Corn cultivation method and cultivation system
By combining isobar cultivation and environmental monitoring systems in the corn planting area, using image data of different heights and color histogram analysis, the problem of low recognition accuracy of corn planting areas is solved, and high-precision recognition and self-direction of corn planting areas are achieved Learn to optimize and accurately judge corn growth.
Patent Information
- Application Number
- CN202411035847.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-11-01
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In large-scale corn planting, when using drones to take ground photos for corn growth and pest identification, excessive height leads to a decrease in recognition accuracy. If the height is too low, multiple photos need to be spliced and insufficient computing power is insufficient, so the corn cannot be achieved Overall intelligent management of planting areas.
Using the planting method of equal ridge distance and plant spacing of 20-30cm, the corn planting area is scanned and real-time data acquisition is obtained through the environmental monitoring system, and the image data sets of different heights are used to match feature points and color histogram analysis to effectively allocate computing power. Perform image recognition, and save and self-learning to optimize the recognition results.
The overall identification of corn planting areas at the optimal height is realized, the recognition accuracy is improved, the overall growth of corn in different growth stages can be accurately judged, and the re-examination results are verified through color histograms to optimize the identification process.
Smart Images

Figure CN118556568B_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of crop cultivation, and relates to a method for cultivating crops, in particular to a corn cultivation method and a cultivation system. Background Art
[0002] In the field of large-scale corn cultivation, unmanned farm operations are used to manage corn.
[0003] At present, drones are used to take ground photos at a certain height, and machine vision technology is used to identify the growth period and pest and disease conditions of corn. However, the higher the height, the blurrier the ground photos taken, resulting in reduced recognition accuracy. If the height is lower, many ground photos need to be stitched together, and the overlapping parts need to be screened out. The required computing power is too high. Due to limited computing power, only local identification of corn planting areas can be carried out through random sampling, and the recognition accuracy is low, and fully intelligent management cannot be achieved.
[0004] Therefore, there is an urgent need to develop a new corn cultivation method and cultivation system to solve the technical problem of insufficient computing power to identify the entire corn planting area. Summary of the invention
[0005] The purpose of the invention is to provide a corn cultivation method and a cultivation system.
[0006] In order to solve the above technical problems, the present invention provides a corn cultivation method, which includes: when planting corn, planning to adopt equal ridge spacing and plant spacing of 20-30cm; scanning the corn planting area to determine the area boundary, and sowing according to the planning information; acquiring corn growth data in the corn planting area in real time to manage the corn.
[0007] Specifically, the corn planting area is scanned through the environmental monitoring system and the corn growth data in the corn planting area is obtained in real time.
[0008] Specifically, the method for acquiring corn growth data in a corn planting area in real time includes: acquiring image data sets at different heights of the corn planting area.
[0009] Specifically, according to the planting requirements, a number of areas to be imaged at the highest altitude are obtained from the image data group at the highest altitude, and computing power is allocated to perform image recognition on the areas to be imaged at the highest altitude, so as to obtain each feature point in each area to be imaged at the highest altitude; according to each area to be imaged at the highest altitude, corresponding areas to be imaged at the lowest altitude are obtained from the image data group at the lowest altitude, and computing power is allocated to perform image recognition on the areas to be imaged at the lowest altitude, so as to obtain each feature point in each area to be imaged at the lowest altitude; each feature point in each area to be imaged at the highest altitude is compared with each feature point in the corresponding area to be imaged at the lowest altitude, that is, when each area to be imaged at the highest altitude is When each feature point in the area to be imaged at a lower height matches each feature point in the area to be imaged at the corresponding lowest height, the image data group at the highest height is set as the target image data group under the planting requirement; or when each feature point in the area to be imaged at each highest height does not match each feature point in the area to be imaged at the corresponding lowest height, each feature point in the area to be imaged at the next height is re-obtained and compared with each feature point in the area to be imaged at the corresponding lowest height, until each feature point in the area to be imaged at one of the heights matches each feature point in the area to be imaged at the corresponding lowest height, and the image data group at that height is set as the target image data group under the planting requirement.
[0010] Specifically, images of corn planting areas at different heights are acquired; and the images of corn planting areas at the same height are respectively formed into corresponding image data groups.
[0011] Specifically, the color histogram corresponding to each image in the image data group is obtained respectively; according to the planting requirements, a number of pixel point concentration areas corresponding to the color histogram are obtained, and then the area to be image recognized corresponding to each pixel point concentration area in the image is obtained.
[0012] Specifically, the target color is determined according to the planting requirements; and the corresponding pixel concentration area is obtained from the color histogram according to the target color.
[0013] Specifically, the number of pixels in the pixel concentration area is obtained, and at the same time, the number of feature points in the area to be imaged corresponding to the pixel concentration area is obtained; the number of pixels obtained is compared with the number of feature points, that is, when the number of pixels obtained is inconsistent with the number of feature points, the target color corresponding to the planting requirements is re-determined until the number of pixels obtained is consistent with the number of feature points.
[0014] Specifically, the image data sets at different time points are saved by compressing the data.
[0015] Specifically, a color histogram corresponding to each image in the image data group at the first time point is obtained, and each color histogram is saved; a color histogram corresponding to each image in the image data group at the next time point is obtained, and the two corresponding color histograms are compared to obtain a difference data string between the next color histogram and the previous color histogram, and each difference data string is saved.
[0016] On the other hand, the present invention provides a cultivation system adopting the corn cultivation method as described above, comprising: an environmental monitoring system; wherein when planting corn, the environmental monitoring system plans to adopt equal ridge spacing and plant spacing of 20-30cm for planting; the environmental monitoring system scans the corn planting area to determine the area boundary, and sows according to the planned information; the environmental monitoring system obtains corn growth data in the corn planting area in real time to manage the corn.
[0017] The beneficial effect of the present invention is that in the process of identifying corn growth data in a corn planting area, the present invention performs mutual verification through coarse-precision recognition and high-precision recognition, can collect and recognize images of the corn planting area at an optimal height, and realize overall recognition of the corn planting area by effectively allocating computing power. At the same time, it can improve recognition accuracy, and compare and verify the color histogram and image recognition during the recognition process, can review the recognition results, and continuously self-learn and optimize to correctly identify different planting needs, and accurately judge the overall growth of corn in different growth stages.
[0018] Other features and advantages of the present invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practicing the present invention.
[0019] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0021] Figure 1 It is a work flow chart of the corn cultivation method provided by an embodiment of the present invention;
[0022] Figure 2 is a workflow diagram for determining a target image data set at a height corresponding to a planting requirement provided by an embodiment of the present invention;
[0023] Figure 3 is a workflow diagram for determining an area to be image recognized provided by an embodiment of the present invention;
[0024] Figure 4 It is a workflow diagram for saving image data sets at different time points provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0025] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. 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] like Figures 1 to 4 As shown, at least one embodiment provides a corn cultivation method, which includes: when planting corn, planning to use equal ridge spacing and plant spacing of 20-30cm; scanning the corn planting area to determine the area boundary, and sowing according to the planning information; acquiring corn growth data in the corn planting area in real time to manage the corn.
[0027] In some embodiments, in the process of identifying corn growth data in a corn planting area, coarse-precision recognition and high-precision recognition are mutually verified, and images of the corn planting area can be collected and recognized at the optimal height. The overall recognition of the corn planting area can be achieved by effectively allocating computing power, and the recognition accuracy can be improved at the same time. In the recognition process, the color histogram is compared and verified with the image recognition, and the recognition results can be reviewed. Continuous self-learning and optimization can correctly identify different planting requirements and accurately judge the overall growth of corn in different growth stages.
[0028] In some embodiments, the corn planting area is scanned by an environmental monitoring system and corn growth data in the corn planting area is acquired in real time.
[0029] Specifically, the environmental monitoring system consists of a host computer, an image acquisition device and several execution devices. The image acquisition device scans the corn planting area and obtains the corn growth data in the corn planting area in real time, and transmits it to the host computer for identification and processing, and then generates corresponding control instructions to each execution device to realize unmanned intelligent management of corn.
[0030] In some embodiments, the method for acquiring corn growth data in a corn planting area in real time includes: acquiring image data sets at different heights of the corn planting area.
[0031] Specifically, the image acquisition equipment is composed of a camera and a drone. The camera is installed on the drone, and the drone is controlled to different heights. The camera obtains image data groups of the corn-planting area at different heights, and the corresponding data is uploaded to the host computer.
[0032] In some embodiments, see Figure 2 , according to the planting requirements, a number of areas to be imaged at the highest heights are obtained from the image data set at the highest height, so as to allocate computing power to perform image recognition on the areas to be imaged at the highest heights, and then obtain each feature point in each area to be imaged at the highest heights; according to each area to be imaged at the highest heights, corresponding areas to be imaged at the lowest heights are obtained from the image data set at the lowest heights, so as to allocate computing power to perform image recognition on the areas to be imaged at the lowest heights, and then obtain each feature point in each area to be imaged at the lowest heights; each feature point in each area to be imaged at the highest height is compared with each feature point in the corresponding area to be imaged at the lowest heights, that is, when each area to be imaged at the highest heights is When each feature point in the area to be imaged matches each feature point in the area to be imaged at the corresponding lowest height, the image data group at the highest height is set as the target image data group under the planting requirement, or when each feature point in the area to be imaged at each highest height does not match each feature point in the area to be imaged at the corresponding lowest height, each feature point in the area to be imaged at the next height is re-obtained and compared with each feature point in the area to be imaged at the corresponding lowest height, until each feature point in the area to be imaged at one of the heights matches each feature point in the area to be imaged at the corresponding lowest height, and the image data group at that height is set as the target image data group under the planting requirement.
[0033] Specifically, different planting requirements correspond to different areas to be image-recognized in the image data set, and thus the feature points finally extracted are also different.
[0034] Specifically, planting requirements refer to the degree of growth, degree of disease, degree of tilt, degree of fruit fullness, etc.
[0035] Specifically, the images are sorted from top to bottom by height. When the feature points in the area to be imaged at the previous height do not match the feature points in the area to be imaged at the corresponding lowest height, the feature points in the area to be imaged at the next height are re-obtained and compared with the feature points in the area to be imaged at the corresponding lowest height, until the feature points in the area to be imaged at a certain height match the feature points in the area to be imaged at the corresponding lowest height, and the image data group at that height is set as the target image data group under the planting requirement.
[0036] Specifically, the feature point may be shape, color, length, width, height, etc.
[0037] In some embodiments, images of corn planting areas at different heights are acquired; and the images of corn planting areas at the same height are respectively grouped into corresponding image data groups.
[0038] Specifically, several photos are taken at different heights, and then the images of each corn-planting area at the same height are respectively formed into corresponding image data groups, and the images in each image data group can be stitched together to show the complete corn-planting area. The higher the height, the lower the pixel of the taken picture, and the greater the difficulty of image recognition. The lower the height, the higher the pixel of the taken picture, and the lower the difficulty of image recognition. At the same time, since the computing power is certain, the higher the height, the lower the accuracy of image recognition, and the lower the height, the more pictures to be taken, and the more computing power required. Therefore, under the premise of confirming the planting needs, only by determining the appropriate height to take images of the corn-planting area can the computing power be effectively utilized, the difficulty of image recognition can be reduced, and the recognition accuracy can be improved.
[0039] In some embodiments, see Figure 3 , respectively obtain the color histogram corresponding to each image in the image data group; obtain a number of pixel point concentration areas corresponding to the color histogram according to the planting requirements, and then obtain the image recognition area corresponding to each pixel point concentration area in the image.
[0040] Specifically, each image can be converted into a color histogram. Combining the planting requirements and the distribution data in the color histogram, the concentrated areas of each pixel point corresponding to the planting requirements can be found from the color histogram, and then the area to be image recognized corresponding to the concentrated area of each pixel point can be quickly extracted from the image.
[0041] Specifically, by converting the color histogram and the image, it is possible to quickly identify the area to be identified.
[0042] In some embodiments, see Figure 3 , determine the target color according to the planting requirements; obtain the corresponding pixel concentration area from the color histogram according to the target color.
[0043] Specifically, different planting requirements correspond to different target colors, and the color histogram reflects the number of pixels of various colors. The corresponding pixel concentration area can be obtained from the color histogram through the target color.
[0044] In some embodiments, see Figure 3, obtain the number of pixels in the pixel concentration area, and at the same time obtain the number of feature points in the image recognition area corresponding to the pixel concentration area; compare the number of pixels obtained with the number of feature points, that is, when the number of pixels obtained is inconsistent with the number of feature points, re-determine the target color corresponding to the planting demand until the number of pixels obtained is consistent with the number of feature points.
[0045] Specifically, by reviewing the number of pixel points and the number of feature points, optimization between planting requirements and target colors is completed through self-learning, thereby improving recognition accuracy.
[0046] In some embodiments, see Figure 4 , image data sets at different time points are saved by compressing data.
[0047] Specifically, data from different time periods need to be saved, so the amount of data will be large, which will increase hardware costs and waste some computing power. Saving image data groups at different time points by compressing data can reduce costs and increase efficiency.
[0048] In some embodiments, see Figure 4 , obtain the color histogram corresponding to each image in the image data group at the first time point, and save each color histogram; obtain the color histogram corresponding to each image in the image data group at the next time point, compare the two corresponding color histograms to obtain the difference data string between the next color histogram and the previous color histogram, and save each difference data string.
[0049] Specifically, since the color histogram can be directly converted into an image, and by storing a difference data string between a next color histogram and a previous color histogram, data compression can be achieved.
[0050] At least one embodiment provides a cultivation system that adopts the corn cultivation method provided in the above embodiments, which includes: an environmental monitoring system; when planting corn, the environmental monitoring system plans to use equal ridge spacing and plant spacing of 20-30cm for planting; the environmental monitoring system scans the corn planting area to determine the area boundary, and sows according to the planning information; the environmental monitoring system obtains corn growth data in the corn planting area in real time to manage the corn.
[0051] To sum up, the present invention verifies each other through coarse-precision recognition and high-precision recognition in the process of identifying corn growth data in the corn-planting area, can collect and recognize images of the corn-planting area at the optimal height, and realize overall recognition of the corn-planting area by effectively allocating computing power. At the same time, it can improve the recognition accuracy, and compare and verify the color histogram and image recognition during the recognition process, which can review the recognition results, and continuously self-learn and optimize to correctly identify different planting requirements, and accurately judge the overall growth of corn in different growth stages.
[0052] In the description of the embodiments of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.
[0053] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.
[0054] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.
Claims
1. A corn cultivation method, It is characterized in that include: When planting corn, plan to plant with equal ridge spacing and 20-30cm spacing between plants; Scan corn planting areas to determine area boundaries and sow according to planning information; Obtain corn growth data in corn planting areas in real time to manage corn; Scan the corn planting area through the environmental monitoring system and obtain the corn growth data in the corn planting area in real time; The method for obtaining corn growth data in a corn planting area in real time comprises: Acquire image data sets of corn planting areas at different heights; According to the planting requirements, a number of areas to be imaged at the highest altitude are obtained from the image data set at the highest altitude, so as to allocate computing power to perform image recognition on the areas to be imaged at each highest altitude, and then obtain each feature point in the area to be imaged at each highest altitude; According to the image recognition regions at the highest heights, the corresponding image recognition regions at the lowest heights are obtained from the image data group at the lowest heights, so as to allocate computing power to perform image recognition on the image recognition regions at the lowest heights, and then obtain each feature point in the image recognition region at each lowest height; Compare each feature point in the image recognition area at each highest height with each feature point in the image recognition area at the corresponding lowest height, that is, When each feature point in the image recognition area at each highest height matches each feature point in the corresponding image recognition area at the lowest height, the image data set at the highest height is set as the target image data set under the planting requirement, or When each feature point in the area to be imaged at each highest height does not match each feature point in the area to be imaged at the corresponding lowest height, each feature point in the area to be imaged at the next height is re-obtained and compared with each feature point in the area to be imaged at the corresponding lowest height, until each feature point in the area to be imaged at one height matches each feature point in the area to be imaged at the corresponding lowest height, and the image data set at that height is set as the target image data set under the planting requirement; Obtain images of corn growing areas at different heights; The images of each corn planting area at the same height are respectively formed into corresponding image data groups; Respectively obtain the color histogram corresponding to each image in the image data group; According to the planting requirements, a plurality of pixel point concentration areas corresponding to the color histogram are obtained, and then the image recognition area corresponding to each pixel point concentration area in the image is obtained; Determine the target color based on planting needs; Obtain the corresponding pixel concentration area from the color histogram according to the target color; Obtain the number of pixels in the pixel concentration area, and at the same time obtain the number of feature points in the image recognition area corresponding to the pixel concentration area; Compare the number of pixels obtained with the number of feature points, that is, When the number of acquired pixel points is inconsistent with the number of feature points, the target color corresponding to the planting requirement is re-determined until the number of acquired pixel points is consistent with the number of feature points.
2. The corn cultivation method according to claim 1, It is characterized in that The image data sets at different time points are saved by compressing the data.
3. The corn cultivation method according to claim 2, It is characterized in that Obtaining a color histogram corresponding to each image in the image data group at the first time point, and saving each color histogram; Obtain the color histogram corresponding to each image in the image data group at the next time point, compare the two corresponding color histograms to obtain the difference data string between the next color histogram and the previous color histogram, and save each difference data string.
4. A cultivation system using the corn cultivation method according to any one of claims 1 to 3, It is characterized in that include: Environmental monitoring system; When planting corn, the environmental monitoring system plans to plant with equal ridge spacing and plant spacing of 20-30 cm; The environmental monitoring system scans the corn planting area to determine the area boundaries and sows according to the planning information; The environmental monitoring system acquires corn growth data in a corn planting area in real time to manage the corn.
5. A computer device comprising a memory, a processor and a computer program stored in the memory, It is characterized in that The processor executes the computer program to implement the steps of the corn breeding method according to claim 1.
6. A computer-readable storage medium having a computer program / instructions stored thereon, It is characterized in that When the computer program / instructions are executed by a processor, the steps of the corn cultivation method according to claim 1 are implemented.
