Seedling identification and graded seedling thinning and transplanting method and device

Abstract

The embodiment of the present invention provides a seedling identification and graded seedling thinning and transplanting method and device, including: based on the identification signal, obtaining a combined image of the seedling in the horizontal direction and the vertical direction; inputting the combined image into the identification model to obtain the characteristic image of the seedling; the characteristic image of the seedling is a matrix characteristic image, a stem characteristic image and a leaf characteristic image; performing image processing on the characteristic image of the seedling to determine the stem thickness parameter, the seedling height parameter and the leaf area parameter; based on the stem thickness parameter, the seedling height parameter and the leaf area parameter, determining the grade of the seedling by the threshold method; based on the grade of the seedling, performing seedling thinning and transplanting according to the preset transplanting rules. This embodiment can accurately complete the identification and graded transplanting tasks in a non-structured light environment.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a seedling identification and graded seedling thinning and transplanting method and device. Background Art

[0002] At present, in the seedling raising stage, due to problems such as uneven light and nutrition or seed defects, the seedlings fail to germinate or grow and develop unevenly. A batch of seedlings have very different stem thickness, seedling height, and leaf area or are missing. In addition, the grown plants are densely spaced and severely blocked by leaves, affecting the subsequent sale and cultivation of the seedlings. Summary of the invention

[0003] In view of the problems existing in the prior art, an embodiment of the present invention provides a seedling identification and graded seedling thinning and transplanting method and device.

[0004] In a first aspect, an embodiment of the present invention provides a seedling identification and graded seedling thinning and transplanting method, comprising:

[0005] Based on the recognition signal, a combined image of the seedling in the horizontal direction and the vertical direction is obtained;

[0006] Inputting the combined image into a recognition model to obtain a characteristic image of the seedling; the characteristic image of the seedling is a matrix characteristic image, a stem characteristic image and a leaf characteristic image;

[0007] Processing the characteristic image of the seedling to determine the stem thickness parameter, seedling height parameter and leaf area parameter;

[0008] Based on the stem diameter parameter, the seedling height parameter and the leaf area parameter, determining the grade of the seedling by a threshold method;

[0009] Thinning and transplanting the seedlings according to preset transplanting rules based on the grade of the seedlings;

[0010] Among them, the recognition model uses sample combined images of seedlings in the horizontal and vertical directions as input data, and uses feature images of seedlings corresponding to the sample combined images of seedlings in the horizontal and vertical directions as output data, and is trained based on a machine learning algorithm.

[0011] Furthermore, the method further comprises:

[0012] Based on the combined image of the seedlings in the horizontal direction and the vertical direction, an RGB matrix corresponding to the combined image is obtained;

[0013] Inputting the RGB matrix corresponding to the combined image into the recognition model to obtain an RGB value matrix;

[0014] The RGB value matrix is ​​converted into a feature image containing the seedling background.

[0015] Furthermore, the step of determining the grade of the seedlings by a threshold method based on the stem diameter parameter, the seedling height parameter and the leaf area parameter specifically includes:

[0016] Based on the stem diameter parameter, the seedling height parameter and the leaf area parameter, the seedling quantitative parameter is determined according to the first relationship model;

[0017] Based on the threshold intervals of each level, the threshold interval to which the seedling quantitative parameter belongs is determined;

[0018] Determining the grade of the seedlings based on the threshold interval to which the seedling quantification parameter belongs;

[0019] Wherein, the first relationship model is:

[0020] Q=N1*g1+N2*g2+N3*g3

[0021] Among them, Q represents the seedling quantification parameter, N1 represents the stem thickness parameter, g1 is the weight coefficient corresponding to the stem thickness parameter, N2 represents the seedling height parameter, g2 is the weight coefficient corresponding to the seedling height parameter, N3 represents the leaf area parameter, g3 is the weight coefficient corresponding to the leaf area parameter.

[0022] Furthermore, the thinning and transplanting of seedlings according to preset transplanting rules based on the grade of the seedlings includes:

[0023] A grouping and numbering step, grouping and numbering the seedlings in a preset order;

[0024] a seedling placement step, placing the seedlings that have been grouped and numbered in corresponding seedling trays according to preset placement rules based on the grade to which the seedlings belong;

[0025] The seedling marking step is to mark the row of the seedling tray as i, the column of the seedling tray as j, and place the seedling tray according to the spacing rule; wherein the spacing rule is: when i=2k≤r, then j=2m≤c; when i=2k+1≤r, then j=2m+1≤c; wherein r is the number of rows of the seedling tray, c is the number of columns of the seedling tray, k is a positive integer in 0, 1, 2, 3...r, and m is a positive integer in 0, 1, 2, 3...c;

[0026] A measuring step, determining the spacing Δd between the planting holes in the seedling tray by measuring, and establishing a seedling tray coordinate system based on the spacing between the planting holes in the seedling tray to determine the coordinates of the planting holes in the seedling tray;

[0027] Transplanting step: transplanting the seedlings based on the coordinates of the planting holes in the seedling tray.

[0028] Further, the characteristic image of the seedling is subjected to image processing to determine a stem thickness parameter, a seedling height parameter and a leaf area parameter, specifically including:

[0029] The characteristic image of the seedling is binarized to determine the stem diameter parameter, the seedling height parameter and the leaf area parameter.

[0030] Furthermore, the seedlings are classified into level zero, level one and level two, and when placing the seedlings, the level zero seedlings are placed at designated locations.

[0031] Furthermore, the method further comprises:

[0032] The coordinates of the planting holes in the seedling tray determined in the measuring step are saved in the first coordinate queue in the form of an array;

[0033] When performing the transplanting step, the planting hole coordinates stored in the first coordinate queue are taken out to transplant the seedlings, and the planting hole coordinates taken out from the first coordinate queue are saved in the second coordinate queue;

[0034] When all the planting hole coordinates stored in the first coordinate queue are taken out and saved in the second coordinate queue, the seedling tray is updated and the seedlings are transplanted based on the planting hole coordinates stored in the second coordinate queue, and the planting hole coordinates taken out from the second coordinate queue are saved in the first coordinate queue.

[0035] In a second aspect, an embodiment of the present invention provides a seedling identification and graded seedling thinning and transplanting device, comprising:

[0036] An acquisition module, used for acquiring a combined image of the seedling in the horizontal direction and the vertical direction based on the recognition signal;

[0037] A recognition module, used for inputting the combined image into a recognition model to obtain a characteristic image of the seedling; the characteristic image of the seedling is a matrix characteristic image, a stem characteristic image and a leaf characteristic image;

[0038] A first determination module is used to perform image processing on the characteristic image of the seedling to determine a stem diameter parameter, a seedling height parameter and a leaf area parameter;

[0039] A second determination module is used to determine the grade of the seedlings by a threshold method based on the stem diameter parameter, the seedling height parameter and the leaf area parameter;

[0040] A transplanting module, used for performing thinning and transplanting of seedlings according to preset transplanting rules based on the grade of the seedlings;

[0041] Among them, the recognition model uses sample combined images of seedlings in the horizontal and vertical directions as input data, and uses feature images of seedlings corresponding to the sample combined images of seedlings in the horizontal and vertical directions as output data, and is trained based on a machine learning algorithm.

[0042] In a third aspect, an embodiment of the present invention further provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the seedling identification and graded thinning and transplanting method as described in the first aspect above are implemented.

[0043] In a fourth aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the seedling identification and graded thinning and transplanting method as described in the first aspect above.

[0044] It can be seen from the above technical scheme that the seedling identification and graded thinning and transplanting method and device provided by the embodiment of the present invention obtains a combined image of the seedling in the horizontal and vertical directions based on the identification signal; the combined image is input into the recognition model to obtain a characteristic image of the seedling; the characteristic image of the seedling is a matrix characteristic image, a stem characteristic image and a leaf characteristic image; the characteristic image of the seedling is subjected to image processing to determine the stem thickness parameter, the seedling height parameter and the leaf area parameter; based on the stem thickness parameter, the seedling height parameter and the leaf area parameter, the grade of the seedling is determined by the threshold method; based on the grade of the seedling, the seedling is thinned and transplanted according to the preset transplanting rules, so that the seedling identification and grading tasks can be completed quickly and accurately, with good robustness, and the seedling identification and graded transplanting tasks can be accurately completed in a non-structured light environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 A schematic diagram of a process of seedling identification and graded seedling thinning and transplanting provided by an embodiment of the present invention;

[0047] Figure 2 A schematic diagram of a coordinate queue loop provided by an embodiment of the present invention;

[0048] Figure 3 A schematic diagram of a process flow of a seedling identification and graded seedling thinning and transplanting method provided by another embodiment of the present invention;

[0049] Figure 4 A schematic diagram of the structure of a seedling identification and graded seedling thinning and transplanting device provided in one embodiment of the present invention;

[0050] Figure 5A schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention clearer, the technical scheme in the embodiments of the present invention will be clearly described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments in 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. The seedling identification and graded seedling thinning and transplanting method provided by the present invention will be explained and illustrated in detail below through specific embodiments.

[0052] Figure 1 A schematic diagram of a process of seedling identification and graded seedling thinning and transplanting provided by an embodiment of the present invention; Figure 1 As shown, the method includes:

[0053] Step 101: Based on the recognition signal, a combined image of the seedling in the horizontal direction and the vertical direction is obtained.

[0054] In this step, it can be understood that there are one or more identification positions. The seedlings are transported to the identification area (i.e., the identification position) by a robot or other transport equipment. The equipment sends out an identification signal, and two horizontally and vertically arranged cameras collect the seedling images to obtain a combined image of the seedlings in the horizontal and vertical directions.

[0055] Step 102: input the combined image into a recognition model to obtain a characteristic image of the seedling; the characteristic image of the seedling is a matrix characteristic image, a stem characteristic image and a leaf characteristic image.

[0056] In this step, it should be noted that the recognition model is trained based on a machine learning algorithm, using sample combination images of seedlings in the horizontal and vertical directions as input data, and characteristic images of seedlings corresponding to the sample combination images of seedlings in the horizontal and vertical directions as output data.

[0057] In this step, it can be understood that the combined image is input into the recognition model to obtain the characteristic image of the matrix seedling (i.e., the seedling); the characteristic image of the seedling is the matrix characteristic image, the stem characteristic image and the leaf characteristic image.

[0058] Step 103: performing image processing on the characteristic image of the seedling to determine a stem diameter parameter, a seedling height parameter and a leaf area parameter.

[0059] In this step, it can be understood that the characteristic image of the seedling is subjected to image processing, such as binarization processing, so as to obtain corresponding stem thickness parameters, seedling height parameters and leaf area parameters.

[0060] Step 104: Based on the stem diameter parameter, the seedling height parameter and the leaf area parameter, the grade of the seedling is determined by a threshold method.

[0061] In this step, it can be understood that by setting a threshold value, the parameters of stem thickness, seedling height and leaf area are compared to determine within which threshold range the parameters of stem thickness, seedling height and leaf area fall, thereby determining the grade of the seedling. For example, the three parameters of stem thickness, seedling height and leaf area of ​​the seedling are saved, and the identification of the seedling is completed. The scoring thresholds of the three grades are set, and the three parameters of the seedling are multiplied by their corresponding weight parameters and the products are added to obtain the score of the seedling. The score of the seedling is determined according to the score threshold of the grade within which the seedling score falls, thereby realizing the grading of the seedling.

[0062] Step 105: Thinning and transplanting the seedlings according to preset transplanting rules based on the grade of the seedlings.

[0063] In this step, it can be understood that thinning and transplanting are performed based on the grade of the seedlings according to preset transplanting rules. For example, the seedlings are transplanted into the corresponding 4 seedling trays at intervals according to the transplanting rules to achieve the thinning and transplanting task.

[0064] It can be seen from the above technical scheme that the seedling identification and graded thinning and transplanting method provided by the embodiment of the present invention obtains a combined image of the seedlings in the horizontal and vertical directions based on the identification signal; the combined image is input into the recognition model to obtain a characteristic image of the seedling; the characteristic image of the seedling is a matrix characteristic image, a stem characteristic image and a leaf characteristic image; the characteristic image of the seedling is subjected to image processing to determine the stem thickness parameter, the seedling height parameter and the leaf area parameter; based on the stem thickness parameter, the seedling height parameter and the leaf area parameter, the grade of the seedling is determined by the threshold method; based on the grade of the seedling, the seedling is thinned and transplanted according to the preset transplanting rules, so that the seedling identification and grading tasks can be completed quickly and accurately, with good robustness, and the seedling identification and graded transplanting tasks can be accurately completed in a non-structured light environment.

[0065] Based on the above embodiment, in this embodiment, the method further includes:

[0066] Based on the combined image of the seedlings in the horizontal direction and the vertical direction, an RGB matrix corresponding to the combined image is obtained; the RGB matrix corresponding to the combined image is input into a recognition model to obtain an RGB value matrix; and the RGB value matrix is ​​converted into a feature image containing a seedling background.

[0067] In this embodiment, it can be understood that the RGB matrix of the combination of the two pictures collected horizontally and vertically is input into the model algorithm, and the model algorithm will output an RGB value matrix of the same size as the original image, which is reconverted into a feature image containing a black background of the seedlings.

[0068] On the basis of the above embodiment, in this embodiment, the level of the seedlings is determined by a threshold method based on the stem thickness parameter, the seedling height parameter and the leaf area parameter, specifically including:

[0069] Based on the stem diameter parameter, the seedling height parameter and the leaf area parameter, the seedling quantitative parameter is determined according to the first relationship model;

[0070] Based on the threshold intervals of each level, the threshold interval to which the seedling quantitative parameter belongs is determined;

[0071] Determining the grade of the seedlings based on the threshold interval to which the seedling quantification parameter belongs;

[0072] Wherein, the first relationship model is:

[0073] Q=N1*g1+N2*g2+N3*g3

[0074] Among them, Q represents the seedling quantification parameter, N1 represents the stem thickness parameter, g1 is the weight coefficient corresponding to the stem thickness parameter, N2 represents the seedling height parameter, g2 is the weight coefficient corresponding to the seedling height parameter, N3 represents the leaf area parameter, g3 is the weight coefficient corresponding to the leaf area parameter.

[0075] In this embodiment, it can be understood that the score thresholds of three levels, 0, 1, and 2, are set, and the seedlings of level 0 are defined as ungerminated seedlings, the seedlings of level 1 are defined as weakly germinated seedlings, and the seedlings of level 2 are defined as strong germinated seedlings. The three parameters of the seedlings (i.e., stem thickness parameter, seedling height parameter, and leaf area parameter) are multiplied by their corresponding weight parameters according to the first relationship model, and the products obtained are added to obtain the score of the seedling. The seedling score is determined to which level the seedling belongs if the seedling score is within the score threshold, thereby achieving the grading of the seedlings.

[0076] It can be seen from the above technical solutions that the seedling identification and grading seedling thinning and transplanting method provided by the embodiment of the present invention can quickly and accurately complete the seedling identification and grading tasks.

[0077] On the basis of the above embodiment, in this embodiment, the thinning and transplanting of seedlings according to preset transplanting rules based on the grade of the seedlings includes:

[0078] A grouping and numbering step, grouping and numbering the seedlings in a preset order;

[0079] a seedling placement step, placing the seedlings that have been grouped and numbered in corresponding seedling trays according to preset placement rules based on the grade to which the seedlings belong;

[0080] The seedling marking step is to mark the row of the seedling tray as i, the column of the seedling tray as j, and place the seedling tray according to the spacing rule; wherein the spacing rule is: when i=2k≤r, then j=2m≤c; when i=2k+1≤r, then j=2m+1≤c; wherein r is the number of rows of the seedling tray, c is the number of columns of the seedling tray, k is a positive integer in 0, 1, 2, 3...r, and m is a positive integer in 0, 1, 2, 3...c;

[0081] A measuring step, determining the spacing Δd between the planting holes in the seedling tray by measuring, and establishing a seedling tray coordinate system based on the spacing between the planting holes in the seedling tray to determine the coordinates of the planting holes in the seedling tray;

[0082] Transplanting step: transplanting the seedlings based on the coordinates of the planting holes in the seedling tray.

[0083] In this embodiment, for example, after obtaining the grade of the seedlings, the rules and numbering are formulated as follows: divide them into two groups A and B in order of three, and number the seedling grabbers in the two groups as A1, A2, A3, B1, B2, and B3. If there are zero-grade seedlings, throw the zero-grade seedlings in each group to the designated location. The last ones left in groups A and B are the first- and second-level seedlings. The seedling trays to be placed are divided into four seedling trays: Ⅰ, Ⅱ, Ⅲ, and Ⅳ. Ⅰ and Ⅲ are used to place the first-level seedlings, Ⅱ and Ⅳ are used to place the second-level seedlings. The first- and second-level seedlings in A1, A2, and A3 in group A can only be placed in seedling trays Ⅰ and Ⅱ, and the first- and second-level seedlings in B1, B2, and B3 in group B can only be placed in seedling trays Ⅲ and Ⅳ. The row of seedling tray Ⅰ is recorded as i and the column is counted as j. The placement rule of the seedling trays to be placed is to place them at intervals, that is: when i=2k≤r, j=2m≤c, when i=2k+1≤r, j=2m+1≤c, where (r, c are the number of rows and columns of the seedling trays to be placed, k=0,1,2,3…r; m=0,1,2,3…c); the hole spacing Δd of the seedling tray to be placed can be obtained by measurement, with the lower left corner of the seedling tray as the origin. O, the row direction is the x-axis, and the column direction is the y-axis to establish a seedling tray coordinate system, then the coordinates of the seedlings to be placed in the seedling tray can be expressed as (j×Δd, i×Δd), and the calculated coordinates are saved in the coordinate queue D11 in the form of an array; when transplanting, the coordinates are taken out from D11 and the seedlings are placed at the coordinate position, and the coordinates are stored in the coordinate queue D12. When there is a full seedling tray in seedling tray I, there is no data in the queue D11 and D12 is full. The seedling tray I is transported out separately by the conveying device, and a new empty seedling tray is added to its position. When transplanting, the coordinates are taken out from D12 and the seedlings of the corresponding level are placed, and the coordinates are stored in the coordinate queue D11. Repeat the above transplanting operation. Similarly, according to the established rules and numbers, the above operations are implemented in seedling trays II, III, and IV to complete the transplanting and thinning tasks of seedlings.

[0084] It can be seen from the above technical scheme that the seedling identification and graded thinning and transplanting method provided by the embodiment of the present invention can simplify the equipment design and improve the mechanization degree of seedling cultivation through grouping and numbering steps, seedling placement steps, seedling marking steps, measurement steps and transplanting steps, while having better robustness.

[0085] On the basis of the above embodiment, in this embodiment, the characteristic image of the seedling is subjected to image processing to determine the stem diameter parameter, the seedling height parameter and the leaf area parameter, specifically including:

[0086] The characteristic image of the seedling is binarized to determine the stem diameter parameter, the seedling height parameter and the leaf area parameter.

[0087] On the basis of the above embodiment, in this embodiment, the seedlings belong to the grades of zero, one and two, and when placing the seedlings, the zero-grade seedlings are placed at the designated position.

[0088] In this embodiment, it should be noted that the levels of the seedlings are zero, one and two; zero-level seedlings are defined as ungerminated seedlings, one-level seedlings are weakly germinated seedlings, and two-level seedlings are strong germinated seedlings. By placing zero-level seedlings in designated locations, targeted mechanical work is facilitated.

[0089] Based on the above embodiment, in this embodiment, the method further includes:

[0090] The coordinates of the planting holes in the seedling tray determined in the measuring step are saved in the first coordinate queue in the form of an array;

[0091] When performing the transplanting step, the planting hole coordinates stored in the first coordinate queue are taken out to transplant the seedlings, and the planting hole coordinates taken out from the first coordinate queue are saved in the second coordinate queue;

[0092] When all the planting hole coordinates stored in the first coordinate queue are taken out and saved in the second coordinate queue, the seedling tray is updated and the seedlings are transplanted based on the planting hole coordinates stored in the second coordinate queue, and the planting hole coordinates taken out from the second coordinate queue are saved in the first coordinate queue.

[0093] In this embodiment, see Figure 2 and Figure 3For example, the determined coordinates are saved in the coordinate queue D11 in the form of an array; when transplanting, the coordinates are taken out from D11 and the seedlings are placed at the coordinate position, and the coordinates are stored in the coordinate queue D12. When there is a full seedling tray in the Ⅰ seedling tray, there is no data in the D11 queue and D12 is full. The conveying device conveys the Ⅰ seedling tray separately, and a new empty seedling tray is added to its position. When transplanting, the coordinates are taken out from D12 and the seedlings of the corresponding level are placed, and the coordinates are stored in the coordinate queue D11. Repeat the above transplanting operation. Similarly, according to the established rules and numbers, the above operations are implemented in the Ⅱ, Ⅲ, and Ⅳ seedling trays to complete the transplanting and thinning tasks of the seedlings.

[0094] In order to better understand the present embodiment, the contents of the embodiments of the present invention are further described below, but the present invention is not limited to the following embodiments.

[0095] First, the six seedling transport devices move the seedlings to the designated identification area, and then the device sends out an identification signal. The cameras arranged in the horizontal and vertical directions in the area collect images of the seedlings. Secondly, the two images collected by the camera are combined into an RGB matrix of an image and input into the model algorithm. The model algorithm finds the stem thickness, seedling height, and leaf area through calculation. Finally, the model algorithm outputs an RGB value matrix of the same size as the original image, which is reconverted into a feature image containing a black background for the seedlings. Next, the feature image is binarized to obtain the three parameters of stem thickness, seedling height, and leaf area. Finally, the three parameters are saved, and the seedling identification task is completed.

[0096] After obtaining the three parameters of stem thickness, seedling height and leaf area of ​​the seedling through the above operations, it is necessary to manually set the stem thickness weight parameter, seedling height weight parameter, leaf area weight parameter and several levels of score thresholds. Add the product of the weight and the corresponding parameter to get the score of the seedling. According to the score of the seedling, check which level of score threshold it belongs to. If it is within a certain level threshold, the seedling belongs to that level, and the seedling grading task is completed.

[0097] After getting the grades of the six seedlings, the rules and numbering are formulated as follows: divide them into two groups, A and B, in order of three in a group, and number the seedling grabbers in the two groups as A1, A2, A3, B1, B2, B3. If there are grade 0 seedlings, throw the grade 0 seedlings in each group to the designated area. The remaining seedlings in groups A and B are grade 1 and 2 seedlings, and the seedling trays to be placed are divided into four seedling trays, Ⅰ, Ⅲ, Ⅳ. Ⅰ and Ⅲ are used to place grade 1 seedlings, Ⅱ and Ⅳ are used to place grade 2 seedlings, and the grade 1 and 2 seedlings with A1, A2, A3 in group A can only be placed in the seedling trays Ⅰ and Ⅱ, and the grade 1 and 2 seedlings with B1, B2, B3 in group B can only be placed in the seedling trays Ⅲ and Ⅳ. The following is an example of the principle of transplanting and thinning seedlings, with the row of the seedling tray I being i and the column being j. The placement rule of the seedling tray to be placed is to place the seedling tray at intervals, that is, when i=2k≤r, j=2m≤c; when i=2k+1≤r, j=2m+1≤c, where (r, c are the number of rows and columns of the seedling tray to be placed, k=0,1,2,3…r; m=0,1,2,3…c). The hole spacing Δd of the seedling tray to be placed can be obtained by measurement. The seedling tray coordinate system is established with the lower left corner of the seedling tray as the origin O, the row direction as the x-axis, and the column direction as the y-axis. The coordinates of the seedling to be placed in the seedling tray are (j×Δd, i×Δd), which can be expressed as (xj, yi). The calculated coordinates are saved in the coordinate queue D11 in the form of an array. When transplanting, the coordinates are taken out from D11 and the seedlings are placed at the coordinate position. At the same time, the coordinates are stored in the coordinate queue D12. When there is a full seedling tray in the I seedling tray, there is no data in the D11 queue and the D12 queue is full. The conveying device transports the I seedling tray out separately, and a new empty seedling tray is added to its position. When transplanting, the coordinates are taken out from D12 and the seedlings of the corresponding level are placed. At the same time, the coordinates are stored in the coordinate queue D11, and the above transplanting operation is repeated. Similarly, the above operations are performed in the II, III, and IV seedling trays according to the established rules and numbers to complete the seedling transplanting and thinning tasks.

[0098] Figure 4 A schematic diagram of the structure of a seedling identification and grading seedling thinning and transplanting device provided in one embodiment of the present invention is shown in FIG. Figure 4 As shown, the device includes: an acquisition module 201, an identification module 202, a first determination module 203, a second determination module 204 and a transplantation module 205, wherein:

[0099] The acquisition module 201 is used to acquire a combined image of the seedling in the horizontal direction and the vertical direction based on the recognition signal;

[0100] The recognition module 202 is used to input the combined image into a recognition model to obtain a characteristic image of the seedling; the characteristic image of the seedling is a matrix characteristic image, a stem characteristic image and a leaf characteristic image;

[0101] The first determination module 203 is used to perform image processing on the characteristic image of the seedling to determine the stem diameter parameter, the seedling height parameter and the leaf area parameter;

[0102] A second determination module 204 is used to determine the grade of the seedlings by a threshold method based on the stem diameter parameter, the seedling height parameter and the leaf area parameter;

[0103] Transplanting module 205, used for performing thinning and transplanting of seedlings according to preset transplanting rules based on the grade of the seedlings;

[0104] Among them, the recognition model uses sample combined images of seedlings in the horizontal and vertical directions as input data, and uses feature images of seedlings corresponding to the sample combined images of seedlings in the horizontal and vertical directions as output data, and is trained based on a machine learning algorithm.

[0105] The seedling identification and graded thinning and transplanting device provided in the embodiment of the present invention can be specifically used to execute the seedling identification and graded thinning and transplanting method of the above embodiment. Its technical principles and beneficial effects are similar. Please refer to the above embodiment for details and will not be repeated here.

[0106] Based on the same inventive concept, an embodiment of the present invention provides an electronic device, see Figure 5 , the electronic device specifically includes the following contents: a processor 301, a communication interface 303, a memory 302 and a communication bus 304;

[0107] Among them, the processor 301, the communication interface 303, and the memory 302 communicate with each other through the communication bus 304; the communication interface 303 is used to realize the information transmission between various modeling software and intelligent manufacturing equipment module libraries and other related equipment; the processor 301 is used to call the computer program in the memory 302, and the processor implements the methods provided by the above-mentioned method embodiments when executing the computer program. For example, when the processor executes the computer program, the following steps are implemented: based on the recognition signal, a combined image of the seedling in the horizontal direction and the vertical direction is obtained; the combined image is input into the recognition model to obtain the characteristic image of the seedling; the seedling The characteristic images are matrix characteristic images, stem characteristic images and leaf characteristic images; the characteristic images of the seedlings are processed to determine stem thickness parameters, seedling height parameters and leaf area parameters; based on the stem thickness parameters, seedling height parameters and leaf area parameters, the grade of the seedlings is determined by a threshold method; based on the grade of the seedlings, thinning and transplanting are performed according to preset transplanting rules; wherein the recognition model uses sample combined images of seedlings in horizontal and vertical directions as input data, and uses characteristic images of seedlings corresponding to the sample combined images of seedlings in horizontal and vertical directions as output data, and is obtained through training based on a machine learning algorithm.

[0108] Based on the same inventive concept, another embodiment of the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it is implemented to execute the methods provided by the above-mentioned method embodiments, for example, based on the recognition signal, a combined image of the seedlings in the horizontal and vertical directions is obtained; the combined image is input into the recognition model to obtain the characteristic image of the seedlings; the characteristic image of the seedlings is a matrix characteristic image, a stem characteristic image and a leaf characteristic image; the characteristic image of the seedlings is image processed to determine the stem thickness parameter, the seedling height parameter and the leaf area parameter; based on the stem thickness parameter, the seedling height parameter and the leaf area parameter, the grade of the seedlings is determined by the threshold method; based on the grade of the seedlings, thinning and transplanting are performed according to preset transplanting rules; wherein the recognition model is obtained by training based on a machine learning algorithm by using sample combined images of seedlings in the horizontal and vertical directions as input data, and using characteristic images of seedlings corresponding to the sample combined images of seedlings in the horizontal and vertical directions as output data.

[0109] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, i.e., they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative effort.

[0110] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.

[0111] In addition, in the present invention, terms such as "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0112] In addition, in the present invention, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0113] In addition, in the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A seedling identification and graded seedling thinning and transplanting method, characterized in that: include: Based on the recognition signal, a combined image of the seedling in the horizontal direction and the vertical direction is obtained; Inputting the combined image into a recognition model to obtain a characteristic image of the seedling; the characteristic image of the seedling is a matrix characteristic image, a stem characteristic image and a leaf characteristic image; Processing the characteristic image of the seedling to determine the stem thickness parameter, seedling height parameter and leaf area parameter; Based on the stem diameter parameter, the seedling height parameter and the leaf area parameter, determining the grade of the seedling by a threshold method; Thinning and transplanting the seedlings according to preset transplanting rules based on the grade of the seedlings; The recognition model is obtained by training based on a machine learning algorithm using sample combination images of seedlings in the horizontal and vertical directions as input data and feature images of seedlings corresponding to the sample combination images of seedlings in the horizontal and vertical directions as output data; The step of determining the grade of the seedlings by a threshold method based on the stem thickness parameter, the seedling height parameter and the leaf area parameter specifically includes: Based on the stem diameter parameter, the seedling height parameter and the leaf area parameter, the seedling quantitative parameter is determined according to the first relationship model; Based on the threshold intervals of each level, the threshold interval to which the seedling quantitative parameter belongs is determined; Determining the grade of the seedlings based on the threshold interval to which the seedling quantification parameter belongs; Wherein, the first relationship model is: Q=N1*g1+N2*g2+N3*g3 Among them, Q represents the seedling quantification parameter, N1 represents the stem thickness parameter, g1 is the weight coefficient corresponding to the stem thickness parameter, N2 represents the seedling height parameter, g2 is the weight coefficient corresponding to the seedling height parameter, N3 represents the leaf area parameter, g3 is the weight coefficient corresponding to the leaf area parameter; The thinning and transplanting of seedlings according to preset transplanting rules based on the grade of the seedlings includes: A grouping and numbering step, grouping and numbering the seedlings in a preset order; a seedling placement step, placing the seedlings that have been grouped and numbered in corresponding seedling trays according to preset placement rules based on the grade to which the seedlings belong; The seedling marking step is to mark the row of the seedling tray as i, the column of the seedling tray as j, and place the seedling tray according to the spacing rule; wherein the spacing rule is: when i=2k≤r, then j=2m≤c; when i=2k+1≤r, then j=2m+1≤c; wherein r is the number of rows of the seedling tray, c is the number of columns of the seedling tray, k is a positive integer in 0, 1, 2, 3...r, and m is a positive integer in 0, 1, 2, 3...c; A measuring step, determining the spacing Δd between the planting holes in the seedling tray by measuring, and establishing a seedling tray coordinate system based on the spacing between the planting holes in the seedling tray to determine the coordinates of the planting holes in the seedling tray; Transplanting step: transplanting the seedlings based on the coordinates of the planting holes in the seedling tray.

2. The seedling identification and graded seedling thinning and transplanting method according to claim 1, characterized in that: The method further comprises: Based on the combined image of the seedlings in the horizontal direction and the vertical direction, an RGB matrix corresponding to the combined image is obtained; Inputting the RGB matrix corresponding to the combined image into the recognition model to obtain an RGB value matrix; The RGB value matrix is ​​converted into a feature image containing the seedling background.

3. The seedling identification and graded seedling thinning and transplanting method according to claim 1, characterized in that: The characteristic image of the seedling is subjected to image processing to determine a stem diameter parameter, a seedling height parameter and a leaf area parameter, specifically comprising: The characteristic image of the seedling is binarized to determine the stem diameter parameter, the seedling height parameter and the leaf area parameter.

4. The seedling identification and graded seedling thinning and transplanting method according to claim 1, characterized in that: The seedlings are classified into zero-level, one-level and two-level grades. When placing the seedlings, the zero-level seedlings are placed at designated locations.

5. The seedling identification and graded seedling thinning and transplanting method according to claim 1, characterized in that: The method further comprises: The coordinates of the planting holes in the seedling tray determined in the measuring step are saved in the first coordinate queue in the form of an array; When performing the transplanting step, the planting hole coordinates stored in the first coordinate queue are taken out to transplant the seedlings, and the planting hole coordinates taken out from the first coordinate queue are saved in the second coordinate queue; When all the planting hole coordinates stored in the first coordinate queue are taken out and saved in the second coordinate queue, the seedling tray is updated and the seedlings are transplanted based on the planting hole coordinates stored in the second coordinate queue, and the planting hole coordinates taken out from the second coordinate queue are saved in the first coordinate queue.

6. A seedling identification and grading seedling thinning and transplanting device, characterized in that: include: An acquisition module, used for acquiring a combined image of the seedling in the horizontal direction and the vertical direction based on the recognition signal; A recognition module, used for inputting the combined image into a recognition model to obtain a characteristic image of the seedling; the characteristic image of the seedling is a matrix characteristic image, a stem characteristic image and a leaf characteristic image; A first determination module is used to perform image processing on the characteristic image of the seedling to determine a stem diameter parameter, a seedling height parameter and a leaf area parameter; A second determination module is used to determine the grade of the seedlings by a threshold method based on the stem diameter parameter, the seedling height parameter and the leaf area parameter; A transplanting module, used for performing thinning and transplanting of seedlings according to preset transplanting rules based on the grade of the seedlings; The recognition model is obtained by training based on a machine learning algorithm using sample combination images of seedlings in the horizontal and vertical directions as input data and feature images of seedlings corresponding to the sample combination images of seedlings in the horizontal and vertical directions as output data; The step of determining the grade of the seedlings by a threshold method based on the stem thickness parameter, the seedling height parameter and the leaf area parameter specifically includes: Based on the stem diameter parameter, the seedling height parameter and the leaf area parameter, the seedling quantitative parameter is determined according to the first relationship model; Based on the threshold intervals of each level, the threshold interval to which the seedling quantitative parameter belongs is determined; Determining the grade of the seedlings based on the threshold interval to which the seedling quantification parameter belongs; Wherein, the first relationship model is: Q=N1*g1+N2*g2+N3*g3 Among them, Q represents the seedling quantification parameter, N1 represents the stem thickness parameter, g1 is the weight coefficient corresponding to the stem thickness parameter, N2 represents the seedling height parameter, g2 is the weight coefficient corresponding to the seedling height parameter, N3 represents the leaf area parameter, g3 is the weight coefficient corresponding to the leaf area parameter; The thinning and transplanting of seedlings according to preset transplanting rules based on the grade of the seedlings includes: A grouping and numbering step, grouping and numbering the seedlings in a preset order; a seedling placement step, placing the seedlings that have been grouped and numbered in corresponding seedling trays according to preset placement rules based on the grade to which the seedlings belong; The seedling marking step is to mark the row of the seedling tray as i, the column of the seedling tray as j, and place the seedling tray according to the spacing rule; wherein the spacing rule is: when i=2k≤r, then j=2m≤c; when i=2k+1≤r, then j=2m+1≤c; wherein r is the number of rows of the seedling tray, c is the number of columns of the seedling tray, k is a positive integer in 0, 1, 2, 3...r, and m is a positive integer in 0, 1, 2, 3...c; A measuring step, determining the spacing Δd between the planting holes in the seedling tray by measuring, and establishing a seedling tray coordinate system based on the spacing between the planting holes in the seedling tray to determine the coordinates of the planting holes in the seedling tray; Transplanting step: transplanting the seedlings based on the coordinates of the planting holes in the seedling tray.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the seedling identification and graded seedling thinning and transplanting method as described in any one of claims 1 to 5 is implemented.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the seedling identification and graded seedling thinning and transplanting method as described in any one of claims 1 to 5 is implemented.

Images