Method, device, medium and equipment for evaluating emergence rate of soybean seedlings

Abstract

The invention discloses a soybean seedling emergence rate evaluation method and device, a medium and equipment, and relates to the field of crop emergence rate statistics, and the method comprises the steps: obtaining a remote sensing image of a soybean seedling in a to-be-detected region, and carrying out background removal processing to obtain a processed remote sensing image; detecting the processed remote sensing image, obtaining length information of bounding boxes, and obtaining union set intervals of the bounding boxes in a plurality of horizontal directions to determine the number of soybean seedlings; based on the length information of the bounding box, the size of a single soybean seedling with uneven growth vigor is adjusted by introducing an expansion factor and a shrinkage factor, the adjusted soybean seedlings are counted, and the number of overlapped soybean seedlings in the to-be-detected area is determined by analyzing the overlapping state of the soybean seedlings; and determining the number of soybean seedlings based on the union set interval of the bounding box, superposing the number of soybean seedlings with uneven growth vigor and the number of overlapped soybean seedlings, and determining an evaluation result of the emergence rate of soybean seedlings in the to-be-detected area.

Description

Technical Field

[0001] The invention relates to the field of crop emergence rate statistics, and in particular to a soybean seedling emergence rate evaluation method, device, medium and equipment. Background Art

[0002] Soybean emergence rate statistics can evaluate soybean variety quality, predict yield, guide field management, and optimize planting strategies. This information is crucial for farm managers, helping them make more scientific planting decisions and improve production efficiency. However, traditional crop emergence statistics mainly rely on manual observation, which is easily affected by subjective factors.

[0003] The current mainstream method is the target detection and image segmentation method based on deep learning. By using VGG16 as the Faster R-CNN pre-training network, corn seedlings with weeds and soil background at different growth stages in the field are detected. Although this method can detect the seedlings in the image, the detection effect of this method is not ideal when the image brightness is dark and there is a certain amount of shadow interference, which leads to the problem of inaccurate statistics of soybean seedling emergence rate. Summary of the invention

[0004] The present invention provides a soybean seedling emergence rate evaluation method, device, medium and equipment, which are used to solve the above-mentioned problem existing in the prior art, that is, how to improve the accuracy of soybean seedling emergence rate estimation in the prior art. The present invention provides a soybean seedling emergence rate evaluation method, which comprises:

[0005] Acquire a remote sensing image of soybean seedlings in the test area, perform background removal processing on the acquired remote sensing image, and obtain a processed remote sensing image;

[0006] The target detection algorithm is used to detect the remote sensing image after background removal, and the bounding boxes corresponding to multiple soybean seedlings are obtained, and the length information of the bounding boxes is obtained; the redundant intervals between adjacent bounding boxes are removed, and the union intervals of multiple bounding boxes in the horizontal direction are obtained, and the number of soybean seedlings is determined according to the union intervals of the bounding boxes;

[0007] Based on the length information of the bounding box, when there are soybean seedlings with uneven growth in the image, the size of the individual soybean seedlings with uneven growth is adjusted by introducing expansion factors and contraction factors, and the adjusted soybean seedlings are counted. When there are overlapping soybean seedlings in the image, the overlapping state of the soybean seedlings in the image is analyzed to determine the number of soybean seedlings in the test area under overlapping conditions.

[0008] The number of soybean seedlings was determined based on the union interval of the bounding box, and was superimposed with the number of soybean seedlings with uneven growth and the number of overlapping soybean seedlings to determine the evaluation result of the soybean seedling emergence rate in the test area.

[0009] Optionally, the size of individual soybean seedlings with uneven growth is adjusted by introducing an expansion factor and a contraction factor, and the adjusted soybean seedlings are counted, specifically including:

[0010] By introducing expansion factor and contraction factor, the unevenly growing individual soybean seedlings were adjusted to standard-growing soybean seedlings of uniform size. The number of soybean seedlings with uneven growth was obtained using the following formula:

[0011]

[0012] w ω =∑w α +∑ρ·w β +∑σ·w γ ,

[0013]

[0014]

[0015] Among them, k represents the number of soybean seedlings actually planted in the test area, W represents the sum of the horizontal widths of the planting pits corresponding to the actual soybean seedlings planted in the test area in the image, and w ω Represents the total horizontal width of the soybean seedling detection frame, w α Represents the horizontal width of the detection frame of each soybean seedling under standard growth, w β represents the horizontal width of the detection frame of each soybean seedling under weak growth, w γ represents the horizontal width of the detection frame of each soybean seedling under strong growth, ρ represents the expansion factor, and σ represents the contraction factor.

[0016] Optionally, the step of analyzing the overlapping state of the soybean seedlings in the image to obtain the number of soybean seedlings in the test area under overlapping conditions specifically includes:

[0017] The following formula is used to obtain the number of soybean seedlings in the overlapping condition:

[0018]

[0019]

[0020] Among them, Num2 represents the total number of overlapping seedlings in the test area, ω w Represents the total width of the detection frame of overlapping soybean seedlings in the test area, W nrepresents the sum of the horizontal widths of the planting pits corresponding to the actual soybean seedlings planted in each test area, and w0 represents the effective value of the width of the horizontal pixel of each standard planting pit in the test area.

[0021] Optionally, the background removal process of the acquired remote sensing image specifically includes:

[0022] The super green feature segmentation algorithm and OTSU algorithm are used to remove the image background.

[0023] Optionally, the use of a target detection algorithm to detect the remote sensing image after background removal, obtaining bounding boxes corresponding to a plurality of soybean seedlings, and obtaining length information of the bounding boxes specifically includes:

[0024] The YOLOv8n algorithm is used to detect the soybean seedling image after background removal, obtain the coordinate information of the bounding boxes corresponding to multiple soybean seedlings, and determine the length information of the bounding boxes.

[0025] The present invention provides a soybean seedling emergence rate evaluation device, comprising:

[0026] An acquisition module is used to acquire a remote sensing image of soybean seedlings in the test area, and to perform background removal processing on the acquired remote sensing image to obtain a processed remote sensing image;

[0027] The extraction module is used to detect the remote sensing image after background removal using a target detection algorithm, obtain bounding boxes corresponding to multiple soybean seedlings, and obtain length information of the bounding boxes; remove redundant intervals between adjacent bounding boxes, obtain the union interval of multiple bounding boxes in the horizontal direction, and determine the number of soybean seedlings according to the union interval of the bounding boxes;

[0028] A counting module is used to adjust the size of individual soybean seedlings with uneven growth by introducing expansion factors and contraction factors based on the length information of the bounding box when there are soybean seedlings with uneven growth in the image, and count the adjusted soybean seedlings; when there are overlapping soybean seedlings in the image, the overlapping state of the soybean seedlings in the image is analyzed to determine the number of soybean seedlings in the test area under overlapping conditions;

[0029] The determination module is used to determine the number of soybean seedlings based on the union interval of the bounding box, and superimpose it with the number of soybean seedlings with uneven growth and the number of overlapping soybean seedlings to determine the evaluation result of the soybean seedling emergence rate in the test area.

[0030] The present invention provides a computer-readable storage medium, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned soybean seedling emergence rate evaluation method is implemented.

[0031] The present invention provides a computer device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the above-mentioned soybean seedling emergence rate evaluation method is implemented when the processor executes the program.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention provides a method for evaluating the emergence rate of soybean seedlings, which obtains a processed remote sensing image by performing background removal processing on an acquired remote sensing image; uses a target detection algorithm to detect the remote sensing image after the background removal processing, obtains length information of bounding boxes corresponding to multiple soybean seedlings, and determines the number of soybean seedlings by selecting the union interval of the bounding boxes in the horizontal direction, thereby avoiding redundant markings and improving the counting accuracy; introduces an expansion factor and a contraction factor to adjust the size of individual soybean seedlings with uneven growth, and counts the adjusted soybean seedlings, analyzes the overlapping state of the soybean seedlings in the image, determines the number of soybean seedlings based on the union interval of the bounding boxes, and superimposes it with the number of soybean seedlings with uneven growth and the number of overlapping soybean seedlings, so as to determine the evaluation result of the emergence rate of soybean seedlings in the test area, thereby improving the accuracy of the estimation of the emergence rate of soybean seedlings. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0034] Figure 1 A flowchart of a soybean seedling emergence rate evaluation method provided by an embodiment of the present invention;

[0035] Figure 2 A schematic diagram comparing the data annotation effects before and after background removal provided by an embodiment of the present invention;

[0036] Figure 3 A schematic diagram of soybean seedlings with uneven growth provided by an embodiment of the present invention;

[0037] Figure 4 A schematic diagram of overlapping seedlings provided by an embodiment of the present invention;

[0038] Figure 5 A schematic diagram of overlapping marks provided by an embodiment of the present invention;

[0039] Figure 6 A schematic diagram of a computer device for a soybean seedling emergence rate assessment method provided in an embodiment of the present invention.

[0040] The above drawings have shown clear embodiments of the present invention, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present invention in any way, but to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0041] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0043] Figure 1 is a flow chart of a soybean seedling emergence rate evaluation method provided by an embodiment of the present invention, such as Figure 1 As shown, a soybean seedling emergence rate evaluation method shown in this embodiment includes:

[0044] S101: Acquire a remote sensing image of soybean seedlings in the test area, perform background removal processing on the acquired remote sensing image, and obtain a processed remote sensing image.

[0045] The drone equipment DJIM300, equipped with a PI camera, was used to collect remote sensing images of soybean seedlings in the test area.

[0046] Optionally, after acquiring the remote sensing images of the soybean seedlings in the test area, the acquired images are preprocessed, including batch cutting and background removal of the images.

[0047] For example, the soybean seedling remote sensing image can be horizontally adjusted and batch cut; the cut image can be subjected to background removal processing to obtain a processed remote sensing image, such as Figure 2 shown.

[0048] Optionally, a super green feature segmentation algorithm and an OTSU algorithm are used to remove the image background.

[0049] In this embodiment, the specific calculation formula of the green vegetation extraction color index is selected as follows:

[0050] ExG=0.441R-0.811G+0.385B+18.78745,

[0051] Where R, G, and B represent the RGB intensity values ​​of each pixel.

[0052] In this step, by performing background removal processing on the remote sensing image of the soybean seedlings, the interference of background information in the soybean seedlings image is reduced, thereby improving the accuracy of the soybean seedlings detection result.

[0053] For example, a mainstream median filter may be used to filter out spots in the soybean seedling image.

[0054] S102: Use a target detection algorithm to detect the remote sensing image after background removal, obtain bounding boxes corresponding to multiple soybean seedlings, and obtain length information of the bounding boxes; remove redundant intervals between adjacent bounding boxes, obtain the union interval of multiple bounding boxes in the horizontal direction, and determine the number of soybean seedlings according to the union interval of the bounding boxes.

[0055] Optionally, the YOLOv8n algorithm is used to detect the soybean seedling image after background removal, obtain coordinate information of bounding boxes corresponding to multiple soybean seedlings, and determine length information of the bounding boxes.

[0056] S103: Based on the length information of the bounding box, when there are soybean seedlings with uneven growth in the image, the size of the individual soybean seedlings with uneven growth is adjusted by introducing expansion factors and contraction factors, and the adjusted soybean seedlings are counted; when there are overlapping soybean seedlings in the image, the overlapping state of the soybean seedlings in the image is analyzed to determine the number of soybean seedlings in the overlapping state in the test area.

[0057] By simulating the growth of soybean seedlings in the test area under uneven growth conditions, such as Figure 3 As shown, the length units in the images are all pixels.

[0058] The number of soybean seedlings with uneven growth was obtained using the following formula:

[0059]

[0060] w ω =∑w α +∑ρ·w β +∑σ·w γ ,

[0061]

[0062]

[0063] Among them, k represents the number of soybean seedlings actually planted in the test area, W represents the sum of the horizontal widths of the planting pits corresponding to the actual soybean seedlings planted in the test area in the image, and w ω Represents the total horizontal width of the soybean seedling detection frame, w α Represents the horizontal width of the detection frame of each soybean seedling under standard growth, w β represents the horizontal width of the detection frame of each soybean seedling under weak growth, w γ represents the horizontal width of the detection frame of each soybean seedling under strong growth, ρ represents the expansion factor, and σ represents the contraction factor.

[0064] There may be a large number of overlapping seedlings in the image to be detected. This embodiment simulates the growth of soybean seedlings in the test area under overlapping conditions, which is divided into three critical situations, such as Figure 4 As shown, the length units in the images are all pixels.

[0065] After analysis, it was found that the key factor affecting the counting of overlapping soybean seedlings was the uneven growth of the seedlings on both sides. Under the premise of fixed planting points, the uneven growth of seedlings on both sides will make the horizontal width of the target detection frame different, which will lead to deviations in the calculation based on the detection frame side length information.

[0066] Through the analysis of the growth of soybean seedlings in the test area, it was found that the growth of soybean seedlings during this period was less than 1.25 times the standard growth rate. Where Δd1=w α -w0,w α represents the horizontal width of each soybean seedling actually detected; at the same time, the planting point of each soybean seedling is located in the center of each planting pit, then Where Δd2=w0-w α , w α Represents the horizontal width of each soybean seedling actually detected. Based on the above conditions, the following formula is used to obtain the number of soybean seedlings in the overlapping case:

[0067]

[0068]

[0069] Among them, Num2 represents the total number of overlapping seedlings in the test area, ω w Represents the total width of the detection frame of overlapping soybean seedlings in the test area, W n represents the sum of the horizontal widths of the planting pits corresponding to the actual soybean seedlings planted in each test area, and w0 represents the effective value of the width of the horizontal pixel of each standard planting pit in the test area.

[0070] S104: Determine the number of soybean seedlings based on the union interval of the bounding box, and superimpose it with the number of soybean seedlings with uneven growth and the number of overlapping soybean seedlings to determine an evaluation result of the soybean seedling emergence rate of the test area.

[0071] For example, during the soybean seedling detection process, there may be a phenomenon of label overlap, such as Figure 5 As shown, the union interval contained in the detection area in the horizontal direction can be used as effective information to obtain the final number of soybean seedlings statistics. The specific formula is as follows:

[0072] w ω =Box1(x l1 , x r1 )∪Box2(x l2 , x r2 )∪…∪Box i (x li , x ri ),

[0073]

[0074] Num=Num1+Num2+Num3.

[0075] Among them, Box i Represents the interval contained by each overlapping detection box, x li 、x ri Respectively represent the x-values ​​of the upper left corner coordinates and the lower right corner coordinates of different overlapping detection boxes, w ω ' is the union interval value contained in the overlapping detection frame, and Num represents the final number of soybean seedlings in the test area.

[0076] The above is a soybean seedling emergence rate evaluation method provided by one or more embodiments of this specification. Based on the same idea, this embodiment provides a soybean seedling emergence rate evaluation device, including:

[0077] An acquisition module is used to acquire a remote sensing image of soybean seedlings in the test area, and to perform background removal processing on the acquired remote sensing image to obtain a processed remote sensing image;

[0078] The extraction module is used to detect the remote sensing image after background removal using a target detection algorithm, obtain bounding boxes corresponding to multiple soybean seedlings, and obtain length information of the bounding boxes; remove redundant intervals between adjacent bounding boxes, obtain the union interval of multiple bounding boxes in the horizontal direction, and determine the number of soybean seedlings according to the union interval of the bounding boxes;

[0079] A counting module is used to adjust the size of individual soybean seedlings with uneven growth by introducing expansion factors and contraction factors based on the length information of the bounding box when there are soybean seedlings with uneven growth in the image, and count the adjusted soybean seedlings; when there are overlapping soybean seedlings in the image, the overlapping state of the soybean seedlings in the image is analyzed to determine the number of soybean seedlings in the test area under overlapping conditions;

[0080] The determination module is used to determine the number of soybean seedlings based on the union interval of the bounding box, and superimpose it with the number of soybean seedlings with uneven growth and the number of overlapping soybean seedlings to determine the evaluation result of the soybean seedling emergence rate in the test area.

[0081] This embodiment provides a computer-readable storage medium, which stores a computer program. The computer program can be used to execute the soybean seedling emergence rate evaluation method provided above.

[0082] This embodiment provides Figure 6 The structural diagram of the computer device shown in FIG. Figure 6 As shown, at the hardware level, the computer device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and may also include hardware required for other services. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to implement the soybean seedling emergence rate evaluation method provided in the above embodiment.

[0083] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0084] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of the present invention.

Claims

1. A method for evaluating soybean seedling emergence rate, characterized in that: include: Acquire a remote sensing image of soybean seedlings in the test area, perform background removal processing on the acquired remote sensing image, and obtain a processed remote sensing image; The target detection algorithm is used to detect the remote sensing image after background removal, and the bounding boxes corresponding to multiple soybean seedlings are obtained, and the length information of the bounding boxes is obtained; the redundant intervals between adjacent bounding boxes are removed, and the union intervals of multiple bounding boxes in the horizontal direction are obtained, and the number of soybean seedlings is determined according to the union intervals of the bounding boxes; Based on the length information of the bounding box, when there are soybean seedlings with uneven growth in the image, the size of the individual soybean seedlings with uneven growth is adjusted by introducing expansion factors and contraction factors, and the adjusted soybean seedlings are counted. When there are overlapping soybean seedlings in the image, the overlapping state of the soybean seedlings in the image is analyzed to determine the number of soybean seedlings in the test area under overlapping conditions. The number of soybean seedlings was determined based on the union interval of the bounding box, and was superimposed with the number of soybean seedlings with uneven growth and the number of overlapping soybean seedlings to determine the evaluation result of the soybean seedling emergence rate in the test area.

2. The method for evaluating soybean seedling emergence rate according to claim 1, wherein: The method of adjusting the size of individual soybean seedlings with uneven growth by introducing an expansion factor and a contraction factor, and counting the adjusted soybean seedlings, specifically includes: By introducing expansion factor and contraction factor, the unevenly growing individual soybean seedlings were adjusted to standard-growing soybean seedlings of uniform size. The number of soybean seedlings with uneven growth was obtained using the following formula: Among them, k represents the number of soybean seedlings actually planted in the test area, W represents the sum of the horizontal widths of the planting pits corresponding to the actual soybean seedlings planted in the test area in the image, and w ω Represents the total horizontal width of the soybean seedling detection frame, w α Represents the horizontal width of the detection frame of each soybean seedling under standard growth, w β represents the horizontal width of the detection frame of each soybean seedling under weak growth, w γ represents the horizontal width of the detection frame of each soybean seedling under strong growth, ρ represents the expansion factor, and σ represents the contraction factor.

3. The method for evaluating soybean seedling emergence rate according to claim 1, wherein: The method of analyzing the overlapping state of the soybean seedlings in the image to obtain the number of soybean seedlings in the test area under overlapping conditions specifically includes: The following formula is used to obtain the number of soybean seedlings in the overlapping condition: Among them, Num2 represents the total number of overlapping seedlings in the test area, ω w Represents the total width of the detection frame of overlapping soybean seedlings in the test area, W n represents the sum of the horizontal widths of the planting pits corresponding to the actual soybean seedlings planted in each test area, and w0 represents the effective value of the width of the horizontal pixel of each standard planting pit in the test area.

4. The method for evaluating soybean seedling emergence rate according to claim 1, wherein: The background removal process of the acquired remote sensing image specifically includes: The super green feature segmentation algorithm and OTSU algorithm are used to remove the image background.

5. The method for evaluating soybean seedling emergence rate according to claim 1, wherein: The target detection algorithm is used to detect the remote sensing image after background removal, and the bounding boxes corresponding to the plurality of soybean seedlings are obtained, and the length information of the bounding boxes is obtained, which specifically includes: The YOLOv8n algorithm is used to detect the soybean seedling image after background removal, obtain the coordinate information of the bounding boxes corresponding to multiple soybean seedlings, and determine the length information of the bounding boxes.

6. A soybean seedling emergence rate evaluation device, characterized in that: include: An acquisition module is used to acquire a remote sensing image of soybean seedlings in the test area, and to perform background removal processing on the acquired remote sensing image to obtain a processed remote sensing image; The extraction module is used to detect the remote sensing image after background removal using a target detection algorithm, obtain bounding boxes corresponding to multiple soybean seedlings, and obtain length information of the bounding boxes; remove redundant intervals between adjacent bounding boxes, obtain the union interval of multiple bounding boxes in the horizontal direction, and determine the number of soybean seedlings according to the union interval of the bounding boxes; A counting module is used to adjust the size of individual soybean seedlings with uneven growth by introducing expansion factors and contraction factors based on the length information of the bounding box when there are soybean seedlings with uneven growth in the image, and count the adjusted soybean seedlings; when there are overlapping soybean seedlings in the image, the overlapping state of the soybean seedlings in the image is analyzed to determine the number of soybean seedlings in the test area under overlapping conditions; The determination module is used to determine the number of soybean seedlings based on the union interval of the bounding box, and superimpose it with the number of soybean seedlings with uneven growth and the number of overlapping soybean seedlings to determine the evaluation result of the soybean seedling emergence rate in the test area.

7. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method for evaluating the emergence rate of soybean seedlings according to any one of claims 1 to 5 is implemented.

8. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for evaluating the emergence rate of soybean seedlings according to any one of claims 1 to 5 is implemented.

Images