Crop positioning and double-precision gridded precision spraying and weeding device and method

Abstract

The present invention provides a crop positioning and double-precision gridded precision spraying and weed control device and method. The present invention uses specific plant tags to mark vegetable plants during transplanting, so that they have machine-readable signal characteristics, and uses the device developed by the present invention to distinguish between crops and weeds in real time, so as to map the weeds and divide them into spraying units and counting units in the form of double-precision grids. When performing double-precision grid division, the first precision is the spraying precision of the nozzle, and the grids divided with this precision correspond to each nozzle of the integrated spraying module, which are spraying units; the second precision is the movement precision of the independent spraying module, which is specifically used to deal with weeds that cannot be sprayed under the first precision. The grids divided under this precision need to be combined into new spraying units with a specific algorithm, which are called counting units. Based on the double-precision weed division method proposed by the present invention, combined with the two methods of integrated spraying and independent spraying, a higher weed control rate can be achieved without reducing the operating speed of the device without increasing the resolution of the nozzle.

Description

Technical Field

[0001] The present invention relates to the field of smart agricultural technology, and in particular to a crop positioning and double-precision gridded precision spraying and weeding device and method. Background Art

[0002] Weeds compete with crops for sunlight, nutrients and other resources, affecting their normal growth. Vegetable crops have a short growth cycle, and if weeds in vegetable fields are not effectively controlled after transplanting, their yield will be seriously affected.

[0003] Mechanical tillage has been a long-standing weed control method, loosening the soil and removing weeds between rows. However, it cannot remove weeds between plants. Manual weeding, a traditional weed control method, is widely used to remove weeds between vegetable plants. However, according to researchers, manual weeding is inefficient and prone to errors, removing only 65% to 85% of weeds on average. Furthermore, manual weeding is expensive, approximately five times the cost of traditional tillage. Therefore, manual weeding is unsustainable in the long term. Herbicide spraying is currently an effective weed control method, but the negative environmental impact of large-scale herbicide spraying cannot be ignored. Furthermore, herbicide residues on crops can affect their quality, and studies have shown that consuming food contaminated with herbicide residues is harmful to human and animal health.

[0004] In contrast, precision herbicide application can achieve both cost-effective and efficient weed control while effectively minimizing the negative impacts of herbicides. Studies have shown that precision spraying can save 30-70% of herbicides. In recent years, national policies have continuously emphasized promoting green agricultural development, reducing pesticide use, promoting green agricultural and rural development, strengthening comprehensive management of agricultural non-point source pollution, and further promoting the reduction of agricultural inputs, including reducing the use of chemical pesticides.

[0005] In summary, it is necessary to develop a stable and powerful inter-row weeding robot to detect weeds between vegetable crops and perform precise spraying and weeding.

[0006] To identify weeds, researchers have proposed methods for classifying crops and weeds using different features (such as color, size, and spectral reflectance). However, due to the complex natural environment, these techniques are difficult to apply in real-world situations. In recent years, deep learning has been increasingly used in computer vision, and weed identification devices based on deep vision methods have achieved high accuracy. However, these deep learning-based identification schemes require pre-acquisition of sample sets, and model training takes a long time, which brings many inconveniences to practical applications.

[0007] To address the task of weed removal, researchers have proposed using mobile weeders that can avoid seedlings to remove weeds within crop fields. However, because weeders damage ground infrastructure during operation, they cannot be used in fields with electrical pipes or lines, and their operation consumes additional energy. In contrast, precision herbicide spraying offers cost-effective and highly adaptable weed control. However, due to nozzle resolution limitations, single-precision weeding devices based on fixed nozzles cannot effectively remove weeds that are too close to crops, and their weed control effectiveness needs to be improved.

[0008] There is currently no effective solution to the problems of related technologies. Summary of the Invention

[0009] In order to overcome the deficiencies of the prior art, the present invention aims to provide a crop positioning and double-precision gridded precision spraying and weeding device and method.

[0010] To achieve the above object, the present invention provides the following solutions:

[0011] A crop positioning and dual-precision gridded precision spraying and weeding device is used in vegetable fields after transplanting. The roots of the crops in the vegetable fields are provided with plant tags. The device is mounted on a movable chassis and includes: an imaging chamber, a color code sensor, an industrial camera, a medicine box, a spray chamber, an integrated spray module, a cover plate, a master control box, an independent spray module, a single-chip microcomputer development module, and a microcomputer.

[0012] A plurality of color mark sensors are arranged at the bottom of the imaging chamber; the color mark sensors are connected to the microcomputer through the single-chip development module; the color mark sensor is used to detect the plant tag and generate a detection signal; the industrial camera is provided at the top inner side of the imaging chamber; the industrial camera is connected to the microcomputer; the medicine box is provided at the top outer side of the imaging chamber; the medicine box is respectively connected to the integrated spraying module and the independent spraying module; a spraying chamber is also provided on one side of the imaging chamber; the integrated spraying module is provided in the spraying chamber; the integrated spraying module and the independent spraying module are both connected to the single-chip development module; the cover plate is provided on the top of the spraying chamber; the main control box and the independent spraying module are provided on the cover plate; the single-chip development module and the microcomputer are connected to each other inside the main control box; the single-chip development module is used to receive the detection signal and generate a detection signal according to the number of the corresponding color mark sensor The detection signal is marked to obtain a marking signal; the microcomputer is used to control the industrial camera to obtain photographed images of crop plants and weeds according to the marking signal, and perform image processing on the photographed images to obtain processed images, and identify the processed images to obtain pixel position information of weeds and crops; the microcomputer is also used to divide the processed image into multiple spraying units according to the resolution of the integrated spraying module based on the processed image and the pixel position information, and divide the spraying unit into multiple counting units according to the movement accuracy of the independent spraying module, and define a spraying flag for each counting unit, and distribute spraying according to the spraying flag and the spraying unit to obtain integrated spraying information, and distribute spraying according to the spraying flag and the counting unit to obtain independent spraying information; the microcomputer is also used to accurately spray the spraying units according to the integrated spraying information and the independent spraying information respectively.

[0013] Preferably, it also includes: a black light-absorbing curtain, a transformer module, a battery and a plurality of white LED lights; the black light-absorbing curtain covers the four sides of the imaging room; the white LED lights are arranged inside the imaging room; the white LED lights are connected to the batteries through the transformer module; the transformer module is arranged in the main control box; the transformer module is used to adjust the brightness of the white LED lights.

[0014] Preferably, it further comprises a relay module; the relay module is arranged in the master control box;

[0015] The single-chip microcomputer development module includes a first single-chip microcomputer development board, a second single-chip microcomputer development board and a third single-chip microcomputer development board; the first single-chip microcomputer development board, the second single-chip microcomputer development board and the third single-chip microcomputer development board are all connected to the microcomputer; the first single-chip microcomputer development board is connected to the color mark sensor; the second single-chip microcomputer development board is connected to the input end of the integrated spraying module through the relay module; the signal output end of the integrated spraying module and the signal output end of the independent spraying module are both connected to the third single-chip microcomputer development board.

[0016] Preferably, the integrated spray module includes a medicine supply pipe, a first mounting bracket and a plurality of first intelligent nozzles; one end of the medicine supply pipe is connected to the medicine box; the other end of the medicine supply pipe is connected to each of the intelligent nozzles; the first mounting bracket is arranged in the spray room; and the intelligent nozzle is arranged on the first mounting bracket.

[0017] Preferably, the independent spraying module includes a second mounting frame, a robotic arm, a bearing, a servo motor and a second intelligent spray head;

[0018] The second mounting frame is connected to one end of the robotic arm through the bearing, and the servo motor is provided on the robotic arm; the other end of the robotic arm is connected to the second intelligent nozzle; the servo motor is connected to the third single-chip microcomputer development board; the servo motor is used to control the spray angle of the second intelligent nozzle.

[0019] Preferably, any one of the first smart nozzle and the second smart nozzle comprises a solenoid valve, a right-angle adapter and an adjustable nozzle;

[0020] The solenoid valve is connected to the adjustable nozzle through the right-angle adapter; the solenoid valve is connected to the relay module; the relay module is used to control the opening and closing state of the solenoid valve to achieve the opening and closing state of the adjustable nozzle.

[0021] Preferably, a liquid medicine pump is provided in the medicine box; the liquid medicine pump is used to ensure that the medicine box supplies medicine to the integrated spraying module and the independent spraying module at a constant pressure.

[0022] A crop positioning and double-precision gridding precision spraying and weeding method is applied to the above-mentioned crop positioning and double-precision gridding precision spraying and weeding device, and the method includes:

[0023] The device is mounted on a movable chassis and controlled to move at a constant speed through a vegetable field, with the crop row located between the imaging chamber and the spraying chamber. A color code sensor is used to continuously detect passing objects. When the color code sensor detects a plant tag, a high-level detection signal is output based on the color code sensor, and the detection signal is transmitted to a single-chip microcomputer development module.

[0024] When the single-chip microcomputer development module receives the detection signal, it sends the number of the color mark sensor that generated the signal to the microcomputer via serial port communication. When the microcomputer receives the serial port data from the single-chip microcomputer development module, it controls the industrial camera to capture images of the crop plants and weeds around the crop plants, crops the images into a square so that the ground width within the image is 12 cm, and calculates the coordinates of the plant tags in the image based on the cropped images and the signal information from the color mark sensor.

[0025] Converting the RGB channels of the captured image into HSV channels, and setting a threshold for each HSV channel according to lighting conditions to segment a binary image of the green plants in the image;

[0026] Performing an opening operation on the binary image using a preset structural element to remove noise connected domains with smaller pixel areas in the image while maintaining the sizes of connected domains of crops and weeds in the image stable, thereby obtaining a cleaned image;

[0027] Based on the cleared image, traverse all pixels in the cleared image, find pixels with a value of 1 and connected domains formed by them, record their position information, and obtain a recorded image;

[0028] Based on the coordinates of the plant tag in the image and the recorded image, a circular tolerance zone with a radius of 7.5 mm is drawn in the image with the coordinates of the plant tag as the center of the circle, and a connected domain intersecting the circular tolerance zone is determined as a connected domain of the crop, all pixels within the connected domain of the crop are marked as crop pixels and their position information is recorded, and the remaining connected domains outside the connected domain of the crop are determined as a connected domain of weeds, all pixels in the connected domain of the weeds are marked as weed pixels and the position information of the weed pixels is recorded, so as to obtain pixel position information of the weeds and the crop;

[0029] Based on the cleared image, the cleared image is evenly divided into a plurality of regions using a resolution of the integrated spraying module as a first precision, and the regions are used as spraying units, and the spraying units are evenly divided into a plurality of counting units using a movement precision of the independent spraying module as a second precision;

[0030] A spray flag is defined for each counting unit, and a two-dimensional array is established based on the spray flag, and all elements of the two-dimensional array are initialized to -1; the value of the spray flag represents the spray priority of the counting unit; and the two-dimensional array is used to store the value of the spray flag of the counting unit;

[0031] Traverse each unit to be sprayed and make the first allocation to the unit to be sprayed: if the unit to be sprayed contains weeds but no crops, it is a unit that needs to be weeded and will not affect the crops when spraying, record its position information and assign the spray flag of the counting unit inside it to 1, indicating that it is allocated as a unit that needs to be sprayed when the spray unit is first allocated; if the unit to be sprayed contains weeds and crops, it is a unit that needs to be weeded but will affect the crops when spraying, and needs to be divided again with higher precision, record its position information and keep the spray flag of the counting unit inside it at -1, indicating that it is not allocated when the spray unit is first allocated; if If a unit to be sprayed contains neither weeds nor crops, it is a unit that does not require weeding and will not affect crops when sprayed. Its location information is recorded and the spray flag of the counting unit within it is set to 0, indicating that it was assigned as a unit that does not require spraying when the spraying unit was first allocated. If a spraying unit contains no weeds but contains crops, it is a unit that does not require weeding but will affect crops when sprayed. It needs to be divided again with higher precision, its location information is recorded, and the spray flag of the counting unit within it is kept unchanged at -1, indicating that it was not assigned when the spraying unit was first allocated. Based on the results of the first spraying unit allocation, integrated spraying information is generated.

[0032] Traverse each undefined counting unit and define it: if the counting unit contains weeds but no crops, assign its spraying flag bit to 2, indicating that a second allocation is required for this counting unit; if the counting unit contains weeds and crops, assign its spraying flag bit to -3, indicating that a second allocation is prohibited for this counting unit; if the counting unit contains no weeds but contains crops, assign its spraying flag bit to -2, indicating that a second allocation is prohibited for this counting unit; if the counting unit contains no weeds and no crops, assign its spraying flag bit to 0, indicating that a second allocation is not required but allowed for this counting unit;

[0033] According to the result of traversing each undefined counting unit and defining it, the counting units that need to be reallocated to the spraying unit are allocated a second time in the order from top to bottom and from left to right, and new spraying units with the same size as the grid divided with the first precision are generated, and independent spraying information is generated according to the new spraying units;

[0034] Based on the integrated spraying information, the integrated spraying information is compiled line by line into a spraying command code and sent to the single-chip microcomputer development module via serial communication. After receiving the spraying command code, the integrated spraying module calculates the opening and closing time of each nozzle in the integrated spraying module based on the travel speed of the device to control the opening and closing of the nozzles and accurately spray the first assigned spraying unit.

[0035] According to the independent spraying information, the independent spraying information is compiled into a spraying command code line by line and sent to the single-chip microcomputer development module through serial communication; the single-chip microcomputer development module calculates the position and spraying time information of the second-allocated spraying unit according to a preset program and the received spraying command code; according to the calculated position and spraying time information of the second-allocated spraying unit, a specific PWM waveform is output to control the nozzle of the independent spraying module to point to the spraying unit, and to control the opening and closing of the nozzle to accurately spray the second-allocated spraying unit.

[0036] Preferably, the calculation formula of the plant tag is:

[0037]

[0038] Where x is the horizontal resolution of the image, y is the vertical resolution of the image, Ser_1 represents the number of the first color sensor that detects the plant tag, Ser_2 represents the number of the second color sensor that detects the plant tag, dt represents the time difference between the two color sensors detecting the crop tag, v represents the forward speed of the device, and point_signal represents the coordinates of the plant tag in the image.

[0039] Preferably, the calculation formula for segmenting the binary image of green plants in the image is:

[0040]

[0041] Among them, O hsv (x,y) represents the original image captured in HSV channels, [O hsv (x,y)] h,s,v Represent the hue, saturation and brightness of the original image respectively, T h , T s and T v The thresholds defined for each channel are divided into maximum and minimum values. The size of the threshold depends on the actual lighting conditions, soil color, etc. t (x,y) represents the binary image after color segmentation.

[0042] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0043] This invention provides a crop positioning and dual-precision gridded precision spraying and weed control device and method. The method of marking crop plants with plant tags simplifies the crop and weed classification algorithm, improving weed identification accuracy and device operation speed. Furthermore, the use of a color code sensor effectively addresses situations where plant tags are obscured by weeds, improving device stability. This invention combines signal transmission technology with computer vision technology for weed identification, achieving rapid and accurate weed identification while reducing device costs. Furthermore, using a precision spraying solution, each nozzle has a very small coverage area, enabling more precise spraying, saving pesticides while reducing environmental pollution. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0045] Figure 1 1 is a schematic diagram of the overall structure of an embodiment of the present invention;

[0046] Figure 2 is a front schematic diagram of an embodiment of the present invention;

[0047] Figure 3 is a schematic diagram on the right side of an embodiment of the present invention;

[0048] Figure 4 is an enlarged schematic diagram of a local control structure of an embodiment of the present invention;

[0049] Figure 5 This is an enlarged schematic diagram of the first intelligent nozzle installation structure according to an embodiment of the present invention;

[0050] Figure 6 is an enlarged schematic diagram of the second intelligent sprinkler installation structure according to an embodiment of the present invention;

[0051] Figure 7 is an enlarged schematic diagram of the servo motor structure according to an embodiment of the present invention;

[0052] Figure 8 is an enlarged schematic diagram of the adjustable nozzle structure according to an embodiment of the present invention;

[0053] Figure 9 This is an enlarged schematic diagram of a plant tag structure according to an embodiment of the present invention;

[0054] Figure 10 is a flow chart of a method according to an embodiment of the present invention;

[0055] Figure 11This is a schematic diagram of the result of allocating the spraying unit for the first time according to an embodiment of the present invention;

[0056] Figure 12 2. It is a schematic diagram showing the result of dividing a predetermined area into counting units according to an embodiment of the present invention;

[0057] Figure 13 This is a schematic diagram of the result of the second allocation of the spray unit according to an embodiment of the present invention;

[0058] Figure 14 Schematic diagram of the operation of the integrated spraying module according to an embodiment of the present invention;

[0059] Description of reference numerals:

[0060] 1- Imaging chamber; 2- Black light-absorbing curtain; 3- Color mark sensor; 4- Industrial camera; 5- White LED light; 6- Medicine box; 7- Spraying chamber; 8- Integrated spraying module; 9- Cover plate; 10- Master control box; 11- Independent spraying module; 12- Battery; 13- Relay module; 14- Transformer module; 15- First single-chip microcomputer development board; 16- Second single-chip microcomputer development board; 17- Third single-chip microcomputer development board; 18- Microcomputer; 19- Medicine supply pipeline; 20- First intelligent sprinkler; 21- First mounting bracket; 22- Second mounting bracket; 23- Robotic arm; 24- Bearing; 25- Servo motor; 26- Solenoid valve; 27- Right-angle adapter; 28- Sprayer; 29- Plant tag; 30- Crop plant; 31- Weed. DETAILED DESCRIPTION

[0061] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0062] The purpose of the present invention is to provide a crop positioning and double-precision gridded precision spraying weeding device and method, which can improve the weed identification accuracy and weeding speed.

[0063] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0064] like Figures 1 to 9As shown, the present invention provides a weed identification and precise removal device based on signal conduction technology, which is used in vegetable fields after transplanting. It includes an imaging chamber 1 built of aluminum alloy profiles. The imaging chamber 1 is covered by a black light-absorbing curtain 2. Six color mark sensors 3 are arranged at the bottom end of the interior of the imaging chamber 1. The color mark sensors 3 are connected to a microcomputer 18 through a first single-chip microcomputer development board 15. An industrial camera 4 is fixed to the top of the interior of the imaging chamber 1. The industrial camera 4 is connected to a microcomputer 8. Four A white LED lamp 5 is connected to the battery 12 via a transformer module 14. A medicine box 6 with a liquid pump is arranged on the top of the outside of the imaging room 1. The medicine box 6 is connected to the integrated spraying module 8 and the independent spraying module 11 through a medicine supply pipeline. The spraying room 7 is built of aluminum alloy profiles and is located at the rear of the imaging room 1. An integrated spraying module 8 is arranged in the spraying room 7. The circuit input end of the integrated spraying module 8 is connected to the microcomputer 18 through the relay module 13 and the second single-chip microcomputer development board 16. The signal output end of the integrated spraying module 8 is connected to the microcomputer 18 through the third single-chip microcomputer development board 16; the integrated spraying module 8 is composed of a medicine supply pipeline 19, a first intelligent nozzle 20 and a first mounting frame 21. An aluminum alloy cover plate 9 is installed on the top of the spraying chamber 7. A main control box 10 is arranged on the aluminum alloy cover plate 9. A relay module 13, a transformer module 14, a first single-chip microcomputer development board 15, a second single-chip microcomputer development board 16, a third single-chip microcomputer development board 17 and a microcomputer are installed inside the main control box 10. Computer 18, an independent spraying module 11 is arranged on the cover plate 9, and the circuit input end of the independent spraying module 11 is connected to the microcomputer 18 through the third single-chip microcomputer development board 16; the independent spraying module 11 is composed of a second mounting frame 22, a robotic arm 23, a bearing 24, a servo motor 25 and a second intelligent sprinkler head, and the second intelligent sprinkler head is composed of a solenoid valve 26, a right-angle adapter 27, and an adjustable sprinkler head 28; the plant tag 29 is located near the root of the crop plant 30; the weeds 31 are randomly distributed around the crop plant 30.

[0065] Specifically, the imaging chamber 1 is constructed of aluminum alloy profiles and is covered by a black light-absorbing curtain 2 to protect the imaging chamber from interference from ambient light. Six color mark sensors 3 are arranged at the bottom of the interior of the imaging chamber 1. The color mark sensors 3 are connected to a microcomputer 18 via a single-chip microcomputer development board 15. Four white LED lights 5 are installed inside the imaging chamber 1. The white LED lights 5 are connected to a battery 12 via a transformer module 14.

[0066] Furthermore, the six color mark sensors 3 are evenly distributed on the crossbeam at the bottom of the imaging chamber 1 , with three on each side of the traveling direction. The installation height of the color mark sensors 3 can be flexibly adjusted.

[0067] Furthermore, the scanning light generated by the six color mark sensors 3 is deflected downward at a suitable angle relative to the bottom surface of the imaging chamber 1 to avoid mutual interference between the sensors.

[0068] Furthermore, the focal length of the lens of the industrial camera 4 is determined according to the height of the camera from the ground, so as to clearly and comprehensively capture the crops and the weeds around them.

[0069] Furthermore, the brightness of the white LED lamp 5 is adjusted by the transformer module 14 to ensure that the imaging room has appropriate lighting conditions.

[0070] Furthermore, the liquid medicine pump in the medicine box 6 has an automatic pressure relief function to ensure that the medicine box supplies medicine to each spray module at a constant pressure;

[0071] Furthermore, the integrated spraying module 8 includes 12 first intelligent nozzles 20 arranged in a staggered pattern. When the liquid sprays vertically onto the ground through the first intelligent nozzles 20, it forms a circle with a diameter of 10 mm on the ground. Working in tandem, the first intelligent nozzles 20 can simultaneously spray and weed an area 120 mm wide without overlapping.

[0072] Furthermore, the master control box 10 is the control center of the device of the present invention, and is used to process images and sensor data to control the coordinated operation of various modules.

[0073] Furthermore, the independent spraying module 11 is equipped with a second intelligent nozzle that is the same as that in the integrated spraying module 8. The postures of its two joints are controlled by a servo motor 25, so that the missed spraying unit can be re-sprayed.

[0074] Furthermore, the relay module 13 has 13 relays in total, which correspond to the 13 solenoid valves in the spraying module respectively. By controlling the opening and closing of the solenoid valves, the opening and closing of each nozzle are controlled to achieve selective and precise spraying.

[0075] Furthermore, the rotation angle of the servo motor 25 is controlled by the PWM waveform output by the single chip development board 17, so as to adjust the posture of the independent spraying module 11 according to actual conditions and accurately supplement the spraying of the missed area.

[0076] Furthermore, the present invention segments the plants and background in the image based on the color difference between the plants and the soil. The threshold selected during color segmentation is related to the colors of the plants and the soil. Therefore, before using this device, the color segmentation threshold should be adjusted according to the actual situation to ensure that the green plants can be accurately segmented.

[0077] This device can accurately identify and remove weeds in vegetable fields, reducing herbicide usage, environmental pollution from overuse of herbicides, and the impact of herbicide residues on animal health. The weed identification and control solution proposed by this invention is highly economical and adaptable, advancing the development of smart agriculture.

[0078] Through the above-mentioned scheme of the present invention, vegetable seedlings grow in the fields, and weeds may be distributed around the vegetable seedlings. The device is installed on a chassis that can move autonomously and moves along the crop line, and the crop line is located in the middle of the device. By pre-adjusting the voltage value of the transformer module 14, the brightness of the white LED lamp 5 is adjusted to an appropriate range to maintain stable lighting conditions in the imaging room. By teaching the color mark sensor 3, it can quickly and accurately detect plant tags without generating false positives for weeds, etc. By adjusting the exposure of the industrial camera 4, its imaging is clear. Based on the actual lighting conditions, soil color, etc., by adjusting the color segmentation threshold, all green plants can be segmented out without being mixed with a lot of noise.

[0079] like Figure 10 As shown, this embodiment also provides a method for identifying and precisely removing weeds based on signal transduction technology, including:

[0080] Step 1: Detection using sensors: The entire device is mounted on an autonomous chassis and moves at a constant speed through the field, with the crop row located directly between the imaging chamber 1 and the spray chamber 7. The color mark sensor 3 continuously detects passing objects. If no plant tag 29 is detected, the color mark sensor 3 does not generate any signal. If a plant tag 29 is detected, the color mark sensor 3 outputs a high-level signal and transmits the signal to the microcontroller development board 15.

[0081] Step 2: Capture with an industrial camera: After receiving the signal, the MCU development board 15 transmits the sensor ID of the signal-generating sensor to the microcomputer 18 via serial communication. After receiving the serial data from the MCU development board 15, the microcomputer 18 controls the industrial camera 4 to capture an image of the crop plant 30 and surrounding weeds 31, cropping the image width to a 12 cm width of the ground within the frame. Based on the received data, the coordinates of the plant tag 29 in the image are calculated. When the image size is set to 720×720, the calculation rule for the plant tag is shown in Equation 1.

[0082]

[0083] Where x is the horizontal resolution of the image, y is the vertical resolution of the image, Ser_1 represents the number of the first color sensor that detects the plant tag, Ser_2 represents the number of the second color sensor that detects the plant tag, dt represents the time difference between the two color sensors detecting the crop tag, v represents the forward speed of the device, and point_signal represents the coordinates of the plant tag in the image.

[0084] Step 3: Image Processing: After the industrial camera 4 captures the image, it converts the image's RGB channels into HSV channels to facilitate computer processing. A threshold for each channel is set based on the lighting conditions to segment the image into a binary image of the green plants. The color segmentation rule is shown in Equation 2.

[0085]

[0086] Among them O hsv (x,y) represents the original image captured in HSV channels. hsv (x,y)] h,s,v Represent the hue, saturation and brightness of the original image respectively. h ,T s ,T v It is the threshold defined for each channel, divided into maximum and minimum values. The size of the threshold depends on the actual lighting conditions, soil color, etc. t (x,y) represents the binary image after color segmentation.

[0087] Step 4: Clear the connected domains of the image: Use a suitable structural element to perform an opening operation on the binary image to remove the noise connected domains with small pixel areas in the image, while keeping the size of the connected domains of crops and weeds in the image stable.

[0088] Step 5. Determine the location of pixels and connected domains: Based on the image obtained in step 4, traverse all pixels in the image, find pixels with a value of 1 and the connected domains they form, and record their location information.

[0089] Step 6: Determine the pixel locations of weeds and crops: Based on the image coordinates of the plant tag obtained in Step 2 and the image obtained in Step 5, draw a circular tolerance band with a radius of 7.5 mm in the image, centered at the plant tag's coordinates. Connected domains that intersect this tolerance band are considered to be connected domains of crops. All pixels within these connected domains are labeled as crop pixels, and their locations are recorded. Since the image contains only one crop plant, the remaining connected domains are all connected domains of weeds. All pixels within these connected domains are labeled as weed pixels, and their locations are recorded. The pixel locations of weeds and crops are output as the recognition results.

[0090] Step 7: Determine the spraying unit and the segmentation unit: Based on the image obtained in step 4, use the nozzle resolution of 10mm as the first precision and evenly segment it into 12×12 areas as spraying units. Use the movement precision of the independent spraying module as the second precision and further evenly segment the spraying unit into counting units, such as Figure 12 As shown in the figure, assuming the movement accuracy of an independent nozzle is 5 mm, each spray unit is evenly divided into 2×2 areas as counting units. In each image, there are 144 units to be sprayed and 576 counting units.

[0091] Step 8. Define the spray flag: Define a spray flag for each counting unit. The value of the spray flag represents the spray priority of the counting unit. Create a two-dimensional array to store the values of the spray flag of the counting unit. Initialize the array so that all elements are -1.

[0092] Step 9: Perform the first allocation to the spraying units and generate integrated spraying information: traverse each unit to be sprayed and perform the first allocation to it, such as Figure 11As shown. If the unit to be sprayed contains weeds but no crops, it is a unit that needs to be weeded and will not affect the crops when spraying. Its position information is recorded and the spraying flag bits of the 4 counting units inside are all assigned to 1, indicating that it was assigned as a unit that needs to be sprayed when the spraying unit is first allocated; if the unit to be sprayed contains weeds and crops, it is a unit that needs to be weeded but will affect the crops when spraying. It needs to be divided again with higher precision, its position information is recorded and the spraying flag bits of the 4 counting units inside are kept at -1 unchanged, indicating that it was not assigned when the spraying unit was first allocated; if the unit to be sprayed contains weeds and crops, it is a unit that needs to be weeded but will affect the crops when spraying. It needs to be divided again with higher precision, its position information is recorded and the spraying flag bits of the 4 counting units inside are kept at -1 unchanged, indicating that it was not assigned when the spraying unit was first allocated; If a spraying unit contains neither weeds nor crops, it is a unit that does not require weeding and will not affect crops when sprayed. Its location information is recorded, and the spray flags of the four counting units within it are set to 0, indicating that it was assigned as a unit that does not require spraying when the spraying unit was first allocated. If a spraying unit contains no weeds but does contain crops, it is a unit that does not require weeding but will affect crops when sprayed. It needs to be re-divided with higher precision, its location information is recorded, and the spray flags of the four counting units within it are kept unchanged at -1, indicating that the unit was not assigned when the spraying unit was first allocated. Based on the results of the first spraying unit allocation, integrated spraying information is generated.

[0093] Step 10: Perform a second allocation of the counting unit: Traverse each undefined counting unit and define it. If the counting unit contains weeds but no crops, assign its spray flag bit to 2, indicating that a second allocation of this counting unit is required; if the counting unit contains weeds and crops, assign its spray flag bit to -3, indicating that a second allocation of this counting unit is prohibited; if the counting unit contains no weeds but contains crops, assign its spray flag bit to -2, indicating that a second allocation of this counting unit is prohibited; if the counting unit contains no weeds and no crops, assign its spray flag bit to 0, indicating that a second allocation of this counting unit is not required but allowed.

[0094] Step 11, generate independent spraying information: According to the result obtained in step 10, the counting units that need to be reallocated to the spraying units are allocated for the second time in the order from top to bottom and from left to right, such as Figure 13As shown, a new spraying unit with the same size as the grid divided with the first precision is generated. Assuming that the second precision of the grid division is 5mm, for each counting unit that needs to be reallocated, the spraying unit can be expanded in the four directions of upper left, upper right, lower left, and lower right. For example: when expanding to the upper left, select the three counting units on the left, upper side, and upper left side of the counting unit and combine them into a spraying unit. The same applies when expanding in other directions. When the newly combined spraying unit contains a counting unit with a spraying flag value of -2 or -3, the combination is abandoned. Calculate the sum of the spraying flag values of each counting unit in the remaining combinations, select the combination with the largest sum as the allocation result, and record its position information. Based on the result of the first spraying unit allocation, generate independent spraying information.

[0095] Step 12: Precisely spray the first allocated spray unit: Figure 14 As shown, the integrated spraying information obtained in step 9 is compiled line by line into spraying command codes and sent to the single-chip microcomputer development board 16 via serial communication. After receiving the spraying command codes, the single-chip microcomputer development board 16 calculates the opening and closing times of each intelligent nozzle in the integrated spraying module 8 based on the travel speed of the device. The corresponding solenoid valve 26 is controlled by the relay module 14 to control the opening and closing of the nozzle, thereby accurately spraying the first assigned spray unit.

[0096] Step 13: Precisely spray the second assigned spray unit: Based on the independent spray information obtained in step 11, it is compiled line by line into spray command code and sent to the microcontroller development board 17 via serial communication. The microcontroller development board 17 is pre-programmed to calculate the position and spraying time of the second assigned spray unit based on the received spray command code. Based on the calculated position, a specific PWM waveform is output to the servo motor 25 to control the nozzle of the independent spray module to point to the spray unit. The relay module 14 then controls the opening and closing of the intelligent nozzle, precisely spraying the second assigned spray unit.

[0097] The beneficial effects of the present invention are as follows:

[0098] (1) The method of marking crop plants with plant tags in the present invention can simplify the crop and weed classification algorithm, improve the weed identification accuracy and the operating speed of the device.

[0099] (2) The arrangement of the color mark sensor in the present invention can effectively deal with the situation where the plant tag is blocked by weeds, thereby improving the stability of the device.

[0100] (3) The present invention combines signal conduction technology and computer vision technology for weed identification, which not only achieves rapid and accurate weed identification but also reduces the cost of the device.

[0101] (4) The precise spraying scheme of the present invention has a very small coverage area of each nozzle, and the spraying is more precise, which saves pesticides while reducing pollution to the environment.

[0102] (5) The spraying device of the present invention is divided into spraying units with double precision, which can achieve a higher weed control rate under the same nozzle precision without affecting the speed of the device.

[0103] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0104] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A crop positioning and double-precision gridding precision spraying and weeding device, characterized in that: Used in vegetable fields after transplanting, the roots of the crop plants in the vegetable fields are provided with plant tags, the device is arranged on a movable chassis, and the device includes: an imaging chamber, a color code sensor, an industrial camera, a medicine box, a spraying chamber, an integrated spraying module, a cover plate, a main control box, an independent spraying module, a single-chip microcomputer development module and a microcomputer; Six color mark sensors are arranged at the bottom of the imaging chamber. The color mark sensors are evenly distributed on the crossbeam at the bottom of the imaging chamber, with three arranged on each side of the direction of travel. The color mark sensors are connected to the microcomputer via the single-chip development module. The color mark sensors are used to detect the plant tags and generate detection signals. The industrial camera is set at the top of the imaging chamber. The industrial camera is connected to the microcomputer. The integrated spraying module is arranged in the spraying room; the top of the spraying room is provided with the cover plate; the main control box and the independent spraying module are arranged on the cover plate; The integrated spraying module has 12 first intelligent nozzles arranged in a staggered manner. When the liquid medicine passes through the first intelligent nozzles and is sprayed vertically toward the ground, the sprayed liquid forms a circle with a diameter of 10 mm on the ground; The independent spraying module is equipped with a second intelligent nozzle, and the spraying angle of the second intelligent nozzle can be controlled by a servo motor; The single chip microcomputer development module is used to receive the detection signal and mark the detection signal according to the number of the corresponding color mark sensor to obtain a marking signal; The microcomputer is used to control the industrial camera to acquire images of the crop plant and the weeds around it according to the marking signal, perform image processing on the captured images to obtain processed images, and then identify the processed images to obtain pixel position information of the weeds and the crop plant; wherein, after acquiring the captured image, the coordinates of the plant tag in the captured image are calculated, and the calculation of the coordinates of the plant tag in the captured image is based on the time difference between the first color mark sensor that detects the plant tag and the second color mark sensor that detects the plant tag, the forward speed of the device, and the horizontal resolution and vertical resolution of the captured image; The microcomputer is further configured to, based on the processed image and the pixel position information of the weeds and the crop plants, evenly divide the processed image into a plurality of regions with a resolution of the integrated spraying module as a first precision to form a grid of spraying units; further evenly divide each of the spraying units into a plurality of counting units with a movement precision of the independent spraying modules as a second precision to form a grid of counting units; define a spraying flag for each of the counting units, wherein the value of the spraying flag represents the spraying priority of the counting unit; perform a first spraying allocation based on the spraying flag to generate integrated spraying information; and perform a second spraying allocation based on the spraying flag to generate independent spraying information; The microcomputer is also used to control the integrated spraying module and the independent spraying module respectively according to the integrated spraying information and the independent spraying information. The integrated spraying module is controlled to calculate the opening and closing time of each first intelligent nozzle in the integrated spraying module by combining the travel speed of the device, so as to control the opening and closing of the first intelligent nozzle, and accurately spray the spraying unit allocated for the first spraying; the independent spraying module is controlled to calculate the position and spraying time information of the spraying unit allocated for the second spraying according to the independent spraying information, and control the second intelligent nozzle carried by the independent spraying module to point to the spraying unit, and accurately spray the spraying unit allocated for the second spraying, so as to achieve re-spraying of the missed spraying unit.

2. The crop positioning and double-precision gridding precision spraying and weeding device according to claim 1 is characterized in that: The medicine box is provided at the external top of the imaging room; the medicine box is connected to the integrated spray module and the independent spray module respectively; a spray room is also provided on one side of the imaging room; the integrated spray module and the independent spray module are both connected to the single-chip development module; the interior of the main control box is provided with the interconnected single-chip development module and the microcomputer.

3. The crop positioning and double-precision gridding precision spraying and weeding device according to claim 2 is characterized in that: Also includes: Black light-absorbing curtain, transformer module, battery and multiple white LED lights; The black light-absorbing curtain covers the four sides of the imaging room; the white LED lamp is arranged inside the imaging room; the white LED lamp is connected to the battery through the transformer module; the transformer module is arranged in the main control box; the transformer module is used to adjust the brightness of the white LED lamp.

4. The crop positioning and double-precision gridding precision spraying and weeding device according to claim 2 is characterized in that: It also includes a relay module; the relay module is arranged in the master control box; The single-chip microcomputer development module includes a first single-chip microcomputer development board, a second single-chip microcomputer development board and a third single-chip microcomputer development board; the first single-chip microcomputer development board, the second single-chip microcomputer development board and the third single-chip microcomputer development board are all connected to the microcomputer; the first single-chip microcomputer development board is connected to the color mark sensor; the second single-chip microcomputer development board is connected to the input end of the integrated spraying module through the relay module; the signal output end of the integrated spraying module and the signal output end of the independent spraying module are both connected to the third single-chip microcomputer development board.

5. The crop positioning and double-precision gridding precision spraying and weeding device according to claim 4 is characterized in that: The integrated spraying module further includes a medicine supply pipe and a first mounting frame; one end of the medicine supply pipe is connected to the medicine box; the other end of the medicine supply pipe is connected to each of the first intelligent nozzles; the first mounting frame is arranged in the spraying room; The first intelligent nozzle is arranged on the first mounting frame.

6. The crop positioning and double-precision gridding precision spraying and weeding device according to claim 5, characterized in that: The independent spraying module also includes a second mounting frame, a robotic arm, a bearing, and a servo motor; The second mounting frame is connected to one end of the robotic arm through the bearing, and the servo motor is provided on the robotic arm; the other end of the robotic arm is connected to the second intelligent nozzle; the servo motor is connected to the third single-chip microcomputer development board; the servo motor is used to control the spray angle of the second intelligent nozzle.

7. The crop positioning and double-precision gridding precision spraying and weeding device according to claim 6, characterized in that: Either the first smart nozzle or the second smart nozzle comprises a solenoid valve, a right-angle adapter, and an adjustable nozzle; The solenoid valve is connected to the adjustable nozzle through the right-angle adapter; the solenoid valve is connected to the relay module; the relay module is used to control the opening and closing state of the solenoid valve to achieve the opening and closing state of the adjustable nozzle.

8. The crop positioning and double-precision gridding precision spraying and weeding device according to claim 6, characterized in that: A liquid medicine pump is provided in the medicine box; the liquid medicine pump is used to ensure that the medicine box supplies medicine to the integrated spraying module and the independent spraying module at a constant pressure.

Images