Method and system for obtaining corn seedling condition and growth vigor

Abstract

The present application relates to a kind of corn seedling situation acquisition method and system, comprising: obtaining planting area image data, target area is obtained by identifying marker through deep learning model, and corn plant is detected, obtains multiple phenotypic parameters of corn plant;According to marker and plot ridge direction, the target area is cut, and the image of each row plot and single plot length are obtained, and each row plot image is identified, and the coordinate of single plant image is obtained;According to single plant coordinate and plot number, the width of single plot is obtained, and the emergence number of each plot is obtained;According to the longitudinal coordinate difference of adjacent plant, each seedling spacing is calculated, and the width and length of plant are determined according to the longitudinal coordinate difference of single plant, and then the plant area and crown diameter are obtained;Significance analysis is carried out to the merged data, and the excel file of significance analysis, the probability density diagram and heat map of plot seedling number character are obtained.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of agricultural information technology, in particular to a corn seedling condition and seedling vigor acquisition method and system based on unmanned aerial vehicle aerial image machine learning. BACKGROUND

[0002] The corn seedling stage is a key period for determining corn yield. Timely understanding of phenotypic parameters such as seedling number, seedling area, and seedling spacing can control various risks in the planting process. Traditional seedling condition and seedling vigor monitoring mainly relies on manual work, but manual judgment is subjective, data collection cannot guarantee the accuracy of information, data cannot be checked, and manual work is high in cost, low in efficiency, and limited in observation range, which greatly affects subsequent analysis. SUMMARY

[0003] To solve the above problems, the present application provides a corn seedling condition and seedling vigor acquisition method and system based on unmanned aerial vehicle aerial image machine learning, which can collect data in a large area and realize standardization analysis and management of corn seedling phenotypic parameters.

[0004] To achieve the above-mentioned purpose, the present application adopts the following technical scheme: a corn seedling condition and seedling vigor acquisition method based on unmanned aerial vehicle aerial image machine learning, comprising: acquiring planting area image data, identifying a target area by a deep learning model, detecting corn plants, and acquiring a plurality of phenotypic parameters of the corn plants; dividing the target area according to the markers and the direction of the plots to obtain each row of plot images and the length of a single plot, and identifying each row of plot images to obtain the coordinates of a single plant image; obtaining the width of a single plot according to the coordinates of a single plant and the number of plots, and obtaining the seedling number of each plot; calculating each seedling spacing according to the difference in the longitudinal coordinates of adjacent plants, determining the width and length of the plants according to the difference in the longitudinal and transverse coordinates of a single plant, and then obtaining the plant area and crown diameter; merging the seedling condition and seedling vigor data table obtained according to the seedling number, plant area, and crown diameter with the planting plan table according to the row number and column number, and performing significance analysis on the merged data to obtain an excel file of significance analysis, a probability density graph of plot seedling number traits, and a heat map.

[0005] Further, the target area is obtained by identifying the markers by deep learning, and the corn plants are detected, comprising:

[0006] The markers and the corn plants are divided into two categories and labeled, input into a deep learning model for training, and a trained model is obtained;

[0007] The planting area image data is processed based on the trained model, and the markers and the corn plants can be identified respectively to obtain the target area and the plants in the target area.

[0008] Further, the single plot length is:

[0009] Based on the coordinates of the markers, every four marker coordinates determine a row of plots to be measured, and the plots to be measured are cut out from the planting area image through the four coordinates. The difference in the longitudinal coordinates of the cut-out area is the single plot length.

[0010] Further, according to the coordinates of the single plant and the number of plots, the single plot width is obtained, and the emergence number of each plot is obtained, including:

[0011] The difference in the transverse coordinates of the markers is the width of the area to be measured. According to the planting specifications, if it is not wide and narrow ridge, the plot width is: 4 / 5 x (width of the area to be measured / number of plots); if it is wide and narrow ridge, the plot width is: (width of the area to be measured / number of plots)-30, and all plots are obtained according to the ridge direction in turn.

[0012] According to the plot coordinates and the plant coordinates, it is determined how many plants are in the same plot, and the emergence number is obtained.

[0013] Further, the plant area and the crown diameter are obtained, including:

[0014] The width of the plant is the difference in the transverse coordinates w, and the length of the plant is the difference in the longitudinal coordinates h. The plant area is: w*h.

[0015] The orientation of the leaves has no fixed direction, and the longer side is selected as the crown diameter: max(w,h).

[0016] Further, the seedling and seedling trend data table includes: emergence rate, average seedling spacing, seedling spacing variance, seedling area ratio, average leaf length, leaf length variance, and seeding density.

[0017] Further, the probability density map and the heat map of the significant analysis of the plot seedling number trait are obtained, including:

[0018] The probability density map drawing function is:

[0019] sns.kdeplot(data=data,shade=True,vertical=False,cut=0,color='r',alpha=0.5,linewidth=0.5,linestyle='-')

[0020] Wherein, data is the parameter in the seedling and seedling trend data table; shade is the shadow; vertical is drawn with the y-axis; cut is the cut-off bandwidth to the limit value of the axis; color is the drawing color; alpha is the color saturation of the color; linewidth is the line size; linestyle is the straight line; sns.kdeplot() is the probability density map drawing function;

[0021] The function for drawing a heatmap is:

[0022] sns.heatmap(data,annot=False,cmap='coolwarm')

[0023] Here, `data` represents the parameters read from the seedling condition data table; `annot` indicates that no data value should be written to each heatmap cell; `cmap` is the mapping from data values ​​to the color space; and `sns.heatmap()` is the function for drawing the heatmap.

[0024] A corn seedling condition and growth status acquisition system based on machine learning from UAV aerial images includes: a first processing module, which acquires image data of the planting area, identifies target areas by recognizing markers through a deep learning model, detects corn plants, and acquires multiple phenotypic parameters of the corn plants; a second processing module, which segments the target area according to the markers and the row orientation of the plots to obtain images of each row of plots and the length of each plot, and identifies the coordinates of each individual plant image; a third processing module, which obtains the width of each plot and the number of seedlings in each plot based on the coordinates of the individual plant and the number of plots; a fourth processing module, which calculates the spacing between each seedling based on the difference in the ordinate of adjacent plants, and determines the width and length of the plant based on the difference in the ordinate and abordinate of the individual plant, thereby obtaining the plant area and crown diameter; and a significance analysis module, which merges the seedling condition and growth status data table obtained from the number of seedlings, plant area, and crown diameter with the planting plan table according to the row and column numbers, performs significance analysis on the merged data, and obtains an Excel file of significance analysis, a probability density map of the seedling number trait in each plot, and a heat map.

[0025] A computer-readable storage medium storing one or more programs, the one or more programs including instructions that, when executed by a computing device, cause the computing device to perform any of the methods described above.

[0026] A computing device includes: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include instructions for performing any of the methods described above.

[0027] The present invention has the following advantages due to the adoption of the above technical solutions:

[0028] 1. This invention uses an RGB camera mounted on a drone to collect images of seedlings in the field, which can acquire a large amount of effective image data in a short time. The data can be stored for a long time, avoiding the impact of loss of analysis results.

[0029] 2. This invention uses artificial intelligence analysis software to analyze images, which can obtain multiple phenotypic parameters of seedling condition and growth in a short time, saving time.

[0030] 3. This invention reduces the requirements for staff's professional knowledge and experience, and streamlines the operation process.

[0031] 4. This invention enables standardized management of seedling conditions and growth, avoiding the differential effects caused by subjective human judgment.

[0032] 5. This invention can analyze seedling conditions and growth at multiple locations, offering high flexibility.

[0033] 6. This invention can analyze multiple tasks simultaneously, resulting in high efficiency. Attached Figure Description

[0034] Figure 1 This is a schematic diagram of the corn seedling condition and growth status system collection and analysis in an embodiment of the present invention. Detailed Implementation

[0035] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0036] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0037] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0038] This invention proposes a method and system for acquiring maize seedling condition and vigor based on machine learning from UAV aerial images. The UAV captures aerial images, and a machine learning algorithm identifies markers to obtain target areas and detects maize plants, acquiring multiple phenotypic parameters of the plants. The plots are divided according to the row direction to obtain images of each row and the length of each individual plot. The coordinates of each individual plant are obtained by identifying the images of each row. The width of each individual plot is obtained based on the plant coordinates and the number of plots, and the number of seedlings in each plot is counted. The spacing between each seedling can be calculated based on the difference in the ordinate of adjacent plants. The width and length of each individual plant are determined by the difference in the ordinate and transverse coordinates, thus obtaining the plant area and crown diameter. The planting plan table and the analyzed data are merged according to row and column numbers, and a significance analysis is performed to obtain an Excel file of the significance analysis, probability density maps of traits such as the number of seedlings in each plot, and heat maps. This invention can acquire multiple phenotypic parameters of maize, analyze and judge the condition and vigor of maize seedlings, can be used in multiple locations, and offers high throughput, high timeliness, and low cost.

[0039] In one embodiment of the present invention, a method for acquiring corn seedling condition and vigor based on machine learning from drone aerial images is provided. In this embodiment, as... Figure 1 As shown, the method includes the following steps:

[0040] 1) Obtain image data of the planting area, identify the target area by using a deep learning model to identify markers, detect corn plants, and obtain multiple phenotypic parameters of the corn plants;

[0041] 2) The target area is segmented according to the markers and the ridge direction of the plots to obtain the image of each row of plots and the length of a single plot. The image of each row of plots is then identified to obtain the coordinates of the single plant image.

[0042] 3) Based on the coordinates of a single plant and the number of plots, obtain the width of a single plot and the number of seedlings emerging in each plot;

[0043] 4) Calculate the spacing between each seedling based on the difference in the vertical coordinates of adjacent plants, and determine the width and length of the plant based on the difference in the horizontal and vertical coordinates of a single plant, thereby obtaining the plant area and crown diameter.

[0044] 5) Merge the seedling condition data table obtained based on the number of seedlings, plant area and crown diameter with the planting plan table according to the row and column numbers. Perform significance analysis on the merged data to obtain an Excel file of significance analysis, a probability density map of the seedling number traits in the plot and a heat map.

[0045] In this embodiment, the seedling condition data table obtained based on the number of seedlings, plant area, and crown diameter includes: seedling emergence rate, mean spacing between seedlings, variance of spacing between seedlings, percentage of seedling area, mean leaf length, variance of leaf length, and sowing density, etc.

[0046] In step 1) above, obtaining image data of the planting area specifically involves: collecting images of the planting area through drone aerial photography.

[0047] In this embodiment, the flight path was defined, and drone parameters such as resolution, overlap, and shooting interval were set. Shooting was conducted during the corn seedling stage, with wind speeds below level 4 in the aerial photography environment. Data was transmitted via SD card, and the images captured by the drone were stitched together to obtain a complete aerial image of the area.

[0048] In step 1) above, the target region is obtained by recognizing markers through deep learning, and corn plants are detected, including the following steps:

[0049] 1.1) The markers and corn plants are divided into two categories and labeled, and then input into the deep learning model for training to obtain the trained model;

[0050] 1.2) Based on the trained model, the image data of the planting area is processed to identify the markers and corn plants, and the target area and the plants in the target area are obtained.

[0051] In step 2) above, the length of a single cell is: based on the coordinates of the markers, a row of cells to be tested is determined for every four marker coordinates. The cells to be tested are segmented from the planting area image using four coordinates, and the difference in the vertical coordinates of the segmented areas is the length of a single cell.

[0052] In step 3) above, the width of a single plot is obtained based on the coordinates of the individual plant and the number of plots, and the number of seedlings emerging in each plot is obtained, including the following steps:

[0053] 3.1) The difference in the abscissa of the markers is the width of the area to be measured. According to the planting specifications, if it is not a wide-narrow ridge, the cell width is: 4 / 5 × (width of the area to be measured / number of cells); if it is a wide-narrow ridge, the cell width is: (width of the area to be measured / number of cells) - 30. All cells are obtained in sequence according to the ridge direction.

[0054] 3.2) Based on the coordinates of the plot and the plant, determine how many plants are in the same plot and obtain the number of seedlings.

[0055] In step 4) above, all plants in the plot are sorted by coordinates, and the difference in the ordinate of the center point of adjacent plants is the seedling spacing.

[0056] In step 4) above, the plant area and crown diameter are obtained, specifically as follows:

[0057] 4.1) The width of the plant is the difference in the horizontal coordinate w, the length of the plant is the difference in the vertical coordinate h, and the area of ​​the plant is: w*h;

[0058] 4.2) The direction of the leaves is not fixed. The longer side is selected as the crown diameter: max(w,h).

[0059] In step 5) above, the planting plan table and the analyzed data are merged according to the row number and column number. Multiple parameters such as the number of seedlings in each plot are read from the seedling condition data table, and a significance analysis t-test is performed.

[0060] In step 5) above, the probability density map and heat map of the seedling number trait in the plot for significance analysis are obtained, including:

[0061] 5.1) The function for plotting the probability density plot is:

[0062] sns.kdeplot(data=data,shade=True,vertical=False,cut=0,color='r',alpha=0.5,linewidth=0.5,linestyle='-')

[0063] Wherein, data is the parameter read from the seedling condition data table; shade is the shade; vertical is plotted on the y-axis; cut is the cut-off bandwidth to the limit value of the number axis; color is the plotting color; alpha is the color saturation; linewidth is the line size; linestyle is the line style; and sns.kdeplot() is the probability density plotting function.

[0064] 5.2) The function for drawing a heatmap is:

[0065] sns.heatmap(data,annot=False,cmap='coolwarm')

[0066] Here, `data` represents the parameters read from the seedling condition data table; `annot` indicates that no data value should be written to each heatmap cell; `cmap` is the mapping from data values ​​to the color space; and `sns.heatmap()` is the function for drawing the heatmap.

[0067] In summary, when using this invention, the drone plans its flight path, adjusts various parameters, and photographs corn seedlings, then stitches the photographed images together. The stitched images are placed in a designated location, and the seedling condition analysis system is activated. The results, including parameters such as the number of seedlings and the spacing between seedlings, are generated into an Excel file for easy viewing. The planting plan table and the analyzed data are merged according to the row and column numbers, and a significance analysis is performed to obtain an Excel file of the significance analysis, a probability density map of traits such as the number of seedlings in each plot, and a heat map.

[0068] In one embodiment of the present invention, a corn seedling condition and growth status acquisition system based on machine learning from UAV aerial images is provided, comprising:

[0069] The first processing module acquires image data of the planting area, identifies the target area by recognizing markers through a deep learning model, and detects corn plants to obtain multiple phenotypic parameters of the corn plants.

[0070] The second processing module segments the target area according to the markers and the ridge direction of the plots to obtain the image of each row of plots and the length of a single plot. It also identifies each row of plots to obtain the coordinates of a single plant image.

[0071] The third processing module obtains the width of a single cell based on the coordinates of a single plant and the number of cells, and then obtains the number of seedlings emerging in each cell.

[0072] The fourth processing module calculates the spacing between each seedling based on the difference in the vertical coordinates of adjacent plants, and determines the width and length of a plant based on the difference in the horizontal and vertical coordinates of a single plant, thereby obtaining the plant area and crown diameter.

[0073] The significance analysis module merges the seedling condition data table obtained from the number of seedlings, plant area, and crown diameter with the planting plan table according to the row and column numbers. It then performs significance analysis on the merged data to obtain an Excel file of significance analysis, a probability density map of the seedling number traits in the plot, and a heat map.

[0074] In the above embodiments, the first processing module obtains the target region by recognizing markers through deep learning and detects corn plants, including:

[0075] The markers and corn plants were divided into two categories and labeled. The labels were then fed into a deep learning model for training, resulting in a trained model.

[0076] The trained model processes the image data of the planting area, which can identify the markers and corn plants, and obtain the target area and the plants within the target area.

[0077] In the above embodiment, the length of a single cell is: based on the coordinates of the markers, a row of cells to be tested is determined for every four marker coordinates. The cells to be tested are segmented from the planting area image using four coordinates, and the difference in the vertical coordinates of the segmented areas is the length of a single cell.

[0078] In the above embodiments, the width of a single cell is obtained based on the coordinates of a single plant and the number of cells, and the number of seedlings emerging in each cell is obtained, including:

[0079] The difference in the horizontal coordinates of the markers represents the width of the area to be measured. According to the planting specifications, if it is not a wide-narrow ridge, the cell width is: 4 / 5 × (width of the area to be measured / number of cells); if it is a wide-narrow ridge, the cell width is: (width of the area to be measured / number of cells) - 30. All cells are obtained sequentially according to the ridge direction.

[0080] Based on the coordinates of the plot and the plant, the number of plants in the same plot can be determined, and the number of seedlings can be obtained.

[0081] In the above embodiments, the plant area and crown diameter are obtained, including:

[0082] The width of the plant is the difference in the horizontal coordinate w, the length of the plant is the difference in the vertical coordinate h, and the area of ​​the plant is: w*h;

[0083] The direction of the leaves is not fixed; the longer side is selected as the crown diameter: max(w,h).

[0084] In the above embodiments, the probability density map and heat map of the seedling number trait in the plot obtained by significance analysis include:

[0085] The function for plotting probability density maps is:

[0086] sns.kdeplot(data=data,shade=True,vertical=False,cut=0,color='r',alpha=0.5,linewidth=0.5,linestyle='-')

[0087] Where data is the parameter read from the seedling condition data table; shade is the shade; vertical is plotted on the y-axis; cut is the cut-off bandwidth to the limit of the number axis; color is the plotting color; alpha is the color saturation; linewidth is the line size; linestyle is a straight line; and sns.kdeplot() is the probability density plotting function.

[0088] The function for drawing a heatmap is:

[0089] sns.heatmap(data,annot=False,cmap='coolwarm')

[0090] Here, `data` represents the parameters read from the seedling condition data table; `annot` indicates that no data value should be written to each heatmap cell; `cmap` is the mapping from data values ​​to the color space; and `sns.heatmap()` is the function for drawing the heatmap.

[0091] The system provided in this embodiment is used to execute the above-described method embodiments. For specific processes and details, please refer to the above embodiments, which will not be repeated here.

[0092] In one embodiment of the present invention, a computing device structure is provided. This computing device can be a terminal, which may include: a processor, a communication interface, memory, a display screen, and an input device. The processor, communication interface, and memory communicate with each other via a communication bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs, which are executed by the processor to implement the aforementioned methods. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface is used for wired or wireless communication with external terminals. Wireless communication can be achieved through Wi-Fi, a management network, NFC (Near Field Communication), or other technologies. The display screen can be a liquid crystal display or an e-ink display. The input device can be a touch layer covering the display screen, or buttons, a trackball, or a touchpad mounted on the casing of the computing device, or an external keyboard, touchpad, or mouse. The processor can call logical instructions stored in the memory.

[0093] Furthermore, the logical instructions in the aforementioned memory can be implemented as software functional units and sold or used as independent products, and can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0094] In one embodiment of the present invention, a computer program product is provided, the computer program product including a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, and when the program instructions are executed by a computer, the computer is able to perform the methods provided in the above-described method embodiments.

[0095] In one embodiment of the present invention, a non-transitory computer-readable storage medium is provided, which stores server instructions that cause a computer to perform the methods provided in the above embodiments.

[0096] The computer-readable storage medium provided in the above embodiments has a similar implementation principle and technical effect to the above method embodiments, and will not be described again here.

[0097] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0098] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0099] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for obtaining corn seedling condition and vigor, characterized in that, include: Acquire image data of the planting area, identify markers to obtain the target area through a deep learning model, detect corn plants, and obtain multiple phenotypic parameters of corn plants; The target area is segmented based on the markers and the ridge direction of the plots to obtain the image of each row of plots and the length of each individual plot. The coordinates of the individual plant image are then obtained by identifying each row of plot images. Based on the coordinates of a single plant and the number of plots, the width of a single plot is obtained, and the number of seedlings emerging in each plot is acquired. The spacing between each seedling is calculated based on the difference in the vertical coordinates of adjacent plants, and the width and length of a plant are determined based on the difference in the horizontal and vertical coordinates of a single plant, thereby obtaining the plant area and crown diameter. The seedling condition data table obtained based on the number of seedlings, plant area, and crown diameter was merged with the planting plan table according to the row and column numbers. The merged data was then subjected to significance analysis to obtain an Excel file of significance analysis, a probability density map of the seedling number traits in the plot, and a heat map. Based on the coordinates of a single plant and the number of plots, the width of a single plot is obtained, and the number of seedlings emerging in each plot is calculated, including: The difference in the horizontal coordinates of the markers represents the width of the area to be measured. According to the planting specifications, if it is not a wide-narrow ridge, the cell width is: 4 / 5 × (width of the area to be measured / number of cells); if it is a wide-narrow ridge, the cell width is: (width of the area to be measured / number of cells) - 30. All cells are obtained sequentially according to the ridge direction. Based on the coordinates of the plot and the plant, determine how many plants are in the same plot and obtain the number of seedlings; The probability density plots and heatmaps of the seedling number trait in the plots obtained from the significance analysis include: The function for plotting probability density maps is: sns.kdeplot(data=data,shade=True, vertical=False, cut = 0,color='r' ,alpha=0.5 , linewidth=0.5,linestyle='-' ) Where, data is the parameter read from the seedling condition data table; shade is the shade; vertical is plotted on the y-axis; cut is the cut-off bandwidth to the limit value of the number axis; color is the plotting color; alpha is the color saturation; linewidth is the line size; linestyle is the line style; sns.kdeplot() is the probability density plot plotting function; The function for drawing a heatmap is: sns.heatmap(data,annot=False,cmap='coolwarm') Here, `data` represents the parameters read from the seedling condition data table; `annot` means not to write data values ​​in each heatmap cell; `cmap` is the mapping from data values ​​to the color space; and `sns.heatmap()` is the function for drawing the heatmap.

2. The method for obtaining corn seedling condition and vigor as described in claim 1, characterized in that, The target region is obtained by recognizing markers using deep learning, and corn plants are detected, including: The markers and corn plants were divided into two categories and labeled. The labels were then fed into a deep learning model for training, resulting in a trained model. The trained model processes the image data of the planting area, which can identify the markers and corn plants, and obtain the target area and the plants within the target area.

3. The method for obtaining corn seedling condition and vigor as described in claim 1, characterized in that, The length of a single cell is: Based on the coordinates of the markers, a row of test plots is determined for every four marker coordinates. The test plots are then segmented from the planting area image using the four coordinates, and the difference in the vertical coordinates of the segmented areas is the length of a single plot.

4. The method for obtaining corn seedling condition and vigor as described in claim 1, characterized in that, The plant area and crown diameter are obtained, including: The width of the plant is the difference in the horizontal coordinate w, the length of the plant is the difference in the vertical coordinate h, and the area of ​​the plant is: w*h; The direction of the leaves is not fixed; the longer side is selected as the crown diameter: max(w,h).

5. The method for obtaining corn seedling condition and vigor as described in claim 1, characterized in that, The seedling condition data table includes: emergence rate, mean spacing between seedlings, variance of spacing between seedlings, percentage of seedling area, mean leaf length, variance of leaf length, and sowing density.

6. A corn seedling condition and vigor acquisition system, used to implement the corn seedling condition and vigor acquisition method as described in any one of claims 1 to 5, characterized in that, include: The first processing module acquires image data of the planting area, identifies the target area by recognizing markers through a deep learning model, and detects corn plants to obtain multiple phenotypic parameters of the corn plants. The second processing module segments the target area according to the markers and the ridge direction of the plots to obtain the image of each row of plots and the length of a single plot. It also identifies each row of plots to obtain the coordinates of a single plant image. The third processing module obtains the width of a single cell based on the coordinates of a single plant and the number of cells, and then obtains the number of seedlings emerging in each cell. The fourth processing module calculates the spacing between each seedling based on the difference in the vertical coordinates of adjacent plants, and determines the width and length of a plant based on the difference in the horizontal and vertical coordinates of a single plant, thereby obtaining the plant area and crown diameter. The significance analysis module merges the seedling condition data table obtained from the number of seedlings, plant area, and crown diameter with the planting plan table according to the row and column numbers. It then performs significance analysis on the merged data to obtain an Excel file of significance analysis, a probability density map of the seedling number traits in the plot, and a heat map.

7. A computer-readable storage medium for storing one or more programs, characterized in that, The one or more programs include instructions that, when executed by a computing device, cause the computing device to perform any of the methods described in claims 1 to 5.

8. A computing device, characterized in that, include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs including instructions for performing any of the methods described in claims 1 to 5.

Images