Intelligent fertilizer supply system and method for small crop areas

Abstract

The present invention discloses a small-area intelligent fertilization system for crops and a fertilization method thereof. The surface color of the functional leaves of crops in the target area of ​​crop planting is captured by an image acquisition device, and the leaf color value of the functional leaves of crops in the image is extracted by an artificial intelligence image processing algorithm to construct a leaf color distribution model, which is then compared with the standard leaf color value threshold range to accurately determine whether the crops in the target area of ​​the crops are under-fertilized. If there is an under-fertilized area in the target area of ​​the crops, the real-time leaf color value of the functional leaves of the crops in the under-fertilized area is matched with the leaf color value database by a machine learning algorithm to obtain data in a fertilizer formula database, i.e., the optimal fertilization plan, and finally the drone performs accurate fertilization for the under-fertilized area. The present invention solves the problem that the existing application of nitrogen fertilizer to crops is based on empirical observation of crop growth, and there is an inaccurate amount of fertilizer.

Description

Technical Field

[0001] The invention relates to the technical field of crop cultivation, and in particular to a small-area intelligent fertilizer supplementation system for crops and a fertilizer supplementation method thereof. Background Art

[0002] Whether it is agricultural crops or cash crops, fertilizers need to be applied to them during their growth, especially nitrogen fertilizers, which are the most effective type of fertilizer in increasing agricultural production and have always played an important role in agricultural production. However, at present, the world can only rely on empirical guidance for the application of nitrogen fertilizers to crops. The application of nitrogen fertilizers is closely related to environmental and climatic factors. In areas with suitable environment and climate, the nitrogen application period, nitrogen application amount and uniformity of crops will also be affected by the climate conditions of the year such as rainfall and temperature, as well as human factors. This leads to the inevitable occurrence of crop fertilizer shortages during the fertilization process, which in turn affects yield, quality and economic benefits.

[0003] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgment or any form of suggestion that the information constitutes the prior art already known to a person skilled in the art. Summary of the invention

[0004] In order to overcome the defects of the prior art, an intelligent fertilization system for small areas of crops and a fertilization method thereof are now provided to solve the problem that the existing application of nitrogen fertilizer to crops is based on empirical observation of crop growth conditions, resulting in inaccurate fertilization amounts.

[0005] In order to achieve the above purpose, a small-area intelligent fertilizer supplementation system for crops is provided, comprising:

[0006] A drone, wherein the drone is equipped with an image acquisition device and a positioning device for acquiring image information of a planting area, and the drone is mounted with nitrogen fertilizer;

[0007] The controller comprises a control module connected to the drone, the image acquisition device and the positioning device by signal, an image recognition module for identifying the real-time leaf color values ​​of multiple functional leaves on the top of each crop in the planting area image, a modeling module for constructing a leaf color distribution model of the planting area based on the geographical location information of the planting area and the real-time leaf color values, a storage module for storing a leaf color value database of the crop and a fertilizer formula database, a determination module for comparing the real-time leaf color value with a standard leaf color threshold based on the leaf color distribution model to determine the underfertilized area of ​​the leaf color distribution model, and a determination module for determining the underfertilized area of ​​the leaf color distribution model based on the leaf color distribution model. a matching module for matching the real-time leaf color value data in the underfertilized area to a corresponding fertilizing scheme in the fertilization formula database, a planning module for generating a fertilizing path based on the position of the UAV and the distribution of the underfertilized area on the leaf color distribution model, the image recognition module and the storage module are respectively connected to the control module, the modeling module is connected to the image recognition module, the determination module is connected to the modeling module, the matching module is connected to the control module, the storage module and the determination module, and the planning module is connected to the control module, the modeling module and the determination module;

[0008] The control module obtains the fertilization plan and the fertilization path for the underfertilized area, generates an execution signal carrying the fertilization plan and the fertilization path and sends it to the outside, and the drone receives the execution signal and spreads the nitrogen fertilizer to each of the underfertilized areas based on the fertilization path and the fertilization plan.

[0009] Furthermore, it also includes a display device, which is connected to the control module.

[0010] Furthermore, the display device is a touch screen.

[0011] Furthermore, the image acquisition device is a high frame rate color camera.

[0012] Furthermore, the leaf color value is a SPAD value or an LCC value.

[0013] Furthermore, the number of the functional blades is three.

[0014] Furthermore, the leaf color value is a ratio of the leaf color value of the functional leaves on the top layer to the leaf color value of the functional leaves on the bottom layer.

[0015] Furthermore, the drone is a plant protection drone.

[0016] The present invention provides a fertilization method of a small-area intelligent fertilization system for crops, comprising the following steps:

[0017] The drone is above the planting area and collects images of the planting area;

[0018] The control module of the controller acquires the planting area image;

[0019] The image recognition module of the controller recognizes the real-time leaf color values ​​of a plurality of functional leaves on the top of each crop in the planting area image;

[0020] The modeling module of the controller constructs a leaf color distribution model of the planting area based on the geographical location information of the planting area and the real-time leaf color value;

[0021] The determination module of the controller compares the real-time leaf color value with a standard leaf color threshold value based on the leaf color distribution model to determine an underfertilized area of ​​the leaf color distribution model;

[0022] The matching module of the controller matches the real-time leaf color value data in the underfertilized area to a corresponding fertilization scheme in the fertilization formula database in the storage module based on the leaf color value database in the storage module of the controller;

[0023] The control module obtains the fertilization scheme and the fertilization path of the insufficient fertilization area, generates an execution signal carrying the fertilization scheme and the fertilization path, and sends the execution signal to the outside;

[0024] The drone receives the execution signal and spreads the nitrogen fertilizer to each of the insufficient-fertilizer areas based on the fertilization path and the fertilization plan.

[0025] The beneficial effect of the present invention is that the crop small area intelligent fertilizer supplement system of the present invention captures the surface color of the functional leaves of the crops in the target area of ​​crop planting through an image acquisition device, and uses an artificial intelligence image processing algorithm to extract the leaf color value of the functional leaves of the crops in the image to construct a leaf color distribution model, and then compares it with the standard leaf color value threshold range to accurately determine whether the crops in the target area of ​​the crops are underfertilized. If there is an underfertilized area in the target area of ​​the crops, the real-time leaf color value of the functional leaves of the crops in the underfertilized area is matched with the leaf color value database through a machine learning algorithm to obtain data in the fertilizer formula database, that is, the optimal fertilization plan, and finally the drone performs accurate fertilization for the underfertilized area. The crop small area intelligent fertilizer supplement system of the present invention completely changes the drawbacks of traditional naked eye observation, improves the refinement and quantification level of crop cultivation management, and improves the quality and yield of crops, that is, saves fertilizer costs and labor costs, and can avoid a series of adverse effects such as soil compaction caused by fertilizer abuse. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0027] Figure 1 The figure is a schematic diagram of the structure of a small-area intelligent fertilizer supplementation system for crops according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0029] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0030] Reference Figure 1 As shown, the present invention provides a small-area intelligent fertilization system for crops, comprising: a drone 1, an image acquisition device 2, a positioning device 3 and a controller 4.

[0031] The drone 1 is equipped with an image acquisition device 2 and a positioning device 3 for acquiring image information of the planting area. The drone 1 is equipped with nitrogen fertilizer;

[0032] The controller 4 includes a control module 41 , an image recognition module 42 , a storage module 44 , a determination module 45 , a matching module 46 , and a planning module 47 .

[0033] The control module 41 is connected to the drone 1, the image acquisition device 2 and the positioning device 3. The image recognition module 42 and the storage module 44 are connected to the control module 41 respectively. The modeling module 43 is connected to the image recognition module 42. The determination module 45 is connected to the modeling module 43. The matching module 46 is connected to the control module 41, the storage module 44 and the determination module 45. The planning module 47 is connected to the control module 41, the modeling module 43 and the determination module 45.

[0034] The image recognition module 42 is used to recognize the real-time leaf color values ​​of multiple functional leaves on the top of each crop in the planting area image.

[0035] The modeling module 43 is used to construct a leaf color distribution model of the planting area based on the geographical location information and real-time leaf color values ​​of the planting area.

[0036] The storage module 44 is used to store the leaf color value database and the fertilizer formula database of the crops.

[0037] The determination module 45 is used to compare the real-time leaf color value with the standard leaf color threshold value based on the leaf color distribution model to determine the under-fertilized area of ​​the leaf color distribution model.

[0038] The matching module 46 is used to match the real-time leaf color value data in the under-fertilized area to a corresponding fertilization scheme in the fertilization formula database based on the leaf color value database.

[0039] The planning module 47 is used to generate a fertilization path based on the position of the drone 1 and the distribution of the underfertilized area on the leaf color distribution model.

[0040] The control module 41 obtains the fertilization plan and fertilization path of the underfertilized area, generates an execution signal carrying the fertilization plan and fertilization path and sends it to the outside, and the drone 1 receives the execution signal and spreads nitrogen fertilizer to each underfertilized area based on the fertilization path and fertilization plan.

[0041] The crop small area intelligent fertilizer supplementation system of the present invention captures the surface color of the functional leaves of crops in the target area of ​​crop planting through an image acquisition device, and uses an artificial intelligence image processing algorithm to extract the leaf color value of the functional leaves of crops in the image to construct a leaf color distribution model, and then compares it with the standard leaf color value threshold range to accurately determine whether the crops in the target area of ​​the crops are underfertilized. If there is an underfertilized area in the target area of ​​the crops, the real-time leaf color value of the functional leaves of the crops in the underfertilized area is matched with the leaf color value database through a machine learning algorithm to obtain data in the fertilizer formula database, that is, the optimal fertilization plan, and finally the drone performs accurate fertilization for the underfertilized area. The crop small area intelligent fertilizer supplementation system of the present invention completely changes the drawbacks of traditional naked eye observation, improves the refinement and quantification level of crop cultivation management, and improves the quality and yield of crops, that is, saves fertilizer costs and labor costs, and can avoid a series of adverse effects such as soil compaction caused by fertilizer abuse.

[0042] The crop small area intelligent fertilizer supplementation system of the present invention also includes a display device 5. The display device 5 is connected to the control module 41. Specifically, the display device 5 is a touch screen. In actual operation, the plant protection operator can hold the controller, and the display device is used to display the leaf color distribution model and the location information of the underfertilized area, so that the plant protection operator can intuitively manage the target area of ​​the crop on site.

[0043] In a preferred embodiment of crops, the image acquisition device 2 is a high frame rate color camera. The small area intelligent fertilization system for crops of the present invention uses a drone equipped with a high frame rate color camera to patrol the target area. The high frame rate color camera can also avoid the phenomenon that the functional leaves of the plants cannot be identified due to the wind blowing, and has the advantages of high mobility and high efficiency.

[0044] The image recognition module of the small-area intelligent fertilization system for crops of the present invention compares the functional leaf images of plants captured by a high-frame-rate color camera with the data in the leaf color value database through a machine learning algorithm, and can make an accurate judgment at the early stage of over-fertilization / under-fertilization of plants, thereby avoiding the occurrence of plant death due to untimely judgment. On the other hand, the small-area intelligent fertilization system for crops of the present invention uses drones to intelligently, accurately, quickly, and efficiently remedy under-fertilized areas, and can intelligently provide scientific guidance on the types and amounts of fertilizers that plants in under-fertilized areas lack.

[0045] In this embodiment, the leaf color value is a SPAD value or an LCC value. The SPAD (soil and plant analyzer development) value is a relative value of the chlorophyll content of the leaf. The LCC (leaf color chart) value is the reading after comparison with the leaf color card.

[0046] The image recognition template uses a machine learning algorithm to identify and extract the three most recently extended functional leaves at the top of the crop. The leaf color value is obtained by calculating the ratio of the leaf color value of the functional leaves at the top layer to the leaf color value of the functional leaves at the bottom layer, thereby improving the accuracy of the leaf color value of the small-area intelligent fertilization system for crops of the present invention, and avoiding the external environmental factors from having a greater impact on the image acquisition device. Specifically, the image recognition template uses a machine learning algorithm to identify the YOLO (You Only Look Once) algorithm to identify and extract the three most recently extended functional leaves at the top of the crop.

[0047] YOLO and You Only Look Once is a target detection algorithm. The goal of the target detection task is to find all the regions of interest in the image and determine the location and category probability of these regions. YOLO is a one-stage target detection algorithm.

[0048] In this embodiment, the drone 1 is a plant protection drone, which carries nitrogen fertilizer when taking off, and can quickly and efficiently perform fertilizer replenishment after receiving an execution signal.

[0049] The present invention provides a fertilization method of a small-area intelligent fertilization system for crops, comprising the following steps:

[0050] S1: UAV 1 is above the planting area and collects images of the planting area.

[0051] S2: The control module 41 of the controller 4 acquires an image of the planting area.

[0052] S3: The image recognition module 42 of the controller 4 recognizes the real-time leaf color values ​​of the plurality of functional leaves at the top of each crop in the planting area image.

[0053] S4: The modeling module 43 of the controller 4 constructs a leaf color distribution model of the planting area based on the geographical location information and the real-time leaf color values ​​of the planting area.

[0054] S5: The determination module 45 of the controller 4 compares the real-time leaf color value with the standard leaf color threshold value based on the leaf color distribution model to determine the under-fertilized area of ​​the leaf color distribution model.

[0055] S6: The matching module 46 of the controller 4 matches the real-time leaf color value data in the under-fertilized area to a corresponding fertilization scheme in the fertilization formula database in the storage module 44 based on the leaf color value database in the storage module 44 of the controller 4 .

[0056] S7: the control module 41 obtains a fertilization scheme and a fertilization path for the insufficient fertilization area, generates an execution signal carrying the fertilization scheme and the fertilization path, and sends the execution signal to the outside.

[0057] S8: The drone 1 receives the execution signal and spreads nitrogen fertilizer to each underfertilized area based on the fertilizer supplementation path and fertilizer supplementation plan.

[0058] The present invention provides a small-area intelligent fertilizer supplementation system for crops, which is constructed by carrying three major parts, namely, a drone, a high-frame-rate color camera, and a controller. The high-frame-rate color camera is carried by the drone, and the drone is also equipped with a fertilizer spreading device. The controller is integrated with chips or other electrical components such as a leaf color value database, a fertilizer formula database, and an image recognition module.

[0059] First, a drone equipped with a high-frame-rate color camera flies over the target area, takes aerial photos of the target area according to the predetermined flight route, and transmits the captured images back to the controller; the image recognition module in the controller identifies the transmitted images, captures the three functional leaves of each plant, and establishes a SPAD value or LCC value distribution model for the functional leaves by analyzing the surface color of the functional leaves; the judgment module and the matching module are embedded with a machine learning algorithm, which can compare the SPAD value or LCC value distribution model for the functional leaves established by the modeling module with the standard data in the leaf color value database, and determine whether the plants in the target area are under-fertilized.

[0060] If there are plants in the target area that are under-fertilized, the matching module will compare and analyze the target functional leaves with the leaf color value database, retrieve the data in the fertilization formula database, and analyze the amount of fertilizer missing for the corresponding target plants and the optimal amount and application plan of fertilizer. At the same time, the display device will display the relevant information.

[0061] In addition, the drone is implanted with a positioning device, and its position coordinates will be transmitted back to the controller in real time. When the controller determines that the plants in a certain area within the target area are under-fertilized, the planning module will associate the image of the plants under-fertilized with the position coordinates transmitted back by the drone through time parameters, and draw a coordinate graph of the under-fertilized area with the controller position as the zero point coordinate to determine the relative position relationship between the under-fertilized area and the controller. This information will be transmitted to the fertilizer spreading device carried by the drone.

[0062] Finally, the coordinate information carried in the fertilization path is used for autonomous flight. When the drone flies over the area with insufficient fertilizer, the fertilizer spreading device starts to release fertilizer. The fertilizer is blown away by the high-speed airflow and falls evenly in the area with insufficient fertilizer.

[0063] Because it is impossible to fertilize crops in extreme weather conditions such as rain, snow, and fog, but environmental factors such as sunlight intensity will affect the clarity of drone photography. To solve this problem, the drone in the small-area intelligent fertilization system for crops of the present invention adopts a fixed-altitude and fixed-speed cruise mode for aerial image acquisition. When the drone is first launched for operations in the target area, the controller will interpret the images transmitted back by the drone and determine the best flight altitude and flight speed relative to the ground that can meet the image clarity. During the entire operation, the drone will perform flight acquisition according to this relative flight altitude and speed. In addition, when interpreting the photos taken by the drone when it first takes off, the controller will comprehensively analyze the environmental factors during the operation. When processing and analyzing the later images, the controller will perform targeted repair processing on the images based on the environmental factors.

[0064] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the present application.

Claims

1. A small-area intelligent fertilizer supplementation system for crops, characterized in that: include: A drone, wherein the drone is equipped with an image acquisition device and a positioning device for acquiring image information of a planting area, and the drone is mounted with nitrogen fertilizer; The controller comprises a control module connected to the drone, the image acquisition device and the positioning device by signal, an image recognition module for identifying the real-time leaf color values ​​of multiple functional leaves on the top of each crop in the planting area image, a modeling module for constructing a leaf color distribution model of the planting area based on the geographical location information of the planting area and the real-time leaf color values, a storage module for storing a leaf color value database of the crop and a fertilizer formula database, a determination module for comparing the real-time leaf color value with a standard leaf color threshold based on the leaf color distribution model to determine the underfertilized area of ​​the leaf color distribution model, and a determination module for determining the underfertilized area of ​​the leaf color distribution model based on the leaf color distribution model. a matching module for matching the real-time leaf color value data in the underfertilized area to a corresponding fertilizing scheme in the fertilization formula database, a planning module for generating a fertilizing path based on the position of the UAV and the distribution of the underfertilized area on the leaf color distribution model, the image recognition module and the storage module are respectively connected to the control module, the modeling module is connected to the image recognition module, the determination module is connected to the modeling module, the matching module is connected to the control module, the storage module and the determination module, and the planning module is connected to the control module, the modeling module and the determination module; The control module obtains the fertilization scheme and the fertilization path of the insufficient fertilization area, generates an execution signal carrying the fertilization scheme and the fertilization path, and sends the execution signal to the outside, and the drone receives the execution signal and spreads the nitrogen fertilizer to each of the insufficient fertilization areas based on the fertilization path and the fertilization scheme; The leaf color value is a SPAD value or a LCC value; The leaf color value is a ratio of the leaf color value of the functional leaves on the top layer to the leaf color value of the functional leaves on the bottom layer.

2. The small-area intelligent fertilizer supplementation system for crops according to claim 1 is characterized in that: It also includes a display device, which is connected to the control module.

3. The small-area intelligent fertilizer supplementation system for crops according to claim 2 is characterized in that: The display device is a touch screen.

4. The small-area intelligent fertilizer supplementation system for crops according to claim 1 is characterized in that: The image acquisition device is a high frame rate color camera.

5. The small-area intelligent fertilizer supplementation system for crops according to claim 1 is characterized in that: The number of the functional blades is three.

6. The small-area intelligent fertilizer supplementation system for crops according to claim 1 is characterized in that: The UAV is a plant protection UAV.

7. A method for fertilizing crops in a small area using an intelligent fertilizing system as claimed in any one of claims 1 to 6, characterized in that: The following steps are involved: The drone is above the planting area and collects images of the planting area; The control module of the controller acquires the planting area image; The image recognition module of the controller recognizes the real-time leaf color values ​​of a plurality of functional leaves on the top of each crop in the planting area image; The modeling module of the controller constructs a leaf color distribution model of the planting area based on the geographical location information of the planting area and the real-time leaf color value; The determination module of the controller compares the real-time leaf color value with a standard leaf color threshold value based on the leaf color distribution model to determine an underfertilized area of ​​the leaf color distribution model; The matching module of the controller matches the real-time leaf color value data in the underfertilized area to a corresponding fertilization scheme in the fertilization formula database in the storage module based on the leaf color value database in the storage module of the controller; The control module obtains the fertilization scheme and the fertilization path of the insufficient fertilization area, generates an execution signal carrying the fertilization scheme and the fertilization path, and sends the execution signal to the outside; The drone receives the execution signal and spreads the nitrogen fertilizer to each of the insufficient-fertilizer areas based on the fertilization path and the fertilization plan.

Images