Crop shooting height determination method, electronic equipment and storage medium

Abstract

The invention provides a crop shooting height determination method, which comprises the following steps: when the regional area of a target field is greater than a preset area threshold value, calculating the number of field blocks after division of the target field according to the regional area of the target field and the preset area threshold value, according to the method, a target field is averagely divided into a plurality of target areas according to the number of field blocks, a plurality of initial remote sensing images corresponding to shooting of each target area are acquired based on a plurality of preset unmanned aerial vehicle shooting heights, and a plurality of remote sensing images corresponding to each target area are acquired based on the plurality of initial remote sensing images corresponding to each target area. Obtaining a first shooting height corresponding to each target area, and determining an average shooting height of the first shooting heights corresponding to the plurality of target areas as a final shooting height; according to the invention, for a target field with a relatively large area, a relatively large area range can be shot, and meanwhile, the definition of the shot plant image is ensured, so that the growth condition of more plants can be clearly observed.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a method for determining a crop shooting height, an electronic device and a storage medium. Background Art

[0002] Crop detection and evaluation can reflect the plant's planting density, survival rate, growth and other growth status information, which is beneficial to crop research, yield improvement, etc. The rapidly developing remote sensing technology can provide macro and rapid detection, and use spectral vegetation indices to conduct qualitative and quantitative analysis of crop growth conditions. Therefore, in the current crop detection method, remote sensing data can be used for remote sensing detection of planted crops. However, when shooting remote sensing images, especially for fields that are too large, it is very meaningful to set the shooting height to capture remote sensing images with a large range and high clarity. Summary of the invention

[0003] In view of the above technical problems, the technical solution adopted by the present invention is:

[0004] A method for determining a crop shooting height, characterized in that the method comprises the following steps:

[0005] An initial shooting height corresponding to the target field is received; the initial shooting height refers to the lowest shooting height that can capture the entire target field.

[0006] When the area of ​​the target field is larger than a preset area threshold, the number of field blocks after the target field is divided is calculated according to the area of ​​the target field and the preset area threshold, so as to evenly divide the target field into a number of target areas according to the number of field blocks.

[0007] Based on a plurality of preset drone shooting heights, a plurality of initial remote sensing images corresponding to the shooting of each target area are obtained, and based on the plurality of initial remote sensing images corresponding to each target area, a first shooting height corresponding to each target area is obtained; the plurality of drone shooting heights refer to a plurality of shooting heights set from top to bottom with the lowest shooting height as the upper limit height.

[0008] An average shooting height of the first shooting heights corresponding to the plurality of target areas is determined as a final shooting height.

[0009] The present invention has at least the following beneficial effects:

[0010] The present invention provides a method for determining a crop shooting height. The method comprises the following steps: firstly judging the size of a target field. When the area of ​​the target field is greater than a preset area threshold, the number of field blocks after the target field is divided is calculated according to the area of ​​the target field and the preset area threshold, so as to evenly divide the target field into a plurality of target areas according to the number of field blocks, so as to shoot a remote sensing image for each target area separately, thereby ensuring the clarity of the remote sensing image. That is, based on a plurality of preset unmanned aerial vehicle shooting heights, a plurality of initial remote sensing images corresponding to each target area are obtained when shooting, and based on the plurality of initial remote sensing images corresponding to each target area, a first shooting height corresponding to each target area is obtained, so that the unmanned aerial vehicle can shoot a clear plant image and a larger area range at the same time, and finally the average shooting height of the first shooting heights corresponding to the plurality of target areas is determined as the final shooting height. When the present invention targets a target field with a larger area, the method can shoot a larger area range while ensuring the clarity of the shot plant image, thereby facilitating the clear observation of the growth of more plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 This is a flow chart of a method for determining a crop photographing height provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0013] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0014] This embodiment provides a method for determining the crop shooting height. Figure 1 As shown, the method comprises the following steps:

[0015] S100, receiving an initial shooting height corresponding to a target field; the initial shooting height refers to a minimum shooting height capable of shooting a complete target field.

[0016] S200, when the area of ​​the target field is greater than a preset area threshold, the number of field blocks after the target field is divided is calculated according to the area of ​​the target field and the preset area threshold, so as to evenly divide the target field into a plurality of target areas according to the number of field blocks. The preset area threshold can be set by a person skilled in the art according to actual needs, that is, it can be an empirical value. In a specific implementation, after the target field is divided into a plurality of target areas, identifications are set for the ranges of different target areas, so that the drone can quickly and clearly identify the desired shooting area when shooting.

[0017] Specifically, the number of target blocks meets the following conditions:

[0018] E=roundup(C / C0), where E is the number of field blocks, C is the area of ​​the target field, C0 is a preset area threshold, and roundup() is an upward rounding function.

[0019] In another embodiment, when the area of ​​the target field is not greater than a preset area threshold, the initial photographing height is determined as the final photographing height.

[0020] As mentioned above, when the area of ​​the target field is large, if the complete target field is captured in one remote sensing image, the growth status of the plants in the image will be unclear. Therefore, the drone shooting height needs to be lowered. At this time, the complete target field cannot be captured. Therefore, the target field needs to be divided into areas to capture remote sensing images separately to ensure the clarity of the target plants in the remote sensing image.

[0021] Preferably, the preset area threshold is determined by the following steps:

[0022] S01, receiving some target field information; the target field information includes the actual area of ​​the target field, the key shooting height corresponding to the target field and the actual area photographed at the key shooting height; the key shooting height is the shooting height corresponding to the remote sensing image corresponding to the target field determined by the target user when it meets the preset requirements, and the types of crops planted in any two target fields are different; it can be understood that: the key shooting height refers to the maximum shooting height determined by the user from multiple shooting heights whose corresponding remote sensing image clarity meets the requirements based on the experimental values ​​of several shooting heights corresponding to the target field.

[0023] S02, calculating an average height according to the key shooting height in each target field information, and determining the actual shooting area corresponding to the average height as a preset area threshold.

[0024] As mentioned above, when determining the area threshold, several target fields are taken as samples, and the maximum shooting height corresponding to each target field that can be clearly photographed is obtained. It can be known that the shooting heights corresponding to different crops when they are clearly photographed can be known. The average shooting height obtained in this way is universal, and a reasonable area threshold is determined.

[0025] S300, based on a plurality of preset drone shooting heights, obtain a plurality of initial remote sensing images corresponding to each target area when shooting, and based on the plurality of initial remote sensing images corresponding to each target area, obtain a first shooting height corresponding to each target area; the plurality of drone shooting heights refer to a plurality of shooting heights set from top to bottom with the lowest shooting height as the upper limit height. When setting the drone shooting height, it can be set at equal height intervals, for example, the height interval is 2 meters or 5 meters.

[0026] In a specific embodiment, the first shooting height corresponding to any target area is obtained through the following steps:

[0027] S301, for any initial remote sensing image corresponding to any target area, obtain a plant image feature set corresponding to the target plants in the target area from the initial remote sensing image; the plant image feature set includes the average plant size, plant size dispersion, color saturation, and the average interval and interval dispersion between adjacent target plants corresponding to the target plants in the initial remote sensing image; it can be understood that: the target plant refers to any plant of crops planted in the target area, and the plant size is the area size of the plant.

[0028] In a specific embodiment, the average plant size corresponding to the target plant is obtained by the following steps:

[0029] S3011, for any initial remote sensing image, according to the band reflection of each target plant in the initial remote sensing image, identify the planting area corresponding to each target plant in the target area from the initial remote sensing image. Among them, obtaining the planting area of ​​the plant according to the color reflection of the plant leaves is a prior art and will not be repeated here.

[0030] S3012, obtaining the number of pixel points in the planting area corresponding to each target plant in the target area and the number of pixel points in the initial remote sensing image.

[0031] S3013, calculating the actual area corresponding to the initial remote sensing image according to the drone shooting height corresponding to the initial remote sensing image, the drone diagonal field of view and the size information of the initial remote sensing image; it can be understood that: the size information of the initial remote sensing image includes the length and width of the initial remote sensing image.

[0032] Specifically, the actual area A corresponding to the initial remote sensing image meets the following conditions:

[0033] A=L1×W1×4H 2 (tan(θ / 2)) 2 / (L1 2 ×W1 2 ), where L1 is the length of the initial remote sensing image, W1 is the width of the initial remote sensing image, H is the UAV shooting height corresponding to the initial remote sensing image, and θ is the diagonal field of view of the UAV.

[0034] S3014, based on the number of pixel points in the planting area corresponding to each target plant in the target area, the number of pixel points in the initial remote sensing image and the actual area corresponding to the initial remote sensing image, the plant size corresponding to each target plant is calculated to obtain the average plant size corresponding to the target plants.

[0035] Specifically, the plant size K of any target plant meets the following conditions:

[0036] K=D1 / D2×A, where D1 is the number of pixel points in the planting area corresponding to any target plant in the target area, and D2 is the number of pixel points in the initial remote sensing image.

[0037] Specifically, the plant size discreteness refers to the discreteness obtained after performing a variance operation on the plant sizes corresponding to a number of target plants; the interval discreteness refers to the discreteness obtained after performing a variance operation on the intervals between a number of adjacent target plants.

[0038] As mentioned above, remote sensing technology can be used to identify the growth area corresponding to each target plant, and then obtain a number of plant image features. According to the several plant image features, the growth status of the target plant photographed at different shooting heights can be reflected, which is conducive to obtaining the similarity between two adjacent initial remote sensing images, and then determining a reasonable and clear shooting height.

[0039] S302, for a plurality of initial remote sensing images corresponding to any target area, according to a plant image feature set corresponding to each initial remote sensing image, obtain a first image similarity corresponding to each two adjacent initial remote sensing images at the shooting heights of the drone.

[0040] In a specific embodiment, step S302 includes the following steps:

[0041] S3021, for the initial remote sensing images taken at any two adjacent UAV shooting heights, obtain the feature similarity corresponding to each plant image feature.

[0042] Specifically, the feature similarity S of any plant image feature corresponding to two adjacent initial remote sensing images meets the following conditions:

[0043] S=α / β, where α is the minimum value of any plant image feature in any two adjacent initial remote sensing images, and β is the maximum value of any plant image feature in any two adjacent initial remote sensing images.

[0044] Specifically, two adjacent initial remote sensing images refer to two initial remote sensing images corresponding to any two adjacent shots during the process of the drone shooting from top to bottom.

[0045] S3022, the feature similarities corresponding to the features of the plurality of plant images are sequentially summed up to obtain a first image similarity corresponding to the initial remote sensing images at two adjacent UAV shooting heights.

[0046] As described above, by calculating the feature similarity of each plant image feature, the change of each plant image feature corresponding to any two adjacent initial remote sensing images can be reflected. By adding several feature similarities, the image similarity between two adjacent initial remote sensing images can be reflected, which is helpful to judge the influence of different shooting heights on the captured plant status according to the image similarity.

[0047] S303, a curve is drawn according to the shooting order of the initial remote sensing images and the first image similarity corresponding to the initial remote sensing images at each two adjacent drone shooting heights, and when there is an inflection point in the drawn curve, the first image similarity corresponding to the inflection point is determined as the second image similarity. It can be understood that: the curve is drawn with the shooting sequence number as the horizontal coordinate and the corresponding first image similarity as the vertical coordinate.

[0048] S304: For the drone shooting heights of the two initial remote sensing images corresponding to the second image similarity, determine the larger drone shooting height as the first shooting height of the target plant.

[0049] As mentioned above, since the lower the shooting height, the clearer the captured image, as the shooting height decreases, the similarity of the first image will become greater and greater, and after a clear image is captured at a certain shooting height, since subsequent images are all clear, the similarity of the first image will reach a balance, and the first shooting height corresponding to the balance will be used as the first shooting height, so that the drone can capture clear plant images while also capturing a larger area, which is conducive to observing more plant growth conditions.

[0050] It should also be noted that due to the influence of many factors during shooting, such as lighting, wind direction, etc., when the first image similarity reaches equilibrium, it may appear as micro-fluctuations in the drawn curve. At this time, the curve corresponds to more than one inflection point. In this case, considering that the change values ​​of several consecutive first image similarities are all less than the set fluctuation threshold, the first image similarity among several consecutive first image similarities can be determined as the second image similarity to determine the first shooting height of the target plant.

[0051] In another specific embodiment, the method further determines a first shooting height of the target plant by the following steps:

[0052] S10, when there is no inflection point in the drawn curve, for any plant image feature among the plant size discreteness, color saturation and interval discreteness, the several characteristic values ​​corresponding to the plant image feature are normalized to between 0 and 1, so as to obtain several initial characteristic scores corresponding to each initial remote sensing image. For example, when a total of 20 initial remote sensing images are taken, any plant image feature corresponds to 20 characteristic values, all of which are normalized to between 0 and 1. The min-max normalization method can be used. The normalization process of this normalization method is a prior art and will not be repeated here.

[0053] S20, calculating the target feature score corresponding to each initial remote sensing image according to the initial feature scores corresponding to the plant size discreteness, color saturation and interval discreteness respectively.

[0054] Specifically, the target feature score Z corresponding to any initial remote sensing image meets the following conditions:

[0055] Z=(1-Z1)+Z2+(1-Z3), where Z1 is the plant size discreteness score corresponding to any initial remote sensing image, Z2 is the color saturation score corresponding to any initial remote sensing image, and Z3 is the interval discreteness score corresponding to any initial remote sensing image.

[0056] S30, determining the drone shooting height corresponding to the initial remote sensing image corresponding to the maximum target feature score as the first shooting height of the target plant.

[0057] In the above, the situation where there is no inflection point in the fitted curve is taken into consideration. At this time, each initial remote sensing image is analyzed separately. As the shooting height decreases, the edges of the captured plants will become clearer, so the actual size of the plants will become larger, and the spacing between the plants will become smaller. The two offset each other. Therefore, based on the remaining three plant image features, a score for each initial remote sensing image is obtained, and the first shooting height determined based on the score is more reasonable and reliable.

[0058] Furthermore, the method further comprises the following steps:

[0059] S001, according to the growth cycle of the crops planted in the target area, obtain a number of initial remote sensing images corresponding to the target area in each preset time period in the growth cycle when photographed from top to bottom, so as to obtain the first shooting height corresponding to each preset time period. It can be understood that: adopting steps S301-S304 to obtain the first shooting height corresponding to any preset time period, finally obtains the first shooting height corresponding to each preset time period.

[0060] Specifically, the preset time period refers to any time period preselected from the growth cycle of the planted crops, for example, a time period selected in the early growth period of the planted crops, wherein the time period can be one day, one hour, etc.

[0061] S002, according to a number of preset time periods in the growth cycle and the first shooting height corresponding to each preset time period, a relationship curve between the preset time period and the first shooting height is fitted to obtain a target shooting height corresponding to any random time period in the growth cycle.

[0062] In the above-mentioned embodiment, the drone shooting is only for a preset time period to obtain the corresponding first shooting height. Since the growth of crops during the growth cycle changes in real time, it is impossible to know the reasonable shooting heights corresponding to other time periods. By fitting the time period and the first shooting height, the reasonable shooting height corresponding to each random time period can be obtained, so that clearer plant images can be captured in each random time period.

[0063] S400: Determine an average shooting height of first shooting heights corresponding to a plurality of target areas as a final shooting height.

[0064] As mentioned above, since the divided target areas have the same area and size, there will be a gap between the first shooting heights corresponding to different target areas due to the influence of factors such as light, wind, and shooting angle. Since the gap is small, the average shooting height is determined as the final shooting height to provide a reference for the shooting height of other areas planted with the same target plants. This makes it possible to capture a larger area while ensuring the clarity of the captured plant image, which is conducive to clearly observing the growth of more plants.

[0065] An embodiment of the present invention also provides a non-transitory computer-readable storage medium, which can be set in an electronic device to store at least one instruction or at least one program related to implementing a method in a method embodiment. The at least one instruction or the at least one program is loaded and executed by the processor to implement the method for determining the crop shooting height provided in the above embodiment.

[0066] An embodiment of the present invention further provides an electronic device, comprising a processor and the aforementioned non-transitory computer-readable storage medium.

[0067] An embodiment of the present invention further provides a computer program product, which includes program code. When the program product is run on an electronic device, the program code is used to enable the electronic device to execute the steps of the method according to various exemplary embodiments of the present invention described above in this specification.

[0068] Although some specific embodiments of the present invention have been described in detail by way of example, it will be appreciated by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present invention. It will also be appreciated by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the present invention. The scope of the present invention is defined by the appended claims.

Claims

1. A method for determining a crop shooting height, characterized in that: The method comprises the following steps: Receive an initial shooting height corresponding to the target field; the initial shooting height refers to the lowest shooting height that can capture the entire target field; When the area of ​​the target field is greater than a preset area threshold, the number of field blocks after the target field is divided is calculated according to the area of ​​the target field and the preset area threshold, so as to evenly divide the target field into a number of target areas according to the number of field blocks; Based on a plurality of preset drone shooting heights, a plurality of initial remote sensing images corresponding to each target area are obtained, and based on the plurality of initial remote sensing images corresponding to each target area, a first shooting height corresponding to each target area is obtained; the plurality of drone shooting heights refer to a plurality of shooting heights set from top to bottom with the lowest shooting height as the upper limit height; An average shooting height of the first shooting heights corresponding to the plurality of target areas is determined as a final shooting height.

2. The method for determining the crop shooting height according to claim 1, characterized in that: When the area of ​​the target field is not greater than the preset area threshold, the initial shooting height is determined as the final shooting height.

3. The method for determining the crop shooting height according to claim 1, characterized in that: The number of target blocks meets the following conditions: E=roundup(C / C0), where E is the number of field blocks, C is the area of ​​the target field, C0 is a preset area threshold, and roundup() is an upward rounding function.

4. The method for determining the crop shooting height according to claim 1, characterized in that: The preset area threshold is also determined through the following steps: Receive some target field information; the target field information includes the actual area of ​​the target field, the key shooting height corresponding to the target field and the actual area of ​​the area photographed at the key shooting height; the key shooting height is the shooting height corresponding to the remote sensing image of the target field determined by the target user when it meets the preset requirements, and the types of crops planted in any two target fields are different; According to the key shooting height in each target field information, the average height is calculated, and the actual shooting area corresponding to the average height is determined as the preset area threshold.

5. The method for determining the crop shooting height according to claim 1, characterized in that: Get the first shooting height corresponding to any target area through the following steps: For any initial remote sensing image corresponding to any target area, a plant image feature set corresponding to target plants in the target area is obtained from the initial remote sensing image; the plant image feature set includes an average plant size, a plant size dispersion degree, a color saturation, and an average interval and an interval dispersion degree corresponding to the target plants in the initial remote sensing image; For a number of initial remote sensing images corresponding to any target area, the first image similarity corresponding to each two adjacent initial remote sensing images at the shooting heights of the UAV is obtained according to the plant image feature set corresponding to each initial remote sensing image; A curve is drawn according to the shooting order of the initial remote sensing images and the first image similarities corresponding to the initial remote sensing images at every two adjacent UAV shooting heights, and when there is an inflection point in the drawn curve, the first image similarity corresponding to the inflection point is determined as the second image similarity; With respect to the drone shooting heights of the two initial remote sensing images corresponding to the second image similarity, the larger drone shooting height is determined as the first shooting height of the target plant.

6. The method for determining the crop shooting height according to claim 5, characterized in that: Obtain the average plant size corresponding to the target plant by following the steps below: For any initial remote sensing image, according to the band reflectance corresponding to each target plant in the initial remote sensing image, the planting area corresponding to each target plant in the target area is identified from the initial remote sensing image; Obtaining the number of pixel points in the planting area corresponding to each target plant in the target area and the number of pixel points in the initial remote sensing image; According to the drone shooting height corresponding to the initial remote sensing image, the diagonal field of view of the drone and the size information of the initial remote sensing image, the actual area corresponding to the initial remote sensing image is calculated; According to the number of pixels in the planting area corresponding to each target plant in the target area, the number of pixels in the initial remote sensing image and the actual area corresponding to the initial remote sensing image, the plant size corresponding to each target plant is calculated to obtain the average plant size corresponding to the target plants.

7. The method for determining the crop shooting height according to claim 5, characterized in that: The method of obtaining the first image similarity corresponding to each of the initial remote sensing images at each adjacent UAV shooting height according to the plant image feature set corresponding to each initial remote sensing image comprises the following steps: For any two adjacent initial remote sensing images taken by UAVs at different shooting heights, obtain the feature similarity corresponding to each plant image feature; The feature similarities corresponding to the features of a number of plant images are added and calculated in sequence to obtain the first image similarity corresponding to the initial remote sensing images at two adjacent UAV shooting heights.

8. The method for determining the crop shooting height according to claim 5, characterized in that: The method further determines a first shooting height of the target plant through the following steps: When there is no inflection point in the drawn curve, for any plant image feature among plant size discreteness, color saturation and interval discreteness, several feature values ​​corresponding to the plant image feature are normalized to between 0 and 1, so as to obtain several initial feature scores corresponding to each initial remote sensing image; According to the initial feature scores corresponding to the plant size dispersion, color saturation and interval dispersion, the target feature score corresponding to each initial remote sensing image is calculated; The UAV shooting height corresponding to the initial remote sensing image corresponding to the maximum target feature score is determined as the first shooting height of the target plant.

9. A non-transitory computer-readable storage medium, wherein at least one instruction or at least one program is stored in the storage medium, characterized in that: The at least one instruction or the at least one program is loaded and executed by the processor to implement the method for determining the crop shooting height as described in any one of claims 1-8.

10. An electronic device, characterized in that: The invention comprises a processor and the non-transitory computer-readable storage medium as claimed in claim 9.

Images