latitude and longitude coordinates, using Kriging interpolation method to obtain the crop fertilizer deficiency information of the field; A fertilization prescription map of target resolution is obtained according to the crop fertilizer deficiency information.

[0062] Specifically, Figure 3 As shown, the pixel coordinates of the centroid of each area are obtained and converted into longitude and latitude coordinates such as World Geodetic System (WGS) coordinates. Specifically, the projection information and the resolution of the drone image can be used to convert the pixel coordinates of the centroid into a geocentric coordinate system such as the WGS84 coordinate system to obtain the longitude and latitude coordinates of the centroid.

[0063] Then, based on the fertilizer requirement of each area centroid and the longitude and latitude coordinates of the centroid, the crop fertilizer deficit information in the entire field is calculated using the Kriging interpolation method.

[0064] The crop fertilizer deficit information (in matrix form) obtained by Kriging interpolation is assigned corresponding pixel coordinates and crop fertilizer application data, and exported as a tag image file (TIF) format. Finally, the exported fertilization prescription map is resampled to generate a fertilization prescription map with a target resolution (such as a resolution of about 3m) to meet the reading requirements of the plant protection and fertilization drone.

[0065] The crop fertilizer decision-making method based on multispectral images provided in the embodiment of the present application utilizes the GPS coordinates in the image and the Kriging interpolation algorithm to generate a fertilization prescription map of the target resolution. There is no need to perform a large amount of image stitching, thus avoiding the problem of poor timeliness of prescription generation caused by data stitching, significantly reducing the hardware requirements and time cost of data processing, improving work efficiency, saving computing, time and human resources, and improving the efficiency of crop fertilizer decision-making.

[0066] In some embodiments, the multi-spectral drone replaces batteries via a robotic arm of the drone airport.

[0067] Specifically, to ensure the battery replacement efficiency of the multi-spectral UAV, the UAV airport is equipped with multiple sets of batteries. After the multi-spectral UAV lands at the UAV airport, another set of batteries can be directly replaced. The battery replacement can be completed by the UAV airport robotic arm, and the UAV must have the ability to easily disassemble the battery to facilitate the robot to replace the battery.

[0068] The crop fertilizer decision-making method based on multispectral images provided in the embodiment of the present application is that the drone airport is equipped with multiple sets of batteries, and the batteries can be replaced by a robotic arm, which improves the battery replacement efficiency and ensures the timely acquisition of multispectral images, thereby ensuring the efficiency and real-time nature of crop fertilizer decision-making.

[0069] Figure 4 is a schematic diagram of the structure of the crop fertilizer decision-making device based on multispectral images provided by the present invention, such as Figure 4 As shown, the present invention provides a crop fertilizer decision-making device based on multispectral images, including a receiving module 401, a determining module 402, a first acquiring module 403 and a second acquiring module 404.

[0070] The receiving module 401 is used to receive the image to be processed sent by the multispectral UAV through the local area network; the image to be processed is the multispectral image of the field collected by the multispectral UAV; The determination module 402 is used to determine the normalized vegetation index NDVI according to the near-infrared band reflectance and the red band reflectance of the image to be processed, and determine the red-edge chlorophyll vegetation index RECI according to the near-infrared band reflectance and the red-edge band reflectance of the image to be processed; The first acquisition module 403 is used to grade the growth potential of the crops in the image to be processed based on the RECI and the NDVI to obtain a plurality of regions with independent boundaries, and the crops in each region are of the same growth level; The second acquisition module 404 is used to determine the fertilizer requirement of the centroid of each area, and acquire a fertilization prescription map based on the fertilizer requirement of the centroid of all areas.

[0071] Specifically, the above-mentioned crop fertilizer decision-making device based on multispectral images provided by the present invention can implement all the method steps implemented by the above-mentioned crop fertilizer decision-making method embodiment based on multispectral images, and can achieve the same technical effects. The parts and beneficial effects that are the same as those in the method embodiment will not be described in detail here.

[0072] It should be noted that the division of units / modules in the above-mentioned embodiments of the present invention is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, the functional units in the various embodiments of the present application may be integrated into a processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware or in the form of software functional units.

[0073] Figure 5 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 5 As shown, the electronic device may include: a processor 501, a communications interface 502, a memory 503 and a communication bus 504, wherein the processor 501, the communications interface 502 and the memory 503 communicate with each other through the communication bus 504. The processor 501 may call the logic instructions in the memory 503 to execute a crop fertilizer decision method based on multispectral images, the method comprising: Receiving an image to be processed sent by a multispectral UAV through a local area network; the image to be processed is a multispectral image of a field collected by the multispectral UAV; Determine the Normalized Difference Vegetation Index (NDVI) according to the near-infrared band reflectance and the red band reflectance of the image to be processed, and determine the Red Edge Chlorophyll Vegetation Index (RECI) according to the near-infrared band reflectance and the red edge band reflectance of the image to be processed; Based on the RECI and the NDVI, the crops in the image to be processed are graded for growth potential, so as to obtain a plurality of regions with independent boundaries, and the crops in each region are of the same growth potential grade; The fertilizer requirement of the centroid point of each area is determined, and a fertilizer prescription map is obtained based on the fertilizer requirement of the centroid points of all areas.

[0074] Specifically, the processor 501 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or a complex programmable logic device (CPLD), and the processor may also adopt a multi-core architecture.

[0075] The logic instructions in the memory 503 can be implemented in the form of software functional units and can be stored in a processor-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0076] In some embodiments, a computer program product is further provided, the computer program product comprising a computer program, the computer program can be stored on a non-transitory computer-readable storage medium, and when the computer program is executed by a processor, the computer can execute the crop fertilizer decision-making method based on multispectral images provided in the above-mentioned method embodiments, the method comprising: Receiving an image to be processed sent by a multispectral UAV through a local area network; the image to be processed is a multispectral image of a field collected by the multispectral UAV; Determine the Normalized Difference Vegetation Index (NDVI) according to the near-infrared band reflectance and the red band reflectance of the image to be processed, and determine the Red Edge Chlorophyll Vegetation Index (RECI) according to the near-infrared band reflectance and the red edge band reflectance of the image to be processed; Based on the RECI and the NDVI, the crops in the image to be processed are graded to obtain a plurality of regions with independent boundaries, and the crops in each region are of the same growth grade; The fertilizer requirement of the centroid point of each area is determined, and a fertilizer prescription map is obtained based on the fertilizer requirement of the centroid points of all areas.

[0077] Specifically, the above-mentioned computer program product provided in the embodiment of the present application can implement all the method steps implemented by the above-mentioned method embodiments, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as the method embodiment will not be described in detail here.

[0078] In some embodiments, a computer-readable storage medium is further provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program is used to enable a computer to execute the crop fertilizer decision-making method based on multispectral images provided in the above-mentioned method embodiments, the method comprising: Receiving an image to be processed sent by a multispectral UAV through a local area network; the image to be processed is a multispectral image of a field collected by the multispectral UAV; Determine the Normalized Difference Vegetation Index (NDVI) according to the near-infrared band reflectance and the red band reflectance of the image to be processed, and determine the Red Edge Chlorophyll Vegetation Index (RECI) according to the near-infrared band reflectance and the red edge band reflectance of the image to be processed; Based on the RECI and the NDVI, the crops in the image to be processed are graded to obtain a plurality of regions with independent boundaries, and the crops in each region are of the same growth grade; The fertilizer requirement of the centroid point of each area is determined, and a fertilizer prescription map is obtained based on the fertilizer requirement of the centroid points of all areas.

[0079] Specifically, the above-mentioned computer-readable storage medium provided by the present invention can implement all the method steps implemented by the above-mentioned method embodiments, and can achieve the same technical effects. The parts and beneficial effects that are the same as the method embodiments in this embodiment will not be described in detail here.

[0080] It should be noted that the computer-readable storage medium can be any available medium or data storage device that can be accessed by the processor, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO), etc.), optical storage (such as CD, DVD, BD, HVD, etc.), and semiconductor storage (such as ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)), etc.

[0081] In the present invention, "determine B based on A" means that the factor A should be considered when determining B. It is not limited to "B can be determined based on A alone", but should also include: "determine B based on A and C", "determine B based on A, C and E", "determine C based on A, and further determine B based on C", etc. In addition, it can also include taking A as a condition for determining B, for example, "when A meets the first condition, use the first method to determine B"; for another example, "when A meets the second condition, determine B", etc.; for another example, "when A meets the third condition, determine B based on the first parameter", etc. Of course, it can also be a condition that takes A as a factor for determining B, for example, "when A meets the first condition, use the first method to determine C, and further determine B based on C", etc.

[0082] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program codes.

[0083] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer executable instructions. These computer executable instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0084] These processor executable instructions may also be stored in a processor readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the processor readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0085] These processor-executable instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0086] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A crop fertilizer decision-making method based on multispectral images, characterized in that: The edge computing module applied to the drone airport, the method includes: Receiving an image to be processed sent by a multispectral UAV through a local area network; the image to be processed is a multispectral image of a field collected by the multispectral UAV; Determine the Normalized Difference Vegetation Index (NDVI) according to the near-infrared band reflectance and the red band reflectance of the image to be processed, and determine the Red Edge Chlorophyll Vegetation Index (RECI) according to the near-infrared band reflectance and the red edge band reflectance of the image to be processed; Based on the RECI and the NDVI, the crops in the image to be processed are graded to obtain a plurality of regions with independent boundaries, and the crops in each region are of the same growth grade; The fertilizer requirement of the centroid point of each area is determined, and a fertilizer prescription map is obtained based on the fertilizer requirement of the centroid points of all areas.

2. The crop fertilizer decision-making method based on multispectral images according to claim 1 is characterized in that: Determining the fertilizer requirement of the centroid of each area includes: Based on the NDVI of the centroid of each area, the fertilizer requirement of the centroid is determined according to the nitrogen fertilizer optimization algorithm.

3. The crop fertilizer decision-making method based on multispectral images according to claim 2 is characterized in that: The method of obtaining a fertilization prescription map based on the fertilizer requirement of the centroid of all areas includes: Convert the pixel coordinates of the centroid of each area into longitude and latitude coordinates in the geocentric coordinate system; Based on the fertilizer requirement of the centroid of each area and the latitude and longitude coordinates, using Kriging interpolation method to obtain the crop fertilizer deficiency information of the field; A fertilization prescription map of target resolution is obtained according to the crop fertilizer deficiency information.

4. The crop fertilizer decision-making method based on multispectral images according to claim 1 is characterized in that: The step of grading the growth potential of the crops in the image to be processed based on the RECI and the NDVI includes: Eliminating bare soil in the image to be processed based on the RECI; Based on the NDVI, the crops in the image to be processed after the bare soil is removed are graded for their growth potential.

5. The crop fertilizer decision-making method based on multispectral images according to claim 1 is characterized in that: The method further comprises: Dark current correction and radiation correction are performed on the brightness values ​​of the remote sensing image pixels in different bands of the image to be processed to obtain the reflectivity of the image in different bands in the image to be processed.

6. The crop fertilizer decision-making method based on multispectral images according to claim 1 is characterized in that: The method further comprises: According to different flight speeds and image shooting intervals, unjoined images to be processed are extracted at equal time intervals from the received images to be processed; Obtaining the shooting posture and positioning information of the extracted image to be processed when it was shot; Based on the shooting posture and the positioning information, the to-be-processed images of different bands are corrected and registered.

7. The crop fertilizer decision-making method based on multispectral images according to claim 1 is characterized in that: The multi-spectral UAV replaces the battery through the mechanical arm of the UAV field.

8. A crop fertilizer decision-making device based on multispectral images, characterized in that: include: A receiving module, used for receiving an image to be processed sent by a multispectral UAV through a local area network; the image to be processed is a multispectral image of a field collected by the multispectral UAV; A determination module, used to determine the normalized vegetation index NDVI according to the near-infrared band reflectance and the red band reflectance of the image to be processed, and to determine the red-edge chlorophyll vegetation index RECI according to the near-infrared band reflectance and the red-edge band reflectance of the image to be processed; A first acquisition module is used to grade the growth potential of the crops in the image to be processed based on the RECI and the NDVI to obtain a plurality of regions with independent boundaries, and the crops in each region are of the same growth grade; The second acquisition module is used to determine the fertilizer requirement of the centroid point of each area, and obtain a fertilization prescription map based on the fertilizer requirement of the centroid points of all areas.

9. 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 crop fertilizer decision-making method based on multispectral images as described in any one of claims 1 to 7 is implemented.

10. 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 crop fertilizer decision-making method based on multispectral images as described in any one of claims 1 to 7 is implemented.

Images