grafting device and grafting method for vegetable grafting
By using image acquisition and analysis technology in the grafting device to dynamically adjust the docking trajectory, the problem of incision misalignment caused by individual differences in seedlings was solved, achieving high-precision docking of rootstock and scion incisions, and improving grafting survival rate and quality.
Patent Information
- Application Number
- CN202611103692.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-23
- Publication Date
- 2026-11-17
AI Technical Summary
Existing grafting machines rely on fixed mechanical trajectories for grafting when dealing with individual differences in seedling growth. This results in misalignment and loose adhesion between the cut centers of the rootstock and scion, reducing grafting survival rate and quality.
A docking device comprising first and second clamping and transfer modules, an image-sensitive acquisition module, and a control module is adopted. By acquiring and analyzing the incision feature information of the rootstock and scion, the docking trajectory is dynamically adjusted to achieve high-precision overlap and tight fit of the incision center.
It improved the grafting survival rate and quality, enhanced grafting precision, increased the incision fit rate by more than 10%, and increased the survival rate by 5%.
Smart Images

Figure CN123100407A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of agricultural intelligent equipment technology, and in particular to a grafting device and grafting method for vegetable grafting. Background Technology
[0002] Grafting technology is a key agronomic method for improving the resistance of vegetables to stress, preventing soil-borne diseases, and increasing yield and quality.
[0003] Currently, most grafting machines use a mechanical reference positioning method for grafting, relying on preset mechanical coordinates of the clamping and transporting mechanism to complete the grafting operation between the rootstock and scion. However, in actual grafting production, individual growth differences exist among seedlings, and minute cutting errors occur during cutting, leading to random deviations in the position, shape, and area of the cuts on the rootstock and scion. Relying solely on a fixed mechanical trajectory for grafting easily results in problems such as misalignment of the cut centers, loose fit, and localized suspension, significantly reducing grafting survival rate and quality. Summary of the Invention
[0004] This invention provides a grafting device and method for vegetable grafting, which solves the problem that existing grafting devices rely on a fixed mechanical trajectory when grafting seedlings with different growth characteristics, resulting in misalignment of the cut centers of the rootstock and scion, poor adhesion, and thus unstable grafting quality.
[0005] To solve the above-mentioned technical problems, the present invention is implemented as follows: In a first aspect, the present invention provides a grafting device for vegetable grafting, comprising: The workbench has a rootstock photography area, a scion photography area, and a grafting area; A first clamping and transferring module is provided in the rootstock photographing area of the workbench. The first clamping and transferring module is used to carry and fix the cut rootstock and move the rootstock from the rootstock photographing area to the docking area. A first image-sensitive acquisition module is disposed opposite to the first clamping and transferring module. The first image-sensitive acquisition module is used to acquire image information of the rootstock incision. The second clamping and transfer module is located in the scion photography area of the workbench. The second clamping and transfer module is used to carry and fix the cut scion, and move the scion from the scion photography area to the docking area. The second image-sensitive acquisition module is disposed opposite to the second clamping and transferring module. The second image-sensitive acquisition module is used to acquire scion cut image information. The control module is electrically connected to the first clamping and transferring module, the first image-sensitive acquisition module, the second clamping and transferring module, and the second image-sensitive acquisition module, respectively. The control module is used to extract feature information of the rootstock incision and the scion incision based on the rootstock incision image information and the scion incision image information, and control the first clamping and transferring module and the second clamping and transferring module to perform the docking operation of the rootstock incision and the scion incision in the docking area according to the feature information.
[0006] According to the present invention, a grafting device for vegetable grafting includes a first clamping and transferring module comprising: a first linear module, a rootstock gripper, and a first rotary drive component; The first rotary drive is disposed on the worktable, the output end of the first rotary drive is connected to the first linear module, the slide of the first linear module is connected to the rootstock gripper, and the rootstock gripper is used to hold the rootstock. The control module is used to control the rotation angle of the first rotary drive component based on the feature information of the incision in the rootstock.
[0007] According to the present invention, a grafting device for vegetable grafting is provided, wherein the first clamping and transferring module further includes: a seedling top rod and a seedling top driving component; The seedling-topping drive is located on the first linear module. The output end of the seedling-topping drive is connected to the seedling-topping rod. The seedling-topping rod is located on the upper side of the rootstock clamp and abuts against the base of the cotyledon of the rootstock.
[0008] According to the present invention, a docking device for vegetable grafting is provided, wherein the first clamping and transferring module further includes: a micrometer structure; The micrometer structure is located on the first linear module. The micrometer structure has a screw head, which is driven by the top seedling rod to finely adjust the position of the top seedling rod.
[0009] According to the present invention, a grafting device for vegetable grafting is provided, wherein the second clamping and transferring module includes: a second linear module, a scion clamp, and a second rotary drive component; The second rotary drive is disposed on the worktable, and the output end of the second rotary drive is connected to the second linear module. The slide of the second linear module is connected to the scion gripper, and the scion gripper is used to hold the scion. The control module is used to control the rotation angle of the second rotary drive component based on the feature information of the scion cut.
[0010] According to the present invention, a grafting device for vegetable grafting is provided, wherein at least one of the second image-sensitive acquisition module and the first image-sensitive acquisition module includes a bracket, a camera and a supplementary light; The bracket is detachably connected to the worktable. The camera and the fill light are both mounted on the bracket, and the fill light is arranged around the camera.
[0011] Secondly, the present invention provides a grafting method based on the grafting device for vegetable grafting as described in any of the preceding claims, comprising: The first image-sensitive acquisition module is controlled to acquire rootstock incision image information, and the second image-sensitive acquisition module is controlled to acquire scion incision image information; Based on the rootstock cut image information and scion cut image information, feature information is extracted, including: the cut outline boundary, centroid coordinates, and effective cut area of the rootstock and scion, and a cut feature dataset is established; Based on the incision feature dataset, obtain the docking position deviation information of the rootstock incision and the scion incision; Based on the docking position deviation information, the first clamping and transferring module and the second clamping and transferring module are controlled to dynamically correct the docking trajectory of the rootstock incision and the scion incision in the docking area, so as to realize the docking operation in which the center of the rootstock incision and the scion incision coincide.
[0012] According to a grafting method provided by the present invention, the step of extracting feature information based on the rootstock incision image information and the scion incision image information includes: Edge detection algorithms and morphological filtering are used to obtain the cut contour boundaries of rootstock and scion, and to remove interference from hair and stem impurities; The centroid coordinates are obtained by pixel integration of the contour region; The effective cut area is obtained by calculating the number of effective pixels within the statistical contour.
[0013] According to a grafting method provided by the present invention, the step of obtaining the grafting position deviation information of the rootstock incision and the scion incision based on the incision feature dataset includes: The positional deviation of the center of the rootstock incision relative to the center of the first clamping and transferring module is obtained; Obtain the positional deviation of the scion incision center relative to the center of the second clamping and transferring module; The positional deviation includes left-right lateral deviation and front-back longitudinal deviation, with the deviation direction and deviation value.
[0014] According to the grafting method provided by the present invention, the method further includes: conducting a quality assessment on the grafted seedlings after grafting, including the cutting success rate and the incision fit rate of the rootstock and scion; And / or, before extracting feature information based on the rootstock incision image information and the scion incision image information, the method further includes: The rootstock cut image information and scion cut image information are preprocessed. The original image is subjected to grayscale conversion, noise reduction and threshold segmentation to remove impurities and shadow interference in the image and obtain a binary image of the cut.
[0015] The grafting device and method for vegetable grafting provided by this invention, by setting up a first image-sensitive acquisition module and a second image-sensitive acquisition module, acquires images of the cut surfaces of the rootstock and scion after cutting, obtains the cut image information of the rootstock and scion, and controls the grafting operation of the first clamping and transferring module and the second clamping and transferring module according to the cut image information, synchronously corrects the grafting parameters, and compensates for positional deviations, so as to achieve high-precision overlap and tight fit of the center of the rootstock cut and the scion cut, which is beneficial to the stability of grafting quality. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0017] Figure 1 This is a three-dimensional structural schematic diagram of the grafting device for vegetable grafting provided by the present invention.
[0018] Figure 2 This is a three-dimensional structural diagram of the first clamping and transferring module provided by the present invention.
[0019] Figure 3 This is a three-dimensional structural schematic diagram of the second clamping and transferring module provided by the present invention.
[0020] Figure 4 This is a three-dimensional structural schematic diagram of the first image-sensitive acquisition module provided by the present invention.
[0021] Figure 5 This is a schematic flowchart of the docking method based on the docking device for vegetable grafting provided by the present invention.
[0022] Figure 6 This is a schematic diagram of the first clamping and transferring module and the second clamping and transferring module provided by the present invention connecting the rootstock and scion in the docking area.
[0023] Figure 7 This is a three-dimensional structural diagram of the second image-sensitive acquisition module provided by the present invention.
[0024] Figure label: 1. Workbench; 2. First clamping and transferring module; 21. First linear module; 22. Rootstock gripper; 23. First rotary drive component; 24. Seedling lifting rod; 25. Seedling lifting drive component; 26. Micrometer structure; 261. Twisting head; 3. First image sensor acquisition module; 31. Stand; 32. Camera; 33. Fill light; 4. Second clamping and transferring module; 41. Second linear module; 42. Scion gripper; 43. Second rotary drive component; 44. Seedling retainer; 5. Second image sensor acquisition module; 100. Rootstock; 1001. Cotyledon; 101. Scion. Detailed Implementation
[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this 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 this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.
[0026] In the description of the embodiments of the present invention, it should be noted that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the purpose of clarifying the embodiments of the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of the present invention. In addition, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.
[0027] In the description of the embodiments of the present invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "connected" and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in the embodiments of the present invention according to the specific circumstances.
[0028] In embodiments of the present invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.
[0029] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.
[0030] The following is combined Figures 1 to 7 The present invention will provide a detailed description of the grafting device and grafting method for vegetable grafting provided in the embodiments of the present invention through specific implementation examples and application scenarios.
[0031] Firstly, such as Figure 1 As shown, this embodiment provides a grafting device for vegetables, including: a workbench 1, a first clamping and transferring module 2, a first image-sensitive acquisition module 3, a second clamping and transferring module 4, a second image-sensitive acquisition module 5, and a control module.
[0032] Workbench 1 has a rootstock photography area, a scion photography area, and a grafting area.
[0033] The first clamping and transferring module 2 is located in the rootstock photographing area of the workbench 1. The first clamping and transferring module 2 is used to carry and fix the cut rootstock 100, and move the rootstock 100 from the rootstock photographing area to the docking area.
[0034] The first image-sensitive acquisition module 3 is positioned opposite to the first clamping and transferring module 2. The first image-sensitive acquisition module 3 is used to acquire image information of the rootstock incision.
[0035] The second clamping and transfer module 4 is located in the scion photography area of the workbench 1. The second clamping and transfer module 4 is used to carry and fix the cut scion 101, and move the scion 101 from the scion photography area to the docking area.
[0036] The second image-sensitive acquisition module 5 is positioned opposite to the second clamping and transferring module 4. The second image-sensitive acquisition module 5 is used to acquire image information of the scion cut.
[0037] The control module is electrically connected to the first clamping and transferring module 2, the first image-sensitive acquisition module 3, the second clamping and transferring module 4, and the second image-sensitive acquisition module 5, respectively. The control module is used to extract the feature information of the rootstock incision and the scion incision based on the rootstock incision image information and the scion incision image information, and control the first clamping and transferring module 2 and the second clamping and transferring module 4 to perform the docking operation of the rootstock incision and the scion incision in the docking area based on the feature information.
[0038] Understandably, in the process of vegetable grafting, after removing one cotyledon from the rootstock 100, leaving one cotyledon 1001, the scion 101 is cut into a stem and the two cuts are joined together to form a grafted seedling. The vegetables in this embodiment can be cucurbit vegetables, such as watermelon, cucumber, and cantaloupe, or solanaceous vegetables, such as tomato, pepper, and eggplant.
[0039] For the rootstock 100 that has already been cut, the first clamping and transferring module 2 clamps and fixes the rootstock 100 in the rootstock imaging area, and the first image-sensitive acquisition module 3 acquires image data of the rootstock cut. Subsequently, the first clamping and transferring module 2 moves the rootstock 100 from the rootstock imaging area to the docking area.
[0040] The grafting device for vegetables in this embodiment is only used for grafting rootstock 100 and scion 101 that have already been cut. For the cut scion 101, the second clamping and transferring module 4 clamps and fixes the scion 101 in the scion photography area, and the second image-sensitive acquisition module 5 acquires image data of the scion cut. Subsequently, the second clamping and transferring module 4 moves the scion 101 from the scion photography area to the grafting area.
[0041] Specifically, in order to prevent interference between image acquisition of the rootstock incision and image acquisition of the scion incision, the docking area in this embodiment is located between the rootstock photography area and the scion photography area.
[0042] In practical applications, the rootstock 100 and scion 101 are not perfectly straight, and there are slight cutting errors in the cutting of the rootstock 100 and scion 101, resulting in random deviations in their cuts. Existing technologies typically use a fixed rotation angle between the first clamping and transferring module 2 and the second clamping and transferring module 4 to align the rootstock and scion cuts. Because this alignment process is fixed, it cannot accommodate individual differences. This embodiment can acquire feature information of the rootstock and scion cuts by collecting actual images of them, controlling the real-time movement and angle of the first clamping and transferring module 2 and the second clamping and transferring module 4. This adaptively compensates for individual differences, mechanical positioning errors, and cut position offsets caused by cutting errors in the rootstock 100 and scion 101, preventing misalignment of the cuts, improving the tightness of the fit, significantly enhancing grafting accuracy, and effectively increasing the grafting survival rate.
[0043] The feature information includes: the cut outline boundary, centroid coordinates and effective cut area of the rootstock 100 and scion 101.
[0044] The grafting device for vegetable grafting provided by the present invention, by setting up a first image-sensitive acquisition module 3 and a second image-sensitive acquisition module 5, acquires images of the cut surfaces of the rootstock 100 and scion 101 after cutting, obtains the cut image information of the rootstock 100 and scion 101, and controls the grafting operation of the first clamping and transferring module 2 and the second clamping and transferring module 4 according to the cut image information, synchronously corrects the grafting parameters, and compensates for positional deviations, so as to achieve high-precision overlap and tight fit of the center of the rootstock cut and the scion cut, which is beneficial to the stability of grafting quality.
[0045] like Figure 2 As shown, the first clamping and transferring module 2 in this embodiment includes: a first linear module 21, an anvil gripper 22, and a first rotary drive 23.
[0046] The first rotary drive 23 is disposed on the worktable 1. The output end of the first rotary drive 23 is connected to the first linear module 21. The slide of the first linear module 21 is connected to the rootstock gripper 22. The rootstock gripper 22 is used to hold the rootstock 100.
[0047] The control module is used to control the rotation angle of the first rotating drive 23 based on the feature information of the rootstock cut.
[0048] Understandably, the rootstock gripper 22 is used to hold the cut rootstock 100. The gripper is equipped with a flexible pad to buffer the clamping force and prevent damage to the rootstock 100.
[0049] The first linear module 21 is used to drive the extension and retraction of the rootstock gripper 22, and the first rotary drive 23 is used to drive the rotation of the rootstock gripper 22. The first rotary drive 23 drives the rootstock gripper 22 to rotate from the rootstock photography area to the docking area, and the first linear module 21 drives the rootstock gripper 22 to extend in preparation for docking.
[0050] Based on the characteristic information of the rootstock incision, the control module controls the rotation angle of the first rotation drive 23. By fine-tuning the rotation angle, the centering of the rootstock incision and the scion incision is achieved.
[0051] Specifically, the first linear module 21 can be a ball screw slide, and the first rotary drive 23 can be a servo motor. A reducer is also configured between the first rotary drive 23 and the first linear module 21.
[0052] like Figure 2 As shown, the first clamping and transferring module 2 in this embodiment also includes: a seedling-topping rod 24 and a seedling-topping drive component 25.
[0053] The seedling driving component 25 is located on the first linear module 21. The output end of the seedling driving component 25 is connected to the seedling rod 24. The seedling rod 24 is located on the upper side of the rootstock clamp 22 and abuts against the base of the cotyledon 1001 of the rootstock 100.
[0054] Understandably, after the rootstock 100 cuts off one cotyledon 1001, one cotyledon 1001 remains. To prevent the remaining cotyledon 1001 from drooping, this embodiment provides a top seedling stem 24, the top of which abuts against the base of the cotyledon 1001 to support the cotyledon 1001 and prevent the rootstock 100 from tilting and affecting the connection.
[0055] One end of the top seedling rod 24 is connected to the output end of the top seedling drive unit 25, and the other end is bent and abuts against the rootstock 100 from the side facing the rootstock clamp 22. During the docking process between the rootstock 100 and the scion 101, the first linear module 21 drives the rootstock clamp 22 to move forward, and the top seedling drive unit 25 drives the top seedling rod 24 to move obliquely upward, so as to follow the displacement of the rootstock clamp 22 and push the cotyledon 1001 obliquely upward, lifting the cotyledon 1001 to prevent the rootstock 100 from drooping and affecting the subsequent clamping operation.
[0056] Specifically, the seedling drive component 25 can be a cylinder or a linear motor.
[0057] like Figure 2 As shown, the first clamping and transferring module 2 in this embodiment also includes a micrometer structure 26.
[0058] The micrometer structure 26 is located in the first linear module 21. The micrometer structure 26 has a screw head 261, which is driven to cooperate with the top seedling rod 24 to finely adjust the position of the top seedling rod 24.
[0059] Understandably, due to slight differences in the diameter of each batch of rootstock 100, when cutting rootstock 100 from the same batch, it is necessary to fine-tune the position of the top stem 24 relative to the rootstock 100 so that the top stem 24 is tangent to the base of the cotyledon stem of the rootstock 100. In this embodiment, the micrometer is used for fine-tuning the position of the top stem 24. In actual use, by rotating the screw head 261, the rotational motion is converted into a linear micro-displacement of the top stem 24, achieving precise control of the clamping force between the front end of the top stem 24 and the base of the cotyledon 1001 of the rootstock 100.
[0060] like Figure 3 As shown, the second clamping and transferring module 4 in this embodiment includes: a second linear module 41, a scion gripper 42, and a second rotary drive 43.
[0061] The second rotary drive 43 is disposed on the worktable 1. The output end of the second rotary drive 43 is connected to the second linear module 41. The slide of the second linear module 41 is connected to the scion clamp 42. The scion clamp 42 is used to clamp the scion 101.
[0062] The control module is used to control the rotation angle of the second rotary drive 43 based on the characteristic information of the scion cut.
[0063] Understandably, the scion gripper 42 is used to hold the cut scion 101. The gripper is equipped with a flexible pad to buffer the clamping force and avoid damage to the scion 101.
[0064] The second linear module 41 is used to drive the extension and retraction of the scion gripper 42, and the second rotary drive 43 is used to drive the rotation of the scion gripper 42. The second rotary drive 43 drives the scion gripper 42 to rotate from the scion photography area to the docking area, and the second linear module 41 drives the scion gripper 42 to extend in preparation for docking.
[0065] Based on the characteristic information of the scion cut, the control module controls the rotation angle of the second rotation drive 43. By fine-tuning the rotation angle, the scion cut is aligned with the scion cut.
[0066] Specifically, the second linear module 41 can be a ball screw slide, and the second rotary drive 43 can be a servo motor. A reducer is also configured between the second rotary drive 43 and the second linear module 41.
[0067] like Figure 3 As shown, the second clamping and transferring module 4 in this embodiment also includes a seedling-blocking rod 44.
[0068] The seedling blocking rod 44 is located on the second straight module 41, and the seedling blocking rod 44 is located on the lower side of the scion clamp 42.
[0069] To limit the cutting of the scion 101 and prevent it from tilting, this embodiment also includes a seedling-blocking rod 44 in the second clamping and transferring module 4. The seedling-blocking rod 44 is used to limit the stem of the scion 101 to prevent it from tilting during the rotation of the second clamping and transferring module 4, which would affect the grafting quality.
[0070] Because the seedling retainer 44 can rotate with the scion clamp 42, the seedling retainer 44 can always limit the stem of the scion 101 during the rotation stroke of the scion 101, ensuring that the scion 101 always maintains an upright posture.
[0071] like Figure 3 As shown, the second clamping and transfer module 4 in this embodiment also includes a micrometer structure.
[0072] The micrometer structure is located in the second linear module 41. The micrometer structure has a screw head, which is driven by the seedling stop bar 44 to finely adjust the position of the seedling stop bar 44.
[0073] Since the micrometer structure of the second clamping and transfer module 4 has the same function as the micrometer structure 26 of the first clamping and transfer module 2, it will not be described in detail here.
[0074] like Figure 4 and Figure 7 As shown, at least one of the second image sensor acquisition module 5 and the first image sensor acquisition module 3 in this embodiment includes a bracket 31, a camera 32 and a fill light 33.
[0075] The bracket 31 is detachably connected to the worktable 1. The camera 32 and the fill light 33 are both mounted on the bracket 31, and the fill light 33 is arranged around the camera 32.
[0076] Understandably, since the dimensions of rootstocks 100 and scions 101 vary from batch to batch, in order to adapt to the height of rootstocks 100 and scions 101 of different dimensions, the bracket 31 in this embodiment is detachably connected to the workbench 1. Different heights of bracket 31 can be replaced, and the position of bracket 31 on the workbench 1 can be adjusted to adjust the angle between the lens of camera 32 and the rootstocks 100 and scions 101, so that the lens of camera 32 is directly facing the cut of rootstocks 100 and scions 101.
[0077] Specifically, the workbench 1 in this embodiment is provided with multiple mounting holes, and the bracket 31 is provided with multiple mounting positions for mounting the camera 32 and the fill light 33.
[0078] Since the rootstock 100 is cut from the top, the camera 32 of the first image-sensitive acquisition module 3 in this embodiment is positioned from top to bottom opposite the rootstock cut. The scion 101 is cut from the stem, and the second image-sensitive acquisition module 5 in this embodiment is positioned from bottom to top opposite the scion cut.
[0079] In this embodiment, the fill light 33 is a ring-shaped soft fill light, arranged around the camera 32. The fill light 33 has a constant color temperature, which can eliminate the shadows of the rootstock 100, the reflections of the scion 101 stem, and the image interference caused by fluctuations in ambient light, ensuring that the cut images of the rootstock 100 and the scion 101 are clear and complete.
[0080] Secondly, such as Figure 5 As shown, this embodiment provides a docking method based on the above-described docking device for vegetable grafting, including the following steps: Step 511: Control the first image-sensitive acquisition module 3 to acquire rootstock incision image information, and control the second image-sensitive acquisition module 5 to acquire scion incision image information.
[0081] Step 512: Extract feature information based on the rootstock cut image information and scion cut image information. The feature information includes: the cut contour boundary, centroid coordinates, and effective cut area of rootstock 100 and scion 101, and establish a cut feature dataset.
[0082] Step 513: Based on the cut feature dataset, obtain the docking position deviation information of the rootstock cut and the scion cut.
[0083] Step 514: Based on the docking position deviation information, control the first clamping and transferring module 2 and the second clamping and transferring module 4 to dynamically correct the docking trajectory of the rootstock cut and the scion cut in the docking area, so as to realize the docking operation in which the center of the rootstock cut and the scion cut coincide.
[0084] Understandably, in this embodiment, for the cut rootstock 100 and scion 101, the first image-sensitive acquisition module 3 is used to acquire the rootstock cut image information, and the second image-sensitive acquisition module 5 is used to acquire the scion cut image information. Due to individual differences in the rootstock 100 and scion 101, the parameters of the cut images of the rootstock 100 and scion 101 in each batch have slight differences.
[0085] This embodiment extracts feature information from image data and establishes a feature dataset. The feature dataset includes feature information for all rootstocks 100 and scions 101 from the same batch. The cut contour boundary is the geometric basis for successful grafting of the rootstock 100 and scion 101. The centroid coordinates, as the first-order moment statistical feature of the cut area, are a numerical representation of the spatial position of the cut within the image plane. The effective cut area is a comprehensive indicator of cutting quality; its value directly reflects the integrity and usability of the cut. This embodiment simultaneously extracts three types of features—contour boundary, centroid coordinates, and effective cut area—and associates and binds the multidimensional features of the same rootstock 100 or scion 101 to form a digital feature vector for the cut, thus realizing the transformation from raw pixel images to structured data.
[0086] Based on the feature datasets of the rootstock and scion incisions, the deviation information of the docking positions of the rootstock and scion incisions is calculated. During the docking process, the relative positions of the first clamping and transferring module 2 and the second clamping and transferring module 4 are controlled to adjust the relative positions of the rootstock and scion incisions in real time, dynamically correct the docking trajectory of the rootstock and scion incisions, and achieve the coincidence of the centers of the rootstock and scion incisions.
[0087] Specifically, in this embodiment, the control module segments the contours of the rootstock and scion incisions using the YOLO-SEG model. Then, it searches for the adjustment angles of the rootstock and scion incisions using contour projection and a hierarchical search method. Based on the image pyramid mechanism, the search process is unfolded layer by layer from coarse to fine resolution. After determining the coarse displacement by traversing the entire image with a large step size at the coarse resolution layer, fine alignment is confined to a very small local window on the original resolution image. Within this window, the algorithm calculates the intersection-union ratio (IUU) of the scion mask and the rootstock mask at each candidate translation position with a step size, thereby adjusting the relative positions of the rootstock and scion incisions.
[0088] In actual operation, compared with the traditional docking device, the incision docking deviation of this embodiment is reduced from ±0.5mm to within ±0.05mm, the incision fit rate of rootstock 100 and scion 101 is increased by more than 10% on average, and the grafting survival rate is increased by 5%, which fully meets the needs of large-scale and standardized grafting production of vegetables.
[0089] like Figure 6 As shown, in the docking area, the first clamping and transferring module 2 and the second clamping and transferring module 4 rotate relative to each other to the docking position. In this embodiment, the first clamping and transferring module 2 and the second clamping and transferring module 4 can be controlled to adjust the rotation angle so that the geometric center of the rootstock 100 and the scion 101 completely coincides, so that the rootstock 100 and the scion 101 can be docked at a precise angle.
[0090] This embodiment can adaptively compensate for individual differences in rootstock 100 and scion 101, mechanical positioning errors, and cutting errors that cause cut position offsets, completely solving the problem of left-right and front-back misalignment of the cuts of rootstock 100 and scion. This significantly improves grafting accuracy and cut fit, effectively increasing the survival rate of grafted vegetables. Furthermore, this embodiment can accurately extract multi-dimensional feature parameters such as cut contour, centroid, and area, achieving digital recognition of cut status. Compared to traditional indiscriminate mechanical docking, it has a strong adaptability to different working conditions, adapting to seedlings with different growth stages and cutting conditions, making the equipment more versatile.
[0091] This embodiment extracts feature information based on the rootstock incision image information and the scion incision image information, including: Edge detection algorithms and morphological filtering are used to obtain the cut contour boundaries of rootstock 100 and scion 101, and to remove interference from hair and stem impurities.
[0092] The centroid coordinates are obtained by pixel integration of the contour region.
[0093] The effective cut area is obtained by converting the number of effective pixels within the statistical contour.
[0094] Understandably, the cut edges of the rootstock 100 and scion 101 appear as a step change in grayscale between adjacent pixels in the grayscale image. By using edge detection algorithms, such as the Canny operator or the Sobel operator, to extract local extrema and suppress non-maximum values of the grayscale gradient, the straight boundary and sharp corner of the cut surface can be accurately captured. However, the fuzz, epidermal wrinkles, and debris generated during cutting on the surface of the grafted seedling stem will form a large number of short, fragmented edges and closed pseudo-contours in the edge detection results. These pseudo-contours differ significantly from the true cut boundary in geometric features; the fuzzy edges are short, have drastic curvature changes, and have small closed areas. This embodiment uses morphological filtering, such as opening operations to remove isolated burrs and closing operations to connect broken edges, and uses structuring elements to perform morphological reconstruction on the binarized edge image. While preserving the continuity of the main outline of the cut, it effectively filters out the fluffy debris and stem pseudo-impurity edges with lengths and areas below the threshold, thereby obtaining a pure edge image composed only of the true cut outline boundaries of the rootstock 100 and scion 101, providing interference-free geometric primitives for subsequent contour matching and pose calculation.
[0095] After obtaining the clean cut contour boundaries, the cut contours of the rootstock 100 and scion 101 are transformed into closed regions using a contour filling algorithm. Coordinate integration is then performed on all valid pixels within the contour, i.e., the horizontal and vertical coordinates of each pixel within the region are summed and divided by the total number of pixels to obtain the centroid coordinates of the cut region. The centroid coordinates, as a numerical representation of the geometric center of the cut region, are insensitive to local edge noise and minor contour defects, and can stably reflect the spatial position of the cut in the image coordinate system.
[0096] After completing the contour filling, the effective cut area is obtained by traversing and counting the effective pixels inside the cut contours of the rootstock 100 and scion, and by converting the units using the camera 32 calibration parameters. The effective cut area is a key indicator for evaluating the quality of the cut. If the area value is within a preset reasonable range, it indicates that the cut is flat and the bevel is complete, meeting the bonding requirements. If the area is significantly smaller, it suggests that the cut has defects or the bevel is incomplete.
[0097] This embodiment obtains the docking position deviation information of the rootstock incision and scion incision based on the incision feature dataset, including: The positional deviation of the rootstock cut center relative to the center of the first clamping and transferring module 2 is obtained.
[0098] Obtain the positional deviation of the scion cut center relative to the center of the second clamping and transferring module 4.
[0099] Positional deviation includes lateral deviation (left and right) and longitudinal deviation (front and back), with the direction and value of the deviation.
[0100] Understandably, the control module in this embodiment includes a deviation calculation model. This model uses the coordinates of the center of the first clamping and transferring module 2 and the center of the second clamping and transferring module 4 as a reference to calculate the coordinate deviations between the center of the rootstock incision and the center of the first clamping and transferring module 2, and between the center of the scion incision and the center of the second clamping and transferring module 4. The coordinate deviations include the lateral and longitudinal deviations along the workbench 1, as well as the direction of the deviations.
[0101] Specifically, the positional deviation calculation includes lateral (left / right) deviation and longitudinal (front / back) deviation. The deviation calculation formulas are: lateral deviation ΔX = X-coordinate of the cut center - X-coordinate of the clamping and handling mechanism center; longitudinal deviation ΔY = Y-coordinate of the cut center - Y-coordinate of the clamping and handling mechanism center. When ΔX > 0, it is a right deviation; when ΔX < 0, it is a left deviation; and when ΔX = 0, there is no lateral deviation. The logic for determining longitudinal deviation is the same, and will not be elaborated here.
[0102] The docking method in this embodiment also includes: conducting a quality assessment on the grafted seedlings after docking, including the cutting success rate and the cut fit rate of the rootstock 100 and the scion 101.
[0103] Understandably, existing docking devices can only complete the docking operation, lack a quality quantification evaluation system for the cut, and cannot assess the grafting success rate during the docking stage.
[0104] This embodiment includes a data statistics and quantitative evaluation model capable of evaluating the cutting quality and grafting fit quality of rootstock and scion cuts. Specifically, the cutting success rate refers to determining whether a single grafting operation is qualified based on the integrity of the cut outline and area parameters, and counting the number of qualified cuts for a single operation and for the same batch of rootstock 100 and scion 101. The cut fit rate of rootstock 100 and scion 101 refers to the ratio of the effective overlapping area of the rootstock and scion cuts to the total cut area, used to quantitatively characterize the degree of fit and matching between the rootstock 100 and scion cuts.
[0105] This embodiment establishes a quantitative detection and evaluation system for grafting cutting quality and grafting fit quality by acquiring images of rootstock and scion incisions and evaluating the quality of grafting. It accurately counts the cutting success rate and the fit effect of the incision, enabling operators to accurately grasp the quality of batch grafting operations, which is conducive to equipment parameter optimization and standardized production control.
[0106] Before extracting feature information based on the rootstock incision image information and the scion incision image information in this embodiment, the following steps are also included: The rootstock and scion incision images are preprocessed. The original images are then converted to grayscale, denoised, and thresholded to remove impurities and shadow interference, resulting in a binarized image of the incision.
[0107] Understandably, after the first image sensor acquisition module 3 acquires the image information of the rootstock incision and the second image sensor acquisition module 5 acquires the image information of the scion incision, the image information needs to be preprocessed. Through grayscale conversion, noise reduction and threshold segmentation, impurities and shadow interference information in the image are removed to obtain a binarized image, so as to improve the feature recognition accuracy of the rootstock incision and the scion incision.
[0108] This embodiment employs a series of processes including grayscale conversion, noise reduction, and threshold segmentation to peel away the complex multi-dimensional interference factors such as color, texture, noise, and shadows from the original image, ultimately outputting a binarized image containing only the cut contour information. The binarized image features small data size, prominent target, clean background, and clear edges, and can be directly input into subsequent Hough transform line detection, contour extraction, or geometric parameter measurement algorithms. This significantly improves the accuracy, robustness, and real-time performance of cut recognition and positioning, providing a reliable visual information foundation for the automatic alignment and precision control of grafting machines.
[0109] 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 grafting device for vegetable grafting, characterized in that, include: The workbench has a rootstock photography area, a scion photography area, and a grafting area; A first clamping and transferring module is provided in the rootstock photographing area of the workbench. The first clamping and transferring module is used to carry and fix the cut rootstock and move the rootstock from the rootstock photographing area to the docking area. A first image-sensitive acquisition module is disposed opposite to the first clamping and transferring module. The first image-sensitive acquisition module is used to acquire image information of the rootstock incision. The second clamping and transfer module is located in the scion photography area of the workbench. The second clamping and transfer module is used to carry and fix the cut scion, and move the scion from the scion photography area to the docking area. The second image-sensitive acquisition module is disposed opposite to the second clamping and transferring module. The second image-sensitive acquisition module is used to acquire scion cut image information. The control module is electrically connected to the first clamping and transferring module, the first image-sensitive acquisition module, the second clamping and transferring module, and the second image-sensitive acquisition module, respectively. The control module is used to extract feature information of the rootstock incision and the scion incision based on the rootstock incision image information and the scion incision image information, and control the first clamping and transferring module and the second clamping and transferring module to perform the docking operation of the rootstock incision and the scion incision in the docking area according to the feature information.
2. The grafting device for vegetable grafting according to claim 1, characterized in that, The first clamping and transferring module includes: a first linear module, an anvil gripper, and a first rotary drive component; The first rotary drive is disposed on the worktable, the output end of the first rotary drive is connected to the first linear module, the slide of the first linear module is connected to the rootstock gripper, and the rootstock gripper is used to hold the rootstock. The control module is used to control the rotation angle of the first rotary drive component based on the feature information of the incision in the rootstock.
3. The grafting device for vegetable grafting according to claim 2, characterized in that, The first clamping and transferring module further includes: a seedling-supporting rod and a seedling-supporting drive component; The seedling-topping drive is located on the first linear module. The output end of the seedling-topping drive is connected to the seedling-topping rod. The seedling-topping rod is located on the upper side of the rootstock clamp and abuts against the base of the cotyledon of the rootstock.
4. The grafting device for vegetable grafting according to claim 3, characterized in that, The first clamping and transferring module also includes: a micrometer structure; The micrometer structure is located on the first linear module. The micrometer structure has a screw head, which is driven by the top seedling rod to finely adjust the position of the top seedling rod.
5. The grafting device for vegetable grafting according to claim 1, characterized in that, The second clamping and transferring module includes: a second linear module, a scion gripper, and a second rotary drive; The second rotary drive is disposed on the worktable, and the output end of the second rotary drive is connected to the second linear module. The slide of the second linear module is connected to the scion gripper, and the scion gripper is used to hold the scion. The control module is used to control the rotation angle of the second rotary drive component based on the feature information of the scion cut.
6. The grafting device for vegetable grafting according to claim 1, characterized in that, At least one of the second image-sensitive acquisition module and the first image-sensitive acquisition module includes a bracket, a camera, and a fill light; The bracket is detachably connected to the worktable. The camera and the fill light are both mounted on the bracket, and the fill light is arranged around the camera.
7. A docking method based on the docking device for vegetable grafting according to any one of claims 1 to 6, characterized in that, include: The first image-sensitive acquisition module is controlled to acquire rootstock incision image information, and the second image-sensitive acquisition module is controlled to acquire scion incision image information; Based on the rootstock cut image information and scion cut image information, feature information is extracted, including: the cut outline boundary, centroid coordinates, and effective cut area of the rootstock and scion, and a cut feature dataset is established; Based on the incision feature dataset, obtain the docking position deviation information of the rootstock incision and the scion incision; Based on the docking position deviation information, the first clamping and transferring module and the second clamping and transferring module are controlled to dynamically correct the docking trajectory of the rootstock incision and the scion incision in the docking area, so as to realize the docking operation in which the center of the rootstock incision and the scion incision coincide.
8. The docking method according to claim 7, characterized in that, The step of extracting feature information based on the rootstock incision image information and the scion incision image information includes: Edge detection algorithms and morphological filtering are used to obtain the cut contour boundaries of rootstock and scion, and to remove interference from hair and stem impurities; The centroid coordinates are obtained by pixel integration of the contour region; The effective cut area is obtained by calculating the number of effective pixels within the statistical contour.
9. The docking method according to claim 7, characterized in that, The step of obtaining the docking position deviation information of the rootstock incision and the scion incision based on the incision feature dataset includes: The positional deviation of the center of the rootstock incision relative to the center of the first clamping and transferring module is obtained; Obtain the positional deviation of the scion incision center relative to the center of the second clamping and transferring module; The positional deviation includes left-right lateral deviation and front-back longitudinal deviation, with the deviation direction and deviation value.
10. The docking method according to claim 7, characterized in that, Also includes: The quality of the grafted seedlings after grafting was assessed, including the success rate of cutting and the fit rate of the cuts between the rootstock and the scion. And / or, before extracting feature information based on the rootstock incision image information and the scion incision image information, the method further includes: The rootstock cut image information and scion cut image information are preprocessed. The original image is subjected to grayscale conversion, noise reduction and threshold segmentation to remove impurities and shadow interference in the image and obtain a binary image of the cut.
