CN116250523A - Intelligent laser weeding device and weeding method based on machine vision - Google Patents
Intelligent laser weeding device and weeding method based on machine vision Download PDFInfo
- Publication number
- CN116250523A CN116250523A CN202310349843.1A CN202310349843A CN116250523A CN 116250523 A CN116250523 A CN 116250523A CN 202310349843 A CN202310349843 A CN 202310349843A CN 116250523 A CN116250523 A CN 116250523A
- Authority
- CN
- China
- Prior art keywords
- laser
- optical path
- weed
- weeding
- fixing block
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000009333 weeding Methods 0.000 title claims abstract description 54
- 238000000034 method Methods 0.000 title claims abstract description 14
- 241000196324 Embryophyta Species 0.000 claims abstract description 64
- 230000003287 optical effect Effects 0.000 claims description 55
- 230000001502 supplementing effect Effects 0.000 claims description 20
- 239000011521 glass Substances 0.000 claims description 18
- 239000013307 optical fiber Substances 0.000 claims description 8
- 238000002834 transmittance Methods 0.000 claims description 7
- 239000000835 fiber Substances 0.000 claims description 6
- 238000004140 cleaning Methods 0.000 claims description 4
- 239000000428 dust Substances 0.000 claims description 4
- 229910000838 Al alloy Inorganic materials 0.000 claims description 3
- 239000000956 alloy Substances 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 239000003086 colorant Substances 0.000 claims description 2
- 238000013473 artificial intelligence Methods 0.000 claims 2
- 239000000126 substance Substances 0.000 abstract description 7
- 238000013135 deep learning Methods 0.000 abstract description 5
- 239000004009 herbicide Substances 0.000 abstract description 3
- 238000011161 development Methods 0.000 abstract description 2
- 230000002363 herbicidal effect Effects 0.000 abstract description 2
- 238000005259 measurement Methods 0.000 abstract 1
- 230000006378 damage Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 230000003020 moisturizing effect Effects 0.000 description 3
- 239000002689 soil Substances 0.000 description 3
- 241000607479 Yersinia pestis Species 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 208000012260 Accidental injury Diseases 0.000 description 1
- 244000025254 Cannabis sativa Species 0.000 description 1
- 206010059866 Drug resistance Diseases 0.000 description 1
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 244000052616 bacterial pathogen Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 244000038559 crop plants Species 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 238000009342 intercropping Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000035699 permeability Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 238000009423 ventilation Methods 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01M—CATCHING, TRAPPING OR SCARING OF ANIMALS; APPARATUS FOR THE DESTRUCTION OF NOXIOUS ANIMALS OR NOXIOUS PLANTS
- A01M21/00—Apparatus for the destruction of unwanted vegetation, e.g. weeds
- A01M21/04—Apparatus for destruction by steam, chemicals, burning, or electricity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P60/00—Technologies relating to agriculture, livestock or agroalimentary industries
- Y02P60/12—Technologies relating to agriculture, livestock or agroalimentary industries using renewable energies, e.g. solar water pumping
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Environmental Sciences (AREA)
- Zoology (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Wood Science & Technology (AREA)
- Pest Control & Pesticides (AREA)
- Insects & Arthropods (AREA)
- Multimedia (AREA)
- Soil Working Implements (AREA)
- Catching Or Destruction (AREA)
Abstract
The invention discloses an intelligent laser weeding device and a weeding method based on machine vision, wherein the weeding device comprises a supporting auxiliary unit, a weed identification unit and a weed removal unit; according to the invention, through the organic combination of the deep learning, binocular stereoscopic vision technology and laser technology, the laser weeding equipment based on the deep learning and binocular stereoscopic vision is formed, the accurate positioning of the weed center point and the length and width measurement of the weed boundary frame are completed by using a high-performance camera and an embedded artificial intelligent super computing platform, and the laser vibrating mirror is adopted to scan and clear the weed area. The method can realize accurate weeding operation in the seedling stage of crop planting, simultaneously reduce the influence of weeding operation on crop seedlings as much as possible, improve the planting efficiency and effectively increase the yield; meanwhile, the use amount of chemical herbicide is greatly reduced, and the development of green agriculture is effectively promoted.
Description
Technical Field
The invention relates to the field of farmland weeding, in particular to the field of laser weeding devices and methods.
Background
In the emergence period of crops, as the crop plants are smaller at the moment and do not form a compact canopy, weeds have stronger ecological competitiveness, so that the growth and development of the crops are hindered, the plants are thin and weak, and the fruiting rate is reduced; meanwhile, field weeds can be used as hosts for field pathogenic bacteria and pests, so that the difficulty in pest control is increased, and the crop yield is seriously affected. In the general year of grass damage, 10% -20% of yield reduction can be caused, and in severe cases, even five products can be reduced.
Currently, the main weeding means in the world are artificial weeding, chemical weeding, mechanical weeding, biotechnological weeding, physical factor weeding, thermoelectric weeding and the like, and the current weeding means with more use are artificial weeding, mechanical weeding and chemical weeding.
The artificial weeding effect is good, the damage to farmland target crops is small, but a large amount of labor force is needed. The time limit of chemical weeding is less, and meanwhile, the chemical weeding composition has a good weeding effect on inter-seedling weeds, but the screening effect of the chemical herbicide can lead the weeds to generate drug resistance; the abuse of chemical herbicides can cause serious pollution and serious damage to the ecological environment.
The mechanical weeding machine has the advantages of low labor intensity, high weeding efficiency, environmental friendliness, soil loosening while weeding, soil permeability enhancement and soil structure improvement. However, the traditional mechanical weeding equipment still has more problems: on one hand, the weeding time and the crop planting mode are greatly limited, the planting interval of crops is greatly required, the inter-seedling weed is difficult to remove, and the method is not suitable for the inter-cropping, interplanting or close planting conditions; on the other hand, in the fragile seedling period of root system, traditional weeding machinery can cause seedling accidental injury, reduces seedling survival rate, improves manufacturing cost and operating time.
Disclosure of Invention
The invention mainly aims to overcome the defects of the weeding method and reduce weed harm, and provides an intelligent laser weeding device based on machine vision, so as to realize accurate and rapid weeding in fields.
The invention further aims at providing an intelligent laser weeding method based on machine vision.
In order to achieve the main purpose, the invention adopts the following technical scheme:
an intelligent laser weeding device based on machine vision comprises a supporting auxiliary unit, a weed identification unit and a weed removal unit.
Further, support auxiliary unit includes running gear, middle crossbeam, roof shield, shading dustproof bounding wall, LED moisturizing tape fixed bolster, high glass, battery and laser instrument fixed bolster, battery and high glass stationary blade that passes through, running gear is four-wheel independent electric drive chassis, according to crop row spacing free adjustment wheel base size, the middle crossbeam passes through the bolt fastening and is fixed in walking dolly chassis lower surface, battery and laser instrument fixed bolster are fixed in dolly chassis upper surface, roof shield cover is around walking dolly chassis upper surface, shading dustproof bounding wall is the aluminum alloy material, installs around walking dolly chassis lower surface, and opens at optic fibre laser instrument one side and have decurrent vent, LED moisturizing tape fixed bolster welds in shading dustproof bounding wall lower extreme, and LED moisturizing tape fixed bolster lower surface is 1 centimetre apart from shading dustproof bounding wall lower extreme, and high glass thickness 0.5 centimetre is installed in shading dustproof bounding wall's below through high glass stationary blade, and shading dustproof bounding wall lower extreme, high glass, roof shield, dustproof bounding wall constitute jointly and clear up half battery, and clear away the battery and comparatively provide the parallel and level with comparatively fixed laser instrument of the comparatively in the roof shield for walking dolly.
The weed recognition unit comprises an LED light supplementing belt, an embedded artificial intelligent super computing platform, a binocular camera, a camera fixing screw, a mounting bracket clamping bolt and a binocular camera mounting bracket, wherein the LED light supplementing belt is placed towards the ground, the thickness of the LED light supplementing belt is 1 cm, the LED light supplementing belt is stuck to the mounting bracket of the LED light supplementing belt, the lower surface of the light supplementing belt is just contacted with the upper surface of high-transmittance glass, the reflection of light is reduced, the embedded artificial intelligent super computing platform is fixed on a shading dustproof coaming through the screw, the binocular camera lens is placed towards the ground, the camera fixing screw is fixed on the binocular camera mounting bracket, the binocular camera mounting bracket is mounted on a middle cross beam, the mounting bracket clamping bolt at the top of the binocular camera mounting bracket is adjusted, and the position of the binocular camera mounting bracket on the middle cross beam can be controlled.
The weed removing unit comprises an optical fiber laser, a vibrating mirror, a first optical path fixing block, a second optical path fixing block, an optical path fixing bolt, an optical path fixing block clamping bolt, a vibrating mirror control card and a laser optical path connector, wherein the optical fiber laser is arranged on a battery and a laser fixing support, a laser beam is transmitted into the vibrating mirror through the optical fiber and the laser optical path connector, the laser optical path connector is connected with the first optical path fixing block and the second optical path fixing block through the optical path fixing bolt, the first optical path fixing block and the second optical path fixing block are clamped on a middle beam, the positions of the first optical path fixing block and the second optical path fixing block on the middle beam can be controlled by adjusting the optical path fixing block clamping bolt on the top of the first optical path fixing block and the second optical path fixing block, the vibrating mirror is fixed with the laser optical path connector and placed towards the ground, and the vibrating mirror control card is arranged on the side surface of the first optical path fixing block and fixed through screws.
A laser weeding device based on deep learning and binocular stereoscopic vision has a laser power of 100 watts and a minimum ground clearance of 400 mm.
In order to achieve the other purpose, the invention adopts the following technical scheme:
the invention discloses an intelligent laser weeding method based on machine vision, which comprises the following steps:
(1) The walking unit runs at a constant speed, the binocular camera collects field image data, and the data is uploaded to the embedded artificial intelligent super computing platform through the USB interface;
(2) According to a pre-trained weed identification model, the embedded artificial intelligent super computing platform carries out target identification on each frame of collected images, and after weeds are identified, two-dimensional coordinates (X, Y) of weed center points under an image coordinate system are sequentially recorded;
(3) The embedded artificial intelligent super computing platform converts the image acquired by the binocular camera into a gray level image and carries out histogram equalization, further eliminates image distortion, carries out three-dimensional correction to obtain a parallax image, and obtains a three-dimensional coordinate (X) of a target weed center point under a camera coordinate system according to the parallax image and two-dimensional coordinates (X, Y) 0 ,Y 0 ,Z 0 ) Determining a square area where weeds are located, dividing the square area by adopting weeds based on colors, generating boundary coordinates of the weeds, and determining the area where the weeds are located;
(4) The three-dimensional coordinates of the weed center point and the weed boundary coordinates are sent to a laser galvanometer control card through a serial port, the laser galvanometer control card adjusts the height of the integral galvanometer according to the Z-axis coordinate Z0 of the target weed center point, adjusts the laser focus, and controls the galvanometer deflection, so that the laser beam sweeps through the area where the weed is located in a linear interpolation path, and laser cleaning of the weed is realized.
The invention has the beneficial effects that:
1. the invention realizes the organic combination of deep learning and binocular vision, integrates an image processing operation module into an embedded artificial intelligent super computing platform, realizes the accurate positioning of coordinates of a weed center point and a weed boundary on the basis of identifying field weeds, and realizes the intellectualization of the weeding flow of discovery, positioning and cleaning.
2. The invention adopts laser under the control of the vibrating mirror as a weeding means, the self-adaptive scanning of the laser ensures the coverage of the range of weeds, the laser weeding has the advantages of high weeding efficiency and short time consumption, and the laser weeding has no potential threat to the environment and farmland crops, thereby improving the weeding safety.
3. Besides the cost of purchasing equipment in the earlier stage, the laser weeding equipment only needs less maintenance cost in the follow-up process, does not need a large amount of consumables, and is simple and controllable in cost.
Drawings
Fig. 1 is a general schematic of the present invention.
Fig. 2 and 3 are schematic views of the support auxiliary unit of the present invention.
FIG. 4 is a schematic view of a weed identifying unit of the present invention.
Fig. 5 and 6 are schematic diagrams of weed removal units of the present invention.
Fig. 7 is a control flow chart of the present invention.
Detailed Description
The invention will be described in further detail with reference to the drawings and examples.
As shown in fig. 1. The intelligent laser weeding device and method based on machine vision are characterized by comprising a supporting auxiliary unit (1), a weed identification unit (2) and a weed removal unit (3).
As shown in fig. 2 and 3. The supporting auxiliary unit (1) comprises a traveling device (1-1), a middle cross beam (1-2), a roof dust cover (1-3), a shading dust-proof coaming (1-4), an LED light supplementing strip fixing support (1-5), high-transmittance glass (1-6), a battery and laser fixing support (1-7), a battery (1-8) and a high-transmittance glass fixing sheet (1-9), wherein the traveling device is a four-wheel independent electric drive chassis, the wheel distance is freely adjusted according to the row distance of crops, the middle cross beam is fixed on the lower surface of the traveling trolley chassis through bolts and is used for installing a weed identification unit and a weed removal unit, the battery and the laser fixing support are fixed on the upper surface of the trolley chassis, the battery and the laser fixing support are used as a mounting base of the laser and the battery, the roof dust cover covers the periphery of the upper surface of the traveling trolley chassis, the shading dust-proof coaming is made of aluminum alloy material and is mounted on the periphery of the lower surface of the traveling trolley chassis, a downward ventilation opening is formed in one side of the optical fiber laser, the LED light supplementing fixing support is welded on the lower end of the light shading coaming, the LED light supplementing strip fixing support is used for installing the weed identification unit and the weed removal unit, the highest-proof coaming is arranged at the highest thickness of the highest, the highest-transmittance glass is 0 cm, the highest-transmittance glass is higher than the best, the best side of the light shielding unit and the best side is clear and the best side is formed by the best side and the best side, and high-transmittance glass is clear and high-quality, and the best clear, and the side can pass through the best side and the best clear side can be clear and clear. The battery is arranged on the battery and the laser fixing bracket in the dustproof cover of the vehicle roof.
As shown in fig. 4. The weed identification unit (2) comprises an LED light supplementing belt (2-1), an embedded artificial intelligent super computing platform (2-2), a binocular camera (2-3), a camera fixing screw (2-4), a mounting bracket clamping bolt (2-5) and a binocular camera mounting bracket (2-6), wherein the LED light supplementing belt is placed facing the ground, the thickness of the LED light supplementing belt is 1 cm, the LED light supplementing belt is stuck on the LED light supplementing belt fixing bracket, the lower surface of the light supplementing belt is just contacted with the upper surface of high-transparency glass, the reflection of light is reduced, the embedded artificial intelligent super computing platform is fixed on a shading dustproof coaming through screws, a binocular camera lens is placed facing the ground and is fixed on the binocular camera mounting bracket through the camera fixing screw, the binocular camera mounting bracket is mounted on a middle beam, and the mounting bracket clamping bolt at the top of the binocular camera mounting bracket is adjusted, so that the position of the binocular camera mounting bracket on the middle beam can be controlled.
As shown in fig. 5 and 6. The weed removing unit (3) comprises an optical fiber laser (3-1), a vibrating mirror (3-2), a first optical path fixing block (3-3), a second optical path fixing block (3-4), an optical path fixing bolt (3-5), an optical path fixing block clamping bolt (3-6), a vibrating mirror control card (3-7) and a laser optical path connector (3-8), wherein the optical fiber laser is arranged on a battery and a laser fixing bracket, a laser beam is transmitted into the vibrating mirror through the optical fiber and the laser optical path connector, the laser optical path connector is connected with the first optical path fixing block and the second optical path fixing block through the optical path fixing bolt, the first optical path fixing block and the second optical path fixing block are clamped on a middle beam pair, the positions of the first optical path fixing block and the second optical path fixing block on the middle beam can be controlled by the optical path fixing block clamping bolts, the vibrating mirror is fixed with the laser optical path connector and placed facing the ground, the vibrating mirror control card is arranged on the side surface of the first optical path fixing block, the vibrating mirror can control a rectangular area where weeds are located through fixing screws, and the vibrating mirror can control the rectangular area where the weeds to be irradiated, and heat the weeds.
The embodiment is based on the laser weeding device of degree of depth study and binocular stereovision, and is characterized in that the laser power of a fiber laser is 100 watts, and the minimum ground clearance of a vibrating mirror is 400 mm.
The laser weeding method based on deep learning and binocular stereo vision in the embodiment is shown in fig. 7, and is characterized by comprising the following steps:
(1) The walking unit runs at a constant speed, the binocular camera collects field image data, and the data is uploaded to the embedded artificial intelligent super computing platform through the USB interface;
(2) According to a pre-trained weed identification model, the embedded artificial intelligent super computing platform carries out target identification on each frame of collected images, and after weeds are identified, two-dimensional coordinates (X, Y) of weed center points under an image coordinate system are sequentially recorded;
(3) The embedded artificial intelligent super computing platform converts an image acquired by a binocular camera into a gray level image and carries out histogram equalization, image distortion is further eliminated, three-dimensional correction is carried out, a parallax image is obtained, three-dimensional coordinates (X0, Y0 and Z0) of a target weed center point under a camera coordinate system are obtained according to the parallax image and the two-dimensional coordinates (X, Y), a square area where weeds are located is determined, color-based weed segmentation is adopted in the square area, weed boundary coordinates are generated, and the area where the weeds are located is determined;
(4) The three-dimensional coordinates of the weed center point and the weed boundary coordinates are sent to a laser galvanometer control card through a serial port, the laser galvanometer control card adjusts the height of the integral galvanometer according to the Z-axis coordinate Z0 of the target weed center point, adjusts the laser focus, and controls the galvanometer deflection, so that the laser beam sweeps through the area where the weed is located in a linear interpolation path, and laser cleaning of the weed is realized.
The above description is only of the preferred embodiments of the present invention, and is not intended to limit the invention in any way, and any person skilled in the art may make many variations or modifications with the technical content disclosed in the above description without departing from the scope of the invention, as equivalent embodiments. However, any simple modification, equivalent replacement, improvement, etc. of the above embodiments still fall within the scope of the technical solution of the present invention, according to the technical spirit of the present invention, without departing from the technical solution of the present invention.
Claims (6)
1. The intelligent laser weeding device and weeding method based on machine vision are characterized by comprising a supporting auxiliary unit (1), a weed identification unit (2) and a weed removal unit (3).
2. The support auxiliary unit (1) according to claim 1, comprising a running gear (1-1), a middle cross beam (1-2), a roof dust cover (1-3), a shading dust-proof coaming (1-4), an LED light supplementing belt fixing bracket (1-5), high-transmission glass (1-6), a battery and laser fixing bracket (1-7), a battery (1-8) and a high-transmission glass fixing plate (1-9), wherein the running gear is a four-wheel independent electric drive chassis, the wheel tread size is freely adjusted according to the row spacing of crops, the middle cross beam is fixed on the lower surface of the running trolley chassis through bolts, the battery and the laser fixing bracket are fixed on the upper surface of the trolley chassis, the roof dust cover covers the periphery of the upper surface of the running trolley chassis, the light-shielding dustproof coaming is made of aluminum alloy material, is arranged around the lower surface of the chassis of the travelling trolley, and is provided with a downward vent opening at one side of the optical fiber laser, the LED light-supplementing strip fixing support is welded at the lower end of the light-shielding dustproof coaming, the lower surface of the LED light-supplementing strip fixing support is 1 cm away from the lowest end of the light-shielding dustproof coaming, the thickness of high-permeability glass is 0.5 cm, the LED light-supplementing strip fixing support is arranged at the lowest end of the light-shielding dustproof coaming through the high-permeability glass fixing sheet, the upper surface is flush with the lowest end of the light-shielding dustproof coaming, the high-permeability glass, the roof dustproof cover and the light-shielding dustproof coaming jointly form a semi-closed space, so as to provide a better running environment for the weed identification unit and the weed removal unit, the battery is arranged on the battery and the laser fixing bracket in the dustproof cover of the vehicle roof.
3. The weed recognition unit (2) according to claim 1, comprising an LED light supplementing strip (2-1), an embedded artificial intelligence super computing platform (2-2), a binocular camera (2-3), camera fixing screws (2-4), a mounting bracket clamping bolt (2-5) and a binocular camera mounting bracket (2-6), wherein the LED light supplementing strip is placed facing the ground and has a thickness of 1 cm, is stuck on the LED light supplementing strip fixing bracket, the lower surface of the light supplementing strip just contacts with the upper surface of the high-transmittance glass, the reflection of light is reduced, the embedded artificial intelligence super computing platform is fixed on a shading and dustproof coaming through screws, the binocular camera lens is placed facing the ground and is fixed on the binocular camera mounting bracket through the camera fixing screws, the binocular camera mounting bracket is mounted on the middle beam, and the mounting bracket clamping bolt at the top of the binocular camera mounting bracket is adjusted, so that the position of the binocular camera mounting bracket on the middle beam can be controlled.
4. Weed removal unit (3) according to claim 1, comprising a fiber laser (3-1), a vibrating mirror (3-2), a first optical path fixing block (3-3), a second optical path fixing block (3-4), an optical path fixing bolt (3-5), an optical path fixing block clamping bolt (3-6), a vibrating mirror control card (3-7) and a laser optical path connector (3-8), wherein the fiber laser is mounted on a battery and a laser fixing bracket, a laser beam is transmitted into the vibrating mirror through the fiber and the laser optical path connector, the laser optical path connector is connected together with the first optical path fixing block and the second optical path fixing block through the optical path fixing bolt, the first optical path fixing block and the second optical path fixing block are clamped on a middle beam pair, the positions of the first optical path fixing block and the second optical path fixing block on the middle beam can be controlled by adjusting the optical path fixing block clamping bolts at the tops of the first optical path fixing block and the second optical path fixing block, the vibrating mirror is fixed with the laser optical path connector and placed facing the ground, and the vibrating mirror control card is mounted on the side of the first optical path fixing block through a screw fixing bolt.
5. The intelligent laser weeding device based on machine vision according to claim 1, wherein the laser power of the fiber laser is 100 watts, and the minimum ground clearance of the vibrating mirror is 400 mm.
6. The intelligent laser weeding method based on machine vision according to claim 1, comprising the following steps:
1) The support auxiliary unit runs at a constant speed, the binocular camera collects field image data, and the image data is uploaded to the embedded artificial intelligent super computing platform through the USB interface;
2) According to a pre-trained weed identification model, the embedded artificial intelligent super computing platform carries out target identification on each frame of collected images, and after weeds are identified, two-dimensional coordinates (X, Y) of weed center points under an image coordinate system are sequentially recorded;
3) The embedded artificial intelligent super computing platform converts the image acquired by the binocular camera into a gray level image and carries out histogram equalization, further eliminates image distortion, carries out three-dimensional correction to obtain a parallax image, and obtains a three-dimensional coordinate (X) of a target weed center point under a camera coordinate system according to the parallax image and two-dimensional coordinates (X, Y) 0 ,Y 0 ,Z 0 ) Determining a square area where weeds are located, dividing the square area by adopting weeds based on colors, generating boundary coordinates of the weeds, and determining the area where the weeds are located;
4) The three-dimensional coordinates of the weed center point and the weed boundary coordinates are sent to a laser galvanometer control card through a serial port, and the laser galvanometer control card is used for controlling Z according to the Z-axis coordinates of the target weed center point 0 The height of the integral vibrating mirror is adjusted, the laser focus is adjusted, and the laser vibrating mirror control card controls the vibrating mirror to deflect, so that the laser beam sweeps through the area where weeds are located in a linear interpolation path, and laser cleaning of the weeds is realized.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310349843.1A CN116250523B (en) | 2023-04-03 | 2023-04-03 | Intelligent laser weeding device and weeding method based on machine vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310349843.1A CN116250523B (en) | 2023-04-03 | 2023-04-03 | Intelligent laser weeding device and weeding method based on machine vision |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116250523A true CN116250523A (en) | 2023-06-13 |
CN116250523B CN116250523B (en) | 2024-06-14 |
Family
ID=86682657
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310349843.1A Active CN116250523B (en) | 2023-04-03 | 2023-04-03 | Intelligent laser weeding device and weeding method based on machine vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116250523B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117882697A (en) * | 2024-03-14 | 2024-04-16 | 东海实验室 | Accurate and rapid intelligent laser weeding device and method |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101589705A (en) * | 2009-06-26 | 2009-12-02 | 江苏大学 | A kind of laser weeding robot |
CN109511356A (en) * | 2019-01-08 | 2019-03-26 | 安徽农业大学 | A kind of intelligent weed-eradicating robot system and control method based on deep vision |
CN109937999A (en) * | 2019-03-21 | 2019-06-28 | 中国科学院合肥物质科学研究院 | A kind of laser kills the device of mother drug plants |
CN110421353A (en) * | 2019-09-11 | 2019-11-08 | 哈尔滨理工大学 | Auxiliary complex milling machine tool and monitoring method are heated based on machine vision laser |
CN113632779A (en) * | 2021-09-16 | 2021-11-12 | 山东农业大学 | Laser intelligent weeding robot and weeding method |
CN215123657U (en) * | 2021-01-04 | 2021-12-14 | 东北农业大学 | Vibration machinery weeding device based on machine vision |
CN114453729A (en) * | 2022-01-12 | 2022-05-10 | 哈尔滨理工大学 | Laser cooperation device for multi-axis turning and milling composite machine tool and control method |
WO2022101759A1 (en) * | 2020-11-12 | 2022-05-19 | Earth Rover Limited | Light concentration weeding system |
CN114911126A (en) * | 2022-07-15 | 2022-08-16 | 北京科技大学 | Laser three-dimensional projection device based on binocular vision and galvanometer scanning |
WO2022233307A1 (en) * | 2021-05-07 | 2022-11-10 | 天津理工大学 | Weeding robot based on crop stalk positioning, and weeding method |
CN218681465U (en) * | 2022-08-29 | 2023-03-24 | 北京大学现代农业研究院 | Weeding robot |
-
2023
- 2023-04-03 CN CN202310349843.1A patent/CN116250523B/en active Active
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101589705A (en) * | 2009-06-26 | 2009-12-02 | 江苏大学 | A kind of laser weeding robot |
CN109511356A (en) * | 2019-01-08 | 2019-03-26 | 安徽农业大学 | A kind of intelligent weed-eradicating robot system and control method based on deep vision |
CN109937999A (en) * | 2019-03-21 | 2019-06-28 | 中国科学院合肥物质科学研究院 | A kind of laser kills the device of mother drug plants |
CN110421353A (en) * | 2019-09-11 | 2019-11-08 | 哈尔滨理工大学 | Auxiliary complex milling machine tool and monitoring method are heated based on machine vision laser |
WO2022101759A1 (en) * | 2020-11-12 | 2022-05-19 | Earth Rover Limited | Light concentration weeding system |
CN215123657U (en) * | 2021-01-04 | 2021-12-14 | 东北农业大学 | Vibration machinery weeding device based on machine vision |
WO2022233307A1 (en) * | 2021-05-07 | 2022-11-10 | 天津理工大学 | Weeding robot based on crop stalk positioning, and weeding method |
CN113632779A (en) * | 2021-09-16 | 2021-11-12 | 山东农业大学 | Laser intelligent weeding robot and weeding method |
CN114453729A (en) * | 2022-01-12 | 2022-05-10 | 哈尔滨理工大学 | Laser cooperation device for multi-axis turning and milling composite machine tool and control method |
CN114911126A (en) * | 2022-07-15 | 2022-08-16 | 北京科技大学 | Laser three-dimensional projection device based on binocular vision and galvanometer scanning |
CN218681465U (en) * | 2022-08-29 | 2023-03-24 | 北京大学现代农业研究院 | Weeding robot |
Non-Patent Citations (1)
Title |
---|
潘雷;陈树人;栗移新;毛罕平;: "CO_2激光除草应用初步研究", 农机化研究, no. 06, 1 June 2008 (2008-06-01), pages 171 - 173 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117882697A (en) * | 2024-03-14 | 2024-04-16 | 东海实验室 | Accurate and rapid intelligent laser weeding device and method |
Also Published As
Publication number | Publication date |
---|---|
CN116250523B (en) | 2024-06-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP2017535279A (en) | Automatic mobile robot | |
CN108225798B (en) | Tire slippage testing system and method based on edge enhancement identification | |
CN107860366B (en) | Mobile greenhouse crop information measurement and diagnosis system | |
CN108802759A (en) | The nearly sensing system of movable type towards plant phenotype and data capture method | |
CN111460903B (en) | System and method for monitoring growth of field broccoli based on deep learning | |
CN116250523B (en) | Intelligent laser weeding device and weeding method based on machine vision | |
CN107436340A (en) | A kind of plant root cap Integrated Monitoring System and method | |
AU2013340393B2 (en) | Provision of data on the froth in a froth flotation plant | |
CN112610828A (en) | Wheat image acquisition equipment in field | |
CN115280964B (en) | Automatic operation driving method and system of stem and leaf vegetable harvester and harvester | |
CN114431004B (en) | Banana picking device and method based on 3-PRS parallel mechanism | |
EP4044114A1 (en) | Systems and methods for determining residue length within a field | |
CN113971759B (en) | Automatic identifying device and method for soybean inter-plant weed based on computer vision | |
CN114419407A (en) | Inline weed automatic identification method and device for seedling stage of transplanted crops | |
CN112337810B (en) | Vision guiding pearl sorting robot and sorting method thereof | |
CN211423761U (en) | Camera device for agricultural machine with dustproof function | |
CN204929083U (en) | Pay close attention to target search device based on imitative hawk eye vision | |
CN107764835A (en) | A kind of electronic product glass cover-plate detection means and method | |
CN215003685U (en) | Detection robot | |
CN116389857B (en) | Plant phenotype acquisition platform, method, electronic equipment and storage medium | |
CN220016635U (en) | Facility crop phenotype collection device | |
CN213957173U (en) | Weed detection device under complementary scene of farming light | |
CN214198041U (en) | Wheat image acquisition equipment in field | |
CN111157757A (en) | Vision-based crawler speed detection device and method | |
CN210491742U (en) | Distributed positioning device for fruit and automatic picking vehicle |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |