Weeding device and method based on vision and dynamic control

Abstract

The application discloses a weeding device and method based on vision and dynamic control, wherein the device comprises a support, a soil-breaking spike harrow and a soil-breaking driving mechanism; the soil-breaking driving mechanism comprises a first connecting rod, a second connecting rod, a first electric push rod and a second electric push rod; the upper end of the soil-breaking spike harrow is hinged to the support, and the lower end is hinged to the first connecting rod; the other end of the first connecting rod is hinged to the bottom of the second connecting rod; the upper end of the second connecting rod is hinged to a sliding block, and the sliding block is installed on a sliding rail of the support; the first electric push rod is hinged to the support, and the telescopic rod of the first electric push rod is hinged to the hinged position of the first connecting rod and the second connecting rod; the second electric push rod is hinged to the support, and the telescopic rod of the second electric push rod is hinged to the hinged position of the second connecting rod and the sliding block; the weeding device can accurately identify the soil area protruding from the water surface in the paddy field and perform efficient weeding, so that accurate identification and efficient operation on the soil area protruding from the water surface in the paddy field are realized.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of intelligent agriculture, and particularly relates to a weeding device and method based on vision and dynamic control. BACKGROUND

[0002] In rice planting, weeding is one of the basic means to ensure the healthy growth of rice. Rice competes with weeds for resources, and the growth of weeds not only affects the absorption of nutrients, water and light by rice, but also can breed diseases and pests, thereby reducing yield and quality. Therefore, timely and effective weeding is crucial to improving rice yield and improving crop growing environment.

[0003] With the mechanization of agricultural production, mechanized weeding technology has become one of the key means to improve efficiency and reduce labor costs in modern agriculture. Among many mechanical weeding devices, passive weeding machines have been widely used in rice field weeding work due to their simple structure and easy operation.

[0004] Passive weeding machines usually use a traction weeding wheel to work in the rice row, which can quickly remove most of the weeds and save time and effort for manual weeding. Although passive weeding machines are easy to operate, the weeding wheel often cannot effectively enter the soil surface on the soil protruding above the water surface, especially when the soil is relatively hard. The cutting effect of the weeding wheel is greatly reduced, and the cutting ability of the weeding wheel is limited, resulting in some weeds not being effectively removed. This can lead to the survival of weeds in some areas, which can destroy the uniformity of the crop growing environment and affect the stability of rice growth, and even adversely affect yield.

[0005] Therefore, in order to further improve the weeding efficiency and the uniformity of the rice growing environment, agricultural mechanization technology still needs continuous innovation, especially in the design, function improvement and operation mode optimization of weeding machinery, to find more efficient and accurate solutions to ensure the stability of the rice growing environment and the sustainable development of agricultural production SUMMARY

[0006] The present application overcomes the deficiencies of the prior art and provides a weeding device based on vision and dynamic control, which can accurately identify the soil area protruding above the water surface in the paddy field and perform efficient weeding, thereby achieving accurate identification and efficient operation of the soil area protruding above the water surface in the paddy field.

[0007] A second object of the present application is to provide a weeding method for soil protruding above the water surface in a paddy field based on vision and dynamic control.

[0008] The technical solution of the present application to solve the above technical problems is:

[0009] A weeding device based on vision and dynamic control, comprising a support, a visual identification system arranged on the support, a soil height detection system, a weeding system and a control system, wherein,

[0010] The weeding system comprises a soil-breaking harrow arranged on the support and a soil-breaking driving mechanism for driving the soil-breaking harrow to swing, wherein the soil-breaking driving mechanism comprises a first connecting rod, a second connecting rod, a first electric push rod and a second electric push rod, wherein the upper end of the soil-breaking harrow is hinged to the support, the lower end is hinged to one end of the first connecting rod, the other end of the first connecting rod is hinged to the lower end of the second connecting rod, the upper end of the second connecting rod is hinged to a sliding block, the sliding block is installed on a sliding rail of the support, and the sliding rail extends linearly; the base of the first electric push rod is hinged to the support, and the telescopic rod of the first electric push rod is hinged to the hinged position of the first connecting rod and the second connecting rod; the base of the second electric push rod is hinged to the support, and the telescopic rod of the second electric push rod is hinged to the hinged position of the second connecting rod and the sliding block;

[0011] The visual identification system is used to identify the soil protruding from the water surface in the paddy field, and send the area data of the identified soil to the control system;

[0012] The soil height detection system is used to detect the height of the soil in the area, and send the height data of the detected soil to the control system;

[0013] The control system is used to receive the area data of the soil sent by the visual identification system, control the soil height detection system to detect the height of the soil in the area, and control the weeding system to perform weeding operation according to the height data of the soil detected by the soil height detection system.

[0014] Preferably, the bottom of the support is provided with a depth limiting assembly, and the depth limiting assembly is a floating plate installed on the bottom of the support.

[0015] Preferably, the visual identification system is an industrial camera.

[0016] Preferably, the soil height detection system is an ultrasonic sensor.

[0017] A weeding method based on vision and dynamic control, comprising the following steps:

[0018] Step S1: acquiring a paddy field image;

[0019] Step S2: send the collected paddy field image into the constructed visual recognition model, identify whether there is soil protruding from the water surface in the paddy field image through the constructed visual recognition model; when it is identified that there is soil protruding from the water surface, mark the area of the soil protruding from the water surface; then, start the ultrasonic sensor, measure the height of the area of the soil protruding from the water surface through the ultrasonic sensor, to obtain the height data of the soil protruding from the water surface;

[0020] Step S3: the control system calculates the movement stroke and rotation angle of the first push rod and the second push rod through the obtained height data, to realize weeding operation on the area of the soil protruding from the water surface.

[0021] Preferably, in step S2, the construction step of the visual recognition model is:

[0022] Step S201: pretreat the collected paddy field image;

[0023] Step S202: extract features from the pretreated paddy field image, wherein the extracted features include color features, texture features and shape features;

[0024] Step S203: normalize the feature parameters in the color features, texture features and shape features, to construct a high-dimensional feature vector;

[0025] Step S204: input the high-dimensional feature vector as input into a neural network model, train and verify through the neural network model, to obtain a visual recognition model meeting the requirements.

[0026] Preferably, in step S202, the feature parameters in the color features include hue, saturation and brightness; the feature parameters in the texture features include contrast, energy and homogeneity; and the feature parameters in the shape features include perimeter, area and shape complexity.

[0027] Preferably, in step S2, the step of obtaining the height data of the soil protruding from the water surface is: the ultrasonic sensor emits an ultrasonic signal to the area where the soil is located, the ultrasonic signal is reflected back after encountering the soil surface, the ultrasonic sensor receives the reflected ultrasonic signal, and the distance between the ultrasonic sensor and the soil surface is calculated by using the time of signal propagation; then, the ultrasonic sensor repeatedly measures different positions of the area where the soil is located multiple times, to generate a height map of the soil surface, to display the height change of each detection point in the soil.

[0028] Preferably, in step S3, after receiving the height data, the control system first calculates the height difference between the height of each detection position and the water level, and judges whether the soil protrudes from the water surface and the degree of protrusion by comparing the height difference.

[0029] Preferably, in step S3, the step of performing weeding operation on the area of the soil protruding from the water surface is: when the industrial camera monitors the soil protruding from the water surface, the control system calculates the required extension stroke and rotation angle of the first push rod and the second push rod according to the height data measured by the ultrasonic sensor, so as to drive the soil breaking harrow to the target position to perform the weeding operation; when the industrial camera continuously monitors the working area and does not find the soil protruding from the water surface, the visual recognition system outputs the detection result to the control system, and confirms that the working area has been processed; after the control system receives the feedback signal, the first electric push rod and the second electric push rod are controlled to reset.

[0030] Compared with the prior art, the present application has the following advantages and beneficial effects:

[0031] 1. The weeding method based on vision and dynamic control of the present application combines visual recognition technology and deep learning model, which can accurately distinguish the soil area covered by the water layer in the paddy field and the hard soil area protruding from the water surface, so as to realize efficient targeted weeding operation.

[0032] 2. The weeding method based on vision and dynamic control of the present application measures the height of the soil in real time and feeds back to the control system; the control system can dynamically adjust the stroke and force of the first push rod and the second push rod according to the received data information, so as to ensure the accuracy of the soil breaking operation and the self-adaptability to different soil conditions, thereby greatly improving the weeding efficiency and operation quality.

[0033] 3. The weeding method based on vision and dynamic control of the present application designs a closed-loop feedback mechanism and an automatic reset function; after the weeding system completes the weeding operation, when the visual recognition system detects that the working area has been processed, the control system controls the weeding system to automatically reset to the standby state.

[0034] 4. The weeding device based on vision and dynamic control of the present application is highly automated, reduces manual intervention, and improves the stability, reliability and continuous operation ability of the equipment. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 Fig. 1 is a structural schematic diagram of the weeding device based on vision and dynamic control of the present application (not in working state).

[0036] Figure 2 Fig. 2 is a structural schematic diagram of the weeding device based on vision and dynamic control of the present application (in working state).

[0037] Figure 3 Fig. 3 is a schematic diagram of the sliding block.

[0038] Figure 4 Fig. 4 is a motion diagram of the weeding system when working.

[0039] Figure 5 Motion diagram for resetting the weeding system.

[0040] Figure 6 Flowchart of the weeding method based on visual and dynamic control of the present application.

[0041] In the figure: 1-industrial camera; 2-first electric push rod; 3-second electric push rod; 4-first connecting rod; 5-second connecting rod; 6-soil breaking harrow; 7-bracket; 8-floating plate; 9-ultrasonic sensor; 10-sliding block: DETAILED DESCRIPTION

[0042] The present application will be further described in conjunction with the embodiments and the accompanying drawings, but the embodiments of the present application are not limited thereto.

[0043] Referring to Figures 1-6 The weeding device based on visual and dynamic control of the present application comprises a bracket, a visual recognition system, a soil height detection system, a weeding system and a control system arranged on the bracket.

[0044] Referring to Figures 1-6 The weeding system comprises a soil breaking harrow arranged on the bracket and a soil breaking driving mechanism for driving the soil breaking harrow to swing, wherein the soil breaking driving mechanism comprises a first connecting rod, a second connecting rod, a first electric push rod and a second electric push rod, wherein the upper end of the soil breaking harrow is hinged to the bracket, the lower end of the soil breaking harrow is hinged to one end of the first connecting rod, the other end of the first connecting rod is hinged to the lower end of the second connecting rod, the upper end of the second connecting rod is hinged to a sliding block, the sliding block is installed on a sliding rail of the bracket, and the sliding rail extends linearly; the base of the first electric push rod is hinged to the bracket, and the telescopic rod of the first electric push rod is hinged to the hinged position of the first connecting rod and the second connecting rod; the base of the second electric push rod is hinged to the bracket, and the telescopic rod of the second electric push rod is hinged to the hinged position of the second connecting rod and the sliding block.

[0045] Referring to Figures 1-6 The visual recognition system is used for recognizing the soil protruding from the water surface in the paddy field and sending the area data of the recognized soil to the control system; in the present embodiment, the visual recognition system is an industrial camera.

[0046] Referring to Figures 1-6 The soil height detection system is used for detecting the soil height of the area where the soil is located and sending the height data of the detected soil to the control system; in the present embodiment, an ultrasonic sensor is arranged on the bracket.

[0047] Referring to Figures 1-6The control system is used for receiving the region data of the soil sent by the visual recognition system, controlling the soil height detection system to detect the height of the region where the soil is located, and controlling the weeding system to perform weeding operation according to the height data of the soil detected by the soil height detection system.

[0048] Referring to Figures 1-6 The bottom of the support is provided with a depth limiting component, and the depth limiting component is a floating plate installed at the bottom of the support.

[0049] Referring to Figures 1-6 The weeding method based on visual and dynamic control comprises the following steps:

[0050] Step S1: collecting a paddy field image;

[0051] Step S2: sending the collected paddy field image into a constructed visual recognition model, identifying whether there is soil protruding from the water surface in the paddy field image through the constructed visual recognition model, marking the region of the soil protruding from the water surface when it is identified that there is soil protruding from the water surface, and then starting an ultrasonic sensor to measure the height of the region of the soil protruding from the water surface through the ultrasonic sensor to obtain height data of the soil protruding from the water surface.

[0052] Step S3: a control system calculates the movement stroke and rotation angle of a first push rod and a second push rod through the obtained height data to realize weeding operation on the region of the soil protruding from the water surface.

[0053] Referring to Figures 1-6 In step S2, the construction step of the visual recognition model is:

[0054] Firstly, the collected paddy field image is preprocessed, and secondly, the paddy field image after preprocessing is feature extracted.

[0055] In the visual recognition system of the paddy field soil, the selection of the characteristic parameters is crucial for the reflection problem and the identification of the soil protruding from the water surface; in order to ensure that the soil region can be effectively identified in the complex paddy field environment, therefore, a plurality of characteristic parameters which are helpful for distinguishing the soil and the water surface are selected in the embodiment, and the calculation results of the characteristic parameters are taken as the input of the neural network model to improve the recognition accuracy; wherein, the extracted parameters are as follows: color feature, texture feature and shape feature.

[0056] 1) The color feature is extracted by converting the paddy field image from the RGB color space to the HSV color space, and extracting the characteristic parameters such as hue (Hue), saturation (Saturation) and brightness (Value) to help distinguish the soil and the water surface, specifically:

[0057] The RGB image is converted to an HSV image by the following formula:

[0058] C max = max(R, G, B) C min = min(R, G, B) Δ = C max -C min

[0059] For each pixel in the paddy field image, its hue (H), saturation (S) and lightness (V) are calculated, and then the mean and standard deviation of these feature parameters in the image region are calculated:

[0060]

[0061] In the formula: N is the number of pixels in the region, and Hi, Si, Vi are the hue, saturation and lightness values of the i-th pixel in the paddy field image, respectively.

[0062] 2) The texture feature is extracted by local binary pattern (LBP) or gray level co-occurrence matrix (GLCM). Specifically, for each pixel, the LBP value is calculated by comparing the pixel values of its surrounding neighborhood with the center pixel value:

[0063]

[0064] In the formula: s k is the binary value (0 or 1) generated after comparing with the center pixel value; P is the number of neighborhood pixels (usually 8 neighborhood pixels);

[0065] The gray level co-occurrence matrix (GLCM) is used to describe the spatial relationship of pixel gray levels in the paddy field image. The element M(i, j) in the gray level co-occurrence matrix (GLCM) represents the co-occurrence times of gray level i and gray level j at a given distance d and angle θ; the texture features, i.e. contrast, energy and homogeneity, are calculated by the gray level co-occurrence matrix (GLCM):

[0066]

[0067] 3) The shape feature is extracted by edge detection algorithm and contour analysis to extract the shape feature of the soil region, and further identify the convex part of the soil, wherein,

[0068] The Canny edge detection algorithm is used to obtain the edge region in the paddy field image:

[0069] Edges = Canny(I)

[0070] In the formula: I is the input image; Canny(I) is the edge image generated by the Canny algorithm.

[0071] The perimeter, area, shape complexity and other features of the soil region are calculated using contour analysis:

[0072]

[0073] The calculated feature parameters are normalized to form a high-dimensional feature vector, which is used as the input of the neural network model, wherein,

[0074] All the calculated features are integrated into a vector:

[0075] X = [H mean , H max , H min , Contrast, Energy, Homogeneity, Shape Features];

[0076] In the formula, the parameters in Features mainly include the total length of the target region boundary, the area of the target region, and the shape complexity (measured by the relationship between the area and the boundary length);

[0077] In order to avoid the influence of different feature parameters on the performance of the neural network model during training, the high-dimensional feature vector needs to be normalized:

[0078]

[0079] In the formula, μ and σ are the mean and standard deviation of the high-dimensional feature vector, X is the high-dimensional feature vector, and X norm is the normalized high-dimensional feature vector.

[0080] The normalized high-dimensional feature vector is used as the input of the neural network model:

[0081] Output = X norm

[0082] The neural network learns the relationship between these features and finally outputs the recognition result to determine the state of the soil region, i.e., "soil protruding from the water surface". At the same time, when the visual recognition system detects soil protruding from the water surface, it sends a signal to the control system, which starts the ultrasonic sensor to measure the soil height in the target region, specifically:

[0083] The ultrasonic sensor emits an ultrasonic signal to the surface of the soil. When the ultrasonic signal encounters the soil surface, it reflects back. After the ultrasonic sensor receives the reflected signal, it calculates the distance using the time difference between transmission and reception, i.e., the time difference between transmission and reception. The formula is as follows:

[0084]

[0085] In the formula: v is the propagation speed of ultrasonic waves in air, Δt is the time difference between the emission and reception of ultrasonic waves, and the unit is seconds; the distance from the ultrasonic sensor to the soil surface, i.e., the height information, is calculated.

[0086] The ultrasonic sensor measures different positions in the soil area multiple times and generates a height map of the soil surface to show the height changes of each detection point in the soil area.

[0087] The measurement data of the ultrasonic sensor need to be combined with the recognition results of the visual recognition model to accurately identify the protrusion degree of the soil; after the visual recognition system detects the soil, the task of the ultrasonic sensor is to provide the height information of the area where the soil is located. This process includes the following steps:

[0088] Through the visual recognition model, the area where the soil in the paddy field image is located is identified, and the boundary of the area is marked; the visual recognition model sends a signal to the control system indicating that further height measurement of the marked area is needed; the control system instructs the ultrasonic sensor to measure the height of the specific area according to the visual recognition area; the ultrasonic sensor obtains the height map of the area through distance measurement and transmits the height data to the control system.

[0089] After receiving the height data, the control system needs to further analyze the height data to determine whether the soil protrudes above the water surface and the degree of protrusion; this process includes the following steps:

[0090] The control system calculates the height difference between the height of each measurement point and the height of the water surface:

[0091] Height Difference = H soil -H water

[0092] In the formula: H soil is the height of the soil area, and H water is the height of the water surface.

[0093] By comparing these height difference values, the control system can determine whether the soil protrudes above the water surface and the degree of protrusion; that is, according to the height difference values, the control system determines whether the soil area is above the water surface, and determines the degree of soil protrusion according to the size of the height difference value; if the height difference value is greater than a predetermined threshold value, it indicates that the soil is more protruding, and special operation (such as using a soil breaking harrow for weeding) may be required; if the height difference value is smaller, it indicates that the soil is still covered by water, and the control system can adjust the subsequent operation (such as using the weeding wheel at the bottom of the support for weeding) according to this information; the control system determines the working stroke of the first electric push rod and the second electric push rod according to the analysis result of the degree of soil protrusion. Assuming that the degree of soil protrusion is large, the control system can instruct the first electric push rod and the second electric push rod to make a large adjustment; if the degree of protrusion is small, the action of the first electric push rod and the second electric push rod can be appropriately reduced to avoid unnecessary soil disturbance.

[0094] When the industrial camera detects soil protruding above the water surface, the ultrasonic sensor starts to work, measures the height data of the soil, and then the control system analyzes and processes the data to calculate the required extension stroke of the first electric push rod and the second electric push rod. The extension action of the first electric push rod and the second electric push rod is accurately controlled by electricity, so that it can gradually lower the first connecting rod, the second connecting rod and the soil breaking harrow until the extension rod of the first electric push rod and the second electric push rod reaches the target position. At this time, the second push rod will be locked first to reach the specified position, and then the second push rod will accurately adjust the position of the soil breaking harrow to ensure that the four-bar linkage mechanism (i.e. composed of the first connecting rod, the second connecting rod, the soil breaking harrow and the support) is in a dead point state, thereby realizing dynamic control.

[0095] In the above dynamic control process, the movement stroke and rotation angle of the first electric push rod and the second electric push rod need to be obtained to ensure accurate height adjustment of the soil breaking harrow, wherein the stroke of the first electric push rod can be calculated by the following formula, and the stroke of the second electric push rod can also be calculated by the following formula:

[0096]

[0097] In order to ensure that the soil breaking harrow is in the correct position, in addition to controlling the extension length of the first electric push rod and the second electric push rod, the rotation angle of the first electric push rod and the second electric push rod also needs to be controlled; the rotation angle of the first electric push rod and the second electric push rod can be calculated by the following formula:

[0098]

[0099] After the control system calculates the stroke and rotation angle of the first electric push rod and the second electric push rod, the position of the soil breaking harrow can be determined through the geometric position of the first electric push rod and the second electric push rod. Assuming that the soil breaking harrow is driven by the first electric push rod and the second electric push rod, its position can be represented as:

[0100] x = l2 new · cos (θ2) + l3 new · cos (θ3)

[0101] y = l2 new · sin (θ2) + l3 new · sin (θ3)

[0102] Where x and y represent the position of the soil breaking harrow on the horizontal plane.

[0103] When the industrial camera continuously monitors the working area and finds that the soil protruding above the water surface is no longer detected, the visual recognition system outputs a detection result, confirming that the target area has been processed. At this time, the control system receives the feedback signal, instructing the first electric push rod and the second electric push rod to enter the reset state; during the reset process, the first electric push rod and the second electric push rod begin to retract, gradually restoring the first connecting rod and the second connecting rod to the initial position; as the first electric push rod and the second electric push rod retract, the first connecting rod, the second connecting rod, and the soil breaking harrow gradually rise, and the soil breaking harrow is retracted from below the floating plate to above the floating plate and remains in a non-working state; this reset action ensures that the weeding device returns to standby state, preparing for the next weeding operation.

[0104] Through the above design, the reset action of the first electric push rod and the second electric push rod not only ensures the continuity and stability of the weeding system, but also effectively avoids the misoperation caused by the first electric push rod and the second electric push rod not being reset; the entire weeding process is completed by the visual recognition system and the control system, realizing automatic closed-loop control of weeding operation, and thereby improving the operation efficiency and system reliability.

[0105] In addition, the present application can accurately identify the water layer covered soil and hard soil area protruding above the water surface in the paddy field through the industrial camera combined with the deep learning model and the reflection suppression technology, and intelligently switch the operation mode according to the soil state: using the weeding wheel to cut weeds in the water layer covered area, and driving the soil breaking harrow by the first electric push rod and the second electric push rod for soil breaking and weeding in the hard soil area protruding above the water surface.

[0106] In addition, the application combines the ultrasonic sensor to measure the soil height in real time, and the control system dynamically adjusts the extension stroke of the first electric push rod and the second electric push rod, so as to ensure the accuracy and efficiency of the weeding operation; the closed-loop control mechanism and the automatic reset function adopted by the application ensure the stable operation and continuous operation of the weeding system, thereby significantly improving the operation quality of the whole area weeding.

[0107] The above is the preferred embodiment of the application, but the embodiments of the application are not limited by the above, and any changes, modifications, substitutions, combinations, simplifications made without departing from the spirit and principles of the application shall be equivalent replacement methods and shall be included in the protection scope of the application.

Claims

1. A method of weeding based on vision and dynamic control, characterized by, The application discloses a weeding device, which comprises a support, a visual identification system, a soil height detection system, a weeding system and a control system, wherein the weeding system comprises a soil breaking harrow arranged on the support and a soil breaking driving mechanism for driving the soil breaking harrow to swing, the soil breaking driving mechanism comprises a first connecting rod, a second connecting rod, a first electric push rod and a second electric push rod, the upper end of the soil breaking harrow is hinged to the support, the lower end of the soil breaking harrow is hinged to one end of the first connecting rod, the other end of the first connecting rod is hinged to the lower end of the second connecting rod, the upper end of the second connecting rod is hinged to a sliding block, the sliding block is arranged on a sliding rail of the support, and the sliding rail extends linearly; the base of the first electric push rod is hinged to the support, the telescopic rod of the first electric push rod is hinged to the hinged position of the first connecting rod and the second connecting rod; the base of the second electric push rod is hinged to the support, and the telescopic rod of the second electric push rod is hinged to the hinged position of the second connecting rod and the sliding block; the visual identification system is used for identifying the soil protruding from the water surface in the paddy field and sending the area data of the identified soil to the control system; the soil height detection system is used for detecting the height of the soil in the area and sending the height data of the detected soil to the control system; and the control system is used for receiving the area data of the soil sent by the visual identification system, controlling the soil height detection system to detect the height of the soil in the area and controlling the weeding system to perform weeding operation according to the height data of the soil detected by the soil height detection system. Specifically, the method comprises the following steps: Step S1: collecting a paddy field image; Step S2: sending the collected paddy field image to a constructed visual identification model, identifying whether the soil protruding from the water surface exists in the paddy field image through the constructed visual identification model, marking the area of the soil protruding from the water surface when the soil protruding from the water surface is identified to exist, and then starting an ultrasonic sensor to measure the height of the area of the soil protruding from the water surface through the ultrasonic sensor to obtain the height data of the soil protruding from the water surface; Step S3: the control system calculates the movement stroke and rotation angle of the first push rod and the second push rod through the obtained height data, and realizes the weeding operation on the area of the soil protruding from the water surface.

2. The method of claim 1, wherein the method is a method of weed control based on vision and dynamic control, characterized by, The bottom of the support is provided with a depth limiting component, and the depth limiting component is a floating plate arranged on the bottom of the support.

3. The visual and dynamic control-based weeding method according to claim 2, characterized by, The visual identification system is an industrial camera.

4. The method of claim 3, wherein the method is characterized by, The soil height detection system is an ultrasonic sensor.

5. The visual and dynamic control based weeding method according to claim 4, characterized in that, In step S2, the construction steps of the visual identification model are as follows: Step S201: preprocessing the collected paddy field image; Step S202: extracting features from the preprocessed paddy field image, wherein the extracted features include color features, texture features and shape features; Step S203: normalizing the feature parameters in the color features, the texture features and the shape features to construct a high-dimensional feature vector. Step S204: input the high-dimensional feature vector as input into the neural network model, train and verify the neural network model, and obtain a visual recognition model meeting the requirements.

6. The visual and dynamic control-based weeding method according to claim 5, characterized by, In step S202, the feature parameters in the color feature include hue, saturation, and brightness; the feature parameters in the texture feature include contrast, energy, and homogeneity; and the feature parameters in the shape feature include perimeter, area, and shape complexity.

7. The visual and dynamic control-based weeding method according to claim 6, characterized by, In step S2, the step of obtaining the height data of the soil protruding from the water surface is: an ultrasonic sensor emits an ultrasonic signal to the area where the soil is located, the ultrasonic signal is reflected back after encountering the soil surface, and the ultrasonic sensor calculates the distance between the ultrasonic sensor and the soil surface by using the time of signal propagation after receiving the reflected ultrasonic signal; then, the ultrasonic sensor repeatedly measures different positions of the area where the soil is located multiple times to generate a height map of the soil surface to show the height changes of each detection point in the soil.

8. The visual and dynamic control based weeding method according to claim 7, wherein, In step S3, after receiving the height data, the control system first calculates the height difference between the height of each detection position and the water surface height, and judges whether the soil protrudes from the water surface and the degree of protrusion by comparing the height difference.

9. The visual and dynamic control based weeding method according to claim 8, characterized by, In step S3, the step of performing weeding on the area of the soil protruding from the water surface is: when the industrial camera detects soil protruding from the water surface, the control system calculates the required extension stroke and rotation angle of the first push rod and the second push rod based on the height data measured by the ultrasonic sensor, so as to drive the soil-breaking harrow to the target position for weeding; when the industrial camera continuously monitors the working area and does not find soil protruding from the water surface, the visual recognition system outputs the detection result to the control system to confirm that the working area has been processed; After receiving the feedback signal, the control system controls the first electric push rod and the second electric push rod to reset.

Images