Multifunctional plant grafting platform
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GUANGXI QINZHOU AGRI SCHOOL
- Filing Date
- 2025-07-23
- Publication Date
- 2026-08-07
AI Technical Summary
[0004]本发明提供一种多功能植物嫁接平台的一个目的是解决嫁接质量人工易漏检、病株易混入,以及人工检查效率低的问题
Smart Images

Figure CN120548880B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of plant grafting technology, and more specifically, this invention relates to a multifunctional plant grafting platform. Background Technology
[0002] Plant grafting technology is a core means of improving crop resistance, yield, and quality. However, traditional operation methods still have three major systemic defects: 1. Grafting quality relies on manual visual inspection, making it difficult to identify defects such as misaligned graft unions in real time, leading to defective seedlings entering the maintenance stage; diseased plants lack screening mechanisms before grafting, causing chain infections and deaths after mixing in, resulting in a large workload. 2. Grafting operations, tool disinfection, graft union quality inspection, and seedling irrigation are scattered on different platforms, making operation inconvenient. 3. Fixed-height operating platforms force operators of different body types to work in a bent-over position for extended periods, causing lumbar spine injuries and increasing the difficulty of labor. Based on the above-mentioned shortcomings, there is an urgent need to set up a plant grafting platform that is multifunctional and easy to use. Summary of the Invention
[0003] One object of the present invention is to address at least the aforementioned deficiencies and to provide at least the advantages that will be described later.
[0004] One purpose of this invention is to provide a multifunctional plant grafting platform to solve the problems of easy omissions in manual grafting quality inspection, easy mixing of diseased plants, and low efficiency of manual inspection.
[0005] This invention provides a multifunctional plant grafting platform, comprising: Platform entity; The zoning functional module includes a grafting operation area located on the first side of the platform body, equipped with electronic equipment fixed by clamps, lighting tubes, and a dual-camera system; and a plant treatment area located on the second side of the platform body, including an irrigation trough and a disinfection trough; the dual-camera system includes a first camera aimed at the exit of the grafting operation area for capturing images of the graft union after grafting; and a second camera aimed at the entrance of the grafting operation area for capturing images of the health status of the rootstock and scion to be grafted. The AI quality inspection module is executed by the processor and identifies the perpendicularity of the scion and rootstock based on the first camera image; identifies the characteristics of plant lesions based on the second camera image; generates anomaly prompts and outputs them to the electronic device display interface.
[0006] Preferably, a transparent positioning slot is provided at the exit of the grafting operation area, and a horizontal baseline is marked on the side wall of the support on both sides of the slot; an infrared pair tube is embedded in the slot, and the shooting is triggered when the plant stem is inserted to block the light beam; the first camera is set on the other side wall relative to the baseline, and the captured image covers the stem joint and the baseline.
[0007] Preferably, a V-shaped bracket is installed at the entrance of the grafting operation area, with the surface coated with a yellow matte RAL1021 coating; a 120° wide-angle second camera is fixed 20cm directly above the bracket to capture plant images.
[0008] Preferably, the AI quality inspection module extracts the contour line of the joint from the image and calculates the angle θ between it and the baseline; when |θ|>5°, it is determined that the verticality is abnormal.
[0009] Preferably, the AI quality inspection module calibrates the lighting based on a yellow background, converts the image to the HSV color model, and dynamically adjusts the saturation threshold to identify lesion areas; if the lesion area accounts for more than 5%, an alarm is triggered.
[0010] Preferably, the electronic device interface displays the entry and exit views of the grafting operation area in partitions; when an abnormality occurs, a red frame flashes at the edge of the corresponding view, and a warning icon is superimposed on the abnormal part until the user confirms.
[0011] Preferably, the irrigation trough is equipped with multiple sets of proximity sensors and multiple sets of downward-sloping injection pipes. Each set of proximity sensors corresponds to a set of injection pipes. The outlet of the injection pipe is located above the substrate of the grafted seedling tray. When any set of proximity sensors detects that a grafted seedling has been moved in, the corresponding set of injection pipes is automatically triggered to irrigate with nutrient solution or clean water.
[0012] Preferably, an infrared transmitter is embedded in the upper part of the side wall of the disinfection tank, and an infrared receiver is set at the corresponding position on the opposite side wall; When the sliding cover is closed, the metal frame of the cover blocks all light beams; or when a tool or hand is placed in, the light intensity of the blocked light beam is >80% and lasts for 0.5 seconds, which triggers the atomizing nozzle to spray alcohol spray; the atomizing nozzles are distributed in four staggered positions around the disinfection tank wall.
[0013] Preferably, the side sensor of the grafting operation area consists of two sets of infrared ranging sensors (accuracy ±2mm), which are horizontally installed at the edge of the platform 70cm above the ground. The first set of sensors measures the operator's waist height (60-100cm from the ground), and the second set measures the knee height (40-50cm from the ground). The processor calculates the platform's target height using the formula: H = 0.6 × H w +0.4×H k -5, in cm, control the lifting frame to adjust to the target height so that the grafting operation area is located 5±3cm below the operator's elbow.
[0014] The present invention has at least the following beneficial effects: This invention utilizes a dual-camera collaborative approach and AI dual-path analysis (dual analysis of pre-graft lesion identification and post-graft verticality detection) to achieve pre-graft plant health screening and post-graft grafting quality inspection, primarily aimed at improving grafting quality and reducing the risk of disease transmission.
[0015] This invention sets up grafting operations, tool disinfection, interface quality inspection, and seedling irrigation to be carried out on the same platform, which facilitates observation and control.
[0016] This invention features a proximity sensor-linked injection pipe within the irrigation trough, automatically irrigating grafted seedlings upon placement and reducing manual intervention. The disinfection trough disinfects both tools and hands, preventing contamination.
[0017] The lesion identification method of this invention uses the yellow coating of the V-shaped bracket as a reference for calibrating the illumination, and the dynamic threshold of the HSV color model adapts to changes in ambient light, which helps to reduce the false judgment rate.
[0018] The irrigation tank of this invention adopts a one-to-one combination module design (a set of sensors only triggers the corresponding injection pipe), which helps to avoid ineffective irrigation.
[0019] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Attached Figure Description
[0020] Figure 1 This is a structural schematic diagram of one implementation of the multifunctional plant grafting platform described in this invention.
[0021] The components include: support rod 1; second camera 2; lighting tube 3; pressure sensor 4; bracket 5; slot 6; infrared tube 7; support 8; baseline 9; first camera 10; irrigation trough 11; proximity sensor 12; injection pipe 13; combination module 14; platform body 15; disinfection tank 16; atomizing nozzle 17; infrared receiver 18; cover plate 19; lower edge 20; lifting frame 21; and irrigation trough cover plate 22. Detailed Implementation
[0022] The present invention will be further described in detail below with reference to embodiments, so that those skilled in the art can implement it based on the description.
[0023] It should be noted that, unless otherwise specified, the experimental methods described in the following embodiments are conventional methods, and the reagents and materials mentioned are commercially available. In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "setting" should be interpreted broadly. For example, they can refer to fixed connection or setting, detachable connection or setting, or integral connection or setting. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances. The terms "lateral," "longitudinal," "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, and are only for the convenience of describing this invention and simplifying the description. They 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, and therefore should not be construed as a limitation of this invention.
[0024] See Figure 1 As shown, this invention discloses a multifunctional plant grafting platform, which includes: Platform Entity 15; The zoning functional module includes a grafting operation area located on the first side of the platform body, which is equipped with electronic equipment fixed by clamps, lighting tubes 3, and a dual-camera system; and a plant treatment area located on the second side of the platform body, which includes an irrigation trough 11 and a disinfection trough 16; the dual-camera system includes a first camera 10 aimed at the exit of the grafting operation area for capturing images of the graft union of the plants after grafting; and a second camera 2 aimed at the entrance of the grafting operation area for capturing images of the health status of the rootstock and scion to be grafted. The AI quality inspection module is executed by the processor and, based on the image from the first camera, identifies the perpendicularity of the scion and rootstock; based on the image from the second camera, it identifies the characteristics of plant lesions; generates anomaly alert signals and outputs them to the electronic device display interface. The grafting operation area exit is equipped with a transparent positioning slot, and the side walls of the brackets 8 on both sides of the slot are marked with horizontal baselines 9. An infrared pair tube is embedded in the slot, and the shooting is triggered when the plant stem is inserted to block the light beam. The first camera 10 is set on the other side wall relative to the baseline 9, and the captured image covers the stem joint and the baseline. The entrance to the grafting operation area is equipped with a V-shaped bracket, which is coated with a yellow matte RAL1021 coating. A 120° wide-angle second camera is fixed 20cm directly above the bracket to capture images of the plant. The AI quality inspection module extracts the contour line of the joint area from the image and calculates the angle θ between it and the horizontal baseline; when |θ|>5°, it determines that the verticality is abnormal. The AI quality inspection module uses a yellow background as a reference to calibrate the lighting. After converting the image to the HSV color model, it dynamically adjusts the saturation threshold to identify lesion areas. An alarm is triggered if the lesion area accounts for more than 5%. The electronic device interface displays the entry and exit views of the grafting operation area in partitioned sections; when an anomaly occurs, a red frame flashes at the edge of the corresponding view, and a warning icon is superimposed on the abnormal part until the user confirms.
[0025] According to one embodiment of the present invention, the platform body is constructed of stainless steel welded to form a support platform, and the surface is coated with an epoxy resin anti-rust layer. The functional modules include: a grafting operation area located on the left long side of the platform body, equipped with adjustable pressure clamps (optionally pneumatic parallel grippers), LED strip lighting tubes, and a dual-camera system; and a plant treatment area located on the right long side, including a PVC irrigation tank and a 304 stainless steel disinfection tank. An acrylic transparent positioning slot is provided at the exit of the grafting operation area, with a 2mm wide horizontal baseline laser-etched on the side wall of the slot; a slotted infrared photocell is embedded in the slot, triggering the first camera 10 to capture images when a plant stem is inserted and blocks the light path. An aluminum alloy V-shaped bracket 5 is installed at the entrance of the grafting operation area, with a matte yellow coating of RAL1021 standard color. A 120° wide-angle lens camera is fixed 20cm above the bracket via a support rod 1 as a second camera. The electronic device can be an industrial tablet computer, with its interface divided into left and right sections: the right side displays a real-time view of the grafting operation area exit, and the left side displays a view of the plant treatment area entrance.
[0026] Horizontal baselines 9 are marked on the side walls of the brackets 8 on both sides of the card slot. These baselines are marked 5mm above the plant insertion path to ensure unobstructed image acquisition. The first camera uses a 5-megapixel CMOS sensor, mounted on the opposite side wall of the baseline, with its lens center axis at a 90° angle to the baseline, covering the stem joint and the entire baseline. An infrared pair 7 is embedded in the card slot. When the plant stem is inserted and obstructs the light beam, if the infrared pair detects continuous obstruction for ≥0.5 seconds, the first camera is triggered to capture an image within 10ms. The second camera is equipped with an F2.0 aperture and an automatic white balance module, continuously capturing plant images at 30fps against the yellow coating background of the V-shaped bracket. When the grafting tray is placed in the grafting operation area entrance bracket, its bottom pressure sensor 4 triggers a load greater than a preset threshold, initiating single-frame capture by the second camera (e.g., a preset threshold of 30N) or manual activation. Images from both cameras are transmitted to electronic devices via a USB 3.0 interface, with a resolution of 1920×1080 pixels.
[0027] Verticality detection algorithm: Extract the contour line of the joint area in the image and fit its principal axis equation; read the slope k of the baseline in the image coordinate system, and calculate the angle θ between the two lines θ=arctan|(k1-k2) / (1+k1k2)| (unit: degrees), where k1 is the slope of the principal axis of the grafting joint contour line, and k2 is the slope of the horizontal baseline marked on the side wall of the bracket on both sides of the slot. When θ>5°, the verticality is judged to be abnormal, and an E01 alarm code is sent to the operation interface. Lesion recognition algorithm: Extract the yellow background area of the V-shaped bracket (RGB∈[240,205,75]±5), convert the image to the HSV color model; extract the saturation value S of a single pixel in the image in the HSV color model. pixel The marker satisfies S pixel threshold The pixel area is designated as the lesion candidate area. The lesion pixel ratio η is calculated as (number of lesion pixels / total number of pixels) × 100%. When η > 5%, an E02 alarm code is sent. After receiving the alarm signal, the electronic device generates a 3px wide red flashing border at the edge of the corresponding view, and simultaneously overlays icons on the abnormal area: a tilt angle symbol is displayed when the verticality is abnormal, and a spot pattern is displayed when the lesion is abnormal, until the user clicks the confirmation button to deactivate the alarm.
[0028] This invention achieves a process of pre-grafting plant health screening and post-grafting interface quality inspection through partitioned collaborative design and dual-path image analysis; the combination of mechanical positioning and algorithm detection can reduce the missed detection rate of manual visual inspection; and the automated triggering mechanism reduces operation interruption and improves the continuity of operation.
[0029] 1. Origin of the perpendicularity test formula: The formula for the angle between two straight lines originates from the relationship between the slopes of two straight lines in analytical geometry: θ = arctan|(k1-k2) / (1+k1k2)|. Dimensional verification: The slope k is a dimensionless ratio (Δy / Δx), and the arctan output angle is in degrees (°), which conforms to the consistency of dimensions.
[0030] Extract the contour line of the junction: calculate the image gradient using the Sobel operator, use the pixel gray value change rate ≥50 (range 0-255) as the edge determination threshold, and connect continuous edge points to generate the contour line.
[0031] Fitting the principal axis: Apply the least squares method to the set of contour points to solve the linear equation y=k1x+b.
[0032] The slope k2 of the baseline is calculated using the image coordinates (x1, y1) and (x2, y2) of the laser-etched line: k2 = (y2 - y1) / (x2 - x1).
[0033] 2. Lesion recognition algorithm: Step 1: Illumination Compensation and Color Conversion: Extract the yellow background area of the V-shaped bracket (RGB∈[240,205,75]±5, calculate the RGB mean of this area as the baseline value (e.g., (245, 210, 80)); perform white balance correction on the entire image: R{norm} = R{pixel} × 128 / R{avg}; / / R{avg} is the mean R value of the yellow area (e.g., 245); G{norm} = G{pixel} × 128 / G{avg}; B{norm} = B{pixel} × 128 / B{avg}; The original RGB pixel values are R{pixel}, G{pixel}, and B{pixel}, and the corrected RGB values are R{norm}, G{norm}, and B{norm}. The corrected image is then converted to the HSV color model.
[0034] Step 2: Dynamic Threshold Setting: Calculate the average value V of the V channel (brightness) within the same yellow area. avg Calculate the saturation threshold S threshold : if V avg ≤ 100, S threshold = 40; if 100 <V avg <180, S threshold = 40 + 0.25 × (V avg - 100); / / Note: Coefficient 0.25 = (60-40) / (180-100); if V avg ≥ 180, S threshold = 60; Step 3: Lesion detection: Scan all pixels in the image and mark those that meet the S... pixel threshold For each pixel, the region marked with a pixel is used as a candidate region for lesions, and the percentage of lesion pixels is calculated as follows: η = (number of lesion pixels / total number of pixels) × 100%; Step 4: Alarm determination: If η>5% → trigger E02 alarm, the electronic device displays a red mask over the lesion area, flashing at a frequency of 2Hz.
[0035] Verticality detection example: Input: contour slope k1=0.35, baseline slope k2=0 (horizontal line).
[0036] Calculate: θ = arctan|(0.35-0) / (1+0.35×0)| = arctan(0.35) ≈ 19.3° Result: Because 19.3° > 5°, alarm E01 was triggered.
[0037] Example of lesion identification: Input data: RGB mean of the yellow area: (242, 208, 78); V value of the yellow area avg =152 (under strong fill light); Image resolution 1280×720 (total pixels 921600); Processing procedure: 1. White balance correction: R{norm}=R{pixel}×128 / 242; G{norm}=G{pixel}×128 / 208; B{norm}=B{pixel}×128 / 78; Eliminate color cast in lighting.
[0038] 2. Dynamic threshold calculation: Since 100 < 152 < 180; ∴S threshold =40+0.25×(152-100)=40+13=53; 3. Lesion detection: Scanning revealed 58,400 pixels satisfying S pixel <53; Percentage calculation: η = (58400 / 921600) × 100% ≈ 6.3%; Output result: Since 6.3% > 5% Therefore, the E02 alarm is triggered, and the interface displays: the plant image is overlaid with a red lesion mask, and the status bar displays "lesion percentage: 6.3% > threshold 5%".
[0039] According to another embodiment of the present invention, the irrigation trough is integrally molded from PVC material. An irrigation trough cover 22 is slidably mounted on the upper opening of the irrigation trough 11 along the platform plane, allowing for on-demand covering and use. For example, when the irrigation trough is not in use, the cover 22 covers the upper opening of the irrigation trough 11, creating a larger platform plane and preventing dust, impurities, or other interference. When in use, the upper opening is opened, and the grafted plant is placed in for irrigation. The bottom of the irrigation trough is designed with an inclination angle of 5°±1° to facilitate liquid collection and drainage. A set of proximity sensors 12 includes an infrared transmitter and an infrared receiver. Multiple sets of infrared transmitters and receivers, as well as multiple sets of injection pipes 13, are symmetrically installed on the inner walls of both sides of the trough. Each set of proximity sensors corresponds to one set of injection pipes. The injection pipes are made of 304 stainless steel with an inner diameter of 4mm, fixed to the side wall of the trough at a 30° inclination angle, and the outlet is 3cm above the surface of the grafting seedling substrate. The sensor signal line is connected to the PLC controller. When the grafted seedling is detected to have moved into the groove and blocked a pair of sensors for ≥0.5 seconds, it is determined to be a valid trigger signal.
[0040] The injection pipe connects to an external liquid supply system via a connector. The supply system can be controlled by a peristaltic pump or a solenoid valve. After receiving the sensor signal, the PLC controller starts the supply system after a 0.3-second delay, continuously irrigating with nutrient solution or clean water at a flow rate of 0.5 L / min for 5 seconds. An umbrella-shaped distributor (0.8 mm diameter × 12 holes) is installed at the outlet of the injection pipe to ensure the liquid evenly covers an area with a diameter of 10 cm. After irrigation, the residual liquid at the bottom of the tank is drained into a recovery container through the drain hole.
[0041] As shown in the figure, a set of proximity sensors 12 and a corresponding set of injection pipes 13 form a combined module 14. The system uses a set of sensors to detect which injection pipe is blocking the flow of liquid for irrigation, reducing waste and improving efficiency. After each irrigation cycle, the system automatically records the timestamp and the trigger sensor number, and historical records can be queried through the electronic device interface.
[0042] This invention achieves immediate irrigation of grafted seedlings after transplanting by linking a proximity sensor with a liquid injection mechanism, reducing manual intervention. The inclined tank and distributor design ensure uniform liquid penetration, and a dual anti-misjudgment mechanism reduces the risk of resource waste. This design helps maintain the stability of substrate moisture in the early stages of grafted seedling transplantation.
[0043] According to another embodiment of the present invention, the disinfection tank is made of 304 stainless steel, with a depth of 15cm and dimensions of 25cm × 20cm. An infrared transmitter is embedded 2cm below the upper edge of the left side wall of the tank, and an infrared receiver 18 is installed at the corresponding position on the upper edge of the right side wall. The optical axes of the transmitter and receiver are horizontally aligned with a spacing of 20cm. An aluminum alloy sliding cover 19 is provided on the top of the tank. The lower edge 20 of the cover 19 is 3-4cm wide, and when closed, it completely covers the opening of the tank and blocks all infrared beams.
[0044] The infrared receiver continuously monitors the light intensity signal, initially set to a 100% baseline value in an unobstructed state. When obstruction causes the light intensity to drop to ≤20% (i.e., obstruction rate ≥80%) for 0.5 seconds, the control circuit is triggered to start the disinfection program. Four atomizing nozzles 17 are staggered along the inner wall of the disinfection tank, installed at a height of 5-10cm from the bottom of the tank. After triggering, the nozzles spray 75% medical alcohol solution, with a single spray volume of 0.5ml and a duration of 2 seconds. Temperature sensors are embedded in the side wall of the disinfection tank; when the ambient temperature is >28℃, the spray function is automatically locked, and open flames are prohibited within a 1-meter radius.
[0045] This invention utilizes a dual-condition triggering mechanism based on infrared light intensity threshold and duration, which facilitates the identification of hand or tool placement actions; the staggered distribution of the atomizing nozzles 17 helps achieve disinfection without blind spots. This solution can reduce the risk of hand contamination during operation switching.
[0046] According to one embodiment of the present invention, two sets of infrared ranging sensors are horizontally installed at a height of 70cm ± 1cm above the ground at the four corner edges of the platform body. The first set of sensors is installed at the left and right ends of the front side (e.g., at the left and right ends of the grafting operation area), with a beam angle of 90°, focusing on measuring the operator's waist height (corresponding to a distance of 60-100cm from the ground); the second set of sensors is installed at the lower ends of the front side of the platform, tilted downwards at a 15° angle, measuring the knee height (corresponding to a distance of 40-50cm from the ground). The sensor housing is made of aluminum alloy with an IP65 protection rating. The measurement cycle is 10 times per second, and data is transmitted to the processor via an RS485 bus.
[0047] The processor performs height calculation: acquires the waist height measurement value H. w (Taking the average value from both left and right sensors) and knee height H k (Take the average value of the left and right sensors); Calculate the target height using the formula: H (cm) = 0.6 × H w +0.4×H k -5 (cm); The scissor-type or four-glue synchronous lifting frame 21 (an electric push rod with a rated load of 200kg can be selected) is adjusted to H±0.3cm within 5 seconds so that the grafting operation area is located 5±3cm below the operator's elbow. Dimensional verification: All variables are in cm, and the dimensions of the formulas are consistent.
[0048] The system is set to a maximum lifting speed of 3cm / s. Movement will immediately halt when an obstacle is detected within 10cm of the platform edge (detectable by an attached ultrasonic sensor). The operator can manually fine-tune the height by ±5cm via the electronic interface; manual adjustment has higher priority than automatic adjustment. The final height value and sensor data are recorded after each adjustment. Abnormal conditions (such as excessive motor current) trigger an E03 alarm code.
[0049] This invention utilizes multi-sensor collaborative measurement and ergonomic formula calculations to adapt to a comfortable operating height for operators of different body types; a safety protection mechanism avoids the risk of mechanical collisions. This design can reduce the load on the lower back during long-term work.
[0050] Calculation example: Input: H w =85cm (waist height), H k =45cm (knee height); Calculate: H = 0.6 × 85 + 0.4 × 45 - 5 = 51 + 18 - 5 = 64 cm; Output: Control the lifting frame to move to a height of 64cm, so that the grafting operation area is 5cm below the elbow.
[0051] The test results are as follows: Operator height Adjust the height below the elbow. Meets the requirement of 5±3cm 160cm 6.8cm yes It is evident that by using multi-sensor collaborative measurement and ergonomic formula calculations to adapt to the comfortable operating height for operators of different body types, it is beneficial to reduce the lumbar load during long-term work.
[0052] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. It can be applied to various fields suitable for the present invention. Further modifications can be readily implemented by those skilled in the art.
Claims
1. A multifunctional plant grafting platform, characterized in that, include: Platform entity; The zoning functional module includes a grafting operation area located on the first side of the platform body, the grafting operation area being equipped with electronic equipment fixed by clamps, lighting tubes, and a dual-camera system; the zoning functional module also includes a plant treatment area located on the second side of the platform body, the plant treatment area including an irrigation tank and a disinfection tank; the dual-camera system includes a first camera aimed at the exit of the grafting operation area for capturing images of the graft union after grafting; the dual-camera system also includes a second camera aimed at the entrance of the grafting operation area for capturing images of the health status of the rootstock and scion to be grafted; The AI quality inspection module, executed by the processor, is used to identify the perpendicularity of the scion and rootstock based on the images captured by the first camera; and to identify the characteristics of plant lesions based on the images captured by the second camera. An abnormality alert signal is generated and output to the display interface of the electronic device; The grafting operation area has a transparent positioning slot at its exit, and horizontal baselines are marked on the side walls of the support on both sides of the slot. An infrared pair is embedded in the slot, which triggers the shooting when the plant stem is inserted to block the light beam. The first camera is set on the other side wall opposite to the baseline, and the captured image covers the stem joint and the baseline. The entrance to the grafting operation area is equipped with a V-shaped bracket, the surface of which is coated with a yellow matte coating; a second camera with a 120° wide-angle lens is fixed 20cm directly above the bracket for capturing plant images; The AI quality inspection module extracts the contour line of the joint area from the image and calculates the angle θ between the contour line and the baseline; when |θ|>5°, it is determined that the verticality is abnormal. The AI quality inspection module calibrates the lighting based on a yellow background, converts the image to the HSV color model, and dynamically adjusts the saturation threshold to identify lesion areas. An alarm will be issued if the area of the lesion exceeds 5%; The interface of the electronic device displays the entrance view and the exit view of the grafting operation area in partitioned display; when an abnormality occurs, the edge of the corresponding view flashes a red frame, and a warning icon is superimposed on the abnormal part until the user confirms; The irrigation trough is equipped with multiple sets of proximity sensors and multiple sets of downward-sloping injection pipes. Each set of proximity sensors corresponds to a set of injection pipes, and the outlet of the injection pipe is located above the substrate of the grafting seedling tray. When any set of proximity sensors detects that a grafted seedling has been moved in, the corresponding set of injection pipes is automatically triggered to irrigate with nutrient solution or clean water. An infrared transmitter is embedded in the upper part of the side wall of the disinfection tank, and an infrared receiver is set at the corresponding position on the opposite side wall. When the sliding cover is closed, the metal frame of the cover blocks all the light beams. When a tool or hand is placed in, the light intensity of the blocked light beam is >80% and lasts for 0.5 seconds, which triggers the atomizing nozzle to spray alcohol. There are 4 atomizing nozzles distributed in a staggered manner around the tank wall of the disinfection tank. The side sensors of the grafting operation area consist of two sets of infrared ranging sensors, horizontally installed at a height of 70cm above the ground on the edge of the platform; the first set of sensors measures the operator's waist height, and the second set measures the operator's knee height; the processor calculates the target platform height according to the formula: H=0.6×H w +0.4×H k -5, in cm, control the lifting frame to adjust to the target height so that the grafting operation area is 5±3cm below the operator's elbow.
Citation Information
Patent Citations
Intelligent monitoring system for peony grafting
CN114882365A
Green plant disease identification and early warning management system and method
CN120126003A
Grafting cultivation device for passion fruit seedlings
CN210202638U
Temperature-controllable cultivation device for ornamental malus spectabilis grafted seedlings
CN213881066U
Workbench facilitating grafting of melon and fruit seedlings
CN219536911U