CN109101029A - Inspection evil intelligent carriage and its working method are gone on patrol in a kind of orchard automatically - Google Patents
Inspection evil intelligent carriage and its working method are gone on patrol in a kind of orchard automatically Download PDFInfo
- Publication number
- CN109101029A CN109101029A CN201811018739.XA CN201811018739A CN109101029A CN 109101029 A CN109101029 A CN 109101029A CN 201811018739 A CN201811018739 A CN 201811018739A CN 109101029 A CN109101029 A CN 109101029A
- Authority
- CN
- China
- Prior art keywords
- information
- automatic
- orchard
- pest
- control center
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 44
- 239000002420 orchard Substances 0.000 title claims abstract description 37
- 238000007689 inspection Methods 0.000 title claims abstract description 25
- 230000007246 mechanism Effects 0.000 claims abstract description 87
- 238000001514 detection method Methods 0.000 claims abstract description 59
- 241000607479 Yersinia pestis Species 0.000 claims abstract description 49
- 201000010099 disease Diseases 0.000 claims abstract description 37
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims abstract description 37
- 238000012545 processing Methods 0.000 claims abstract description 30
- 230000006378 damage Effects 0.000 claims abstract description 23
- 235000013399 edible fruits Nutrition 0.000 claims abstract description 12
- 238000004458 analytical method Methods 0.000 claims abstract description 7
- 230000000007 visual effect Effects 0.000 claims description 37
- 230000010365 information processing Effects 0.000 claims description 19
- 238000010586 diagram Methods 0.000 claims description 13
- 230000001681 protective effect Effects 0.000 claims description 7
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 239000000284 extract Substances 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 4
- 230000003993 interaction Effects 0.000 claims description 4
- 230000003321 amplification Effects 0.000 claims description 3
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 3
- 238000011017 operating method Methods 0.000 claims 1
- 230000007613 environmental effect Effects 0.000 abstract 1
- 238000011176 pooling Methods 0.000 description 22
- 230000006870 function Effects 0.000 description 20
- 241000207199 Citrus Species 0.000 description 15
- 235000020971 citrus fruits Nutrition 0.000 description 12
- 206010039509 Scab Diseases 0.000 description 10
- 230000004913 activation Effects 0.000 description 9
- 241000238631 Hexapoda Species 0.000 description 7
- 230000008569 process Effects 0.000 description 7
- 238000013461 design Methods 0.000 description 6
- 238000010606 normalization Methods 0.000 description 6
- 238000005286 illumination Methods 0.000 description 5
- 230000004044 response Effects 0.000 description 5
- 238000012549 training Methods 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000009194 climbing Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000003902 lesion Effects 0.000 description 4
- 210000002569 neuron Anatomy 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 238000007781 pre-processing Methods 0.000 description 3
- 241000196324 Embryophyta Species 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 239000000428 dust Substances 0.000 description 2
- 238000000265 homogenisation Methods 0.000 description 2
- 238000007477 logistic regression Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 229910000831 Steel Inorganic materials 0.000 description 1
- 208000025865 Ulcer Diseases 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000002775 capsule Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 238000001035 drying Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- JEIPFZHSYJVQDO-UHFFFAOYSA-N iron(III) oxide Inorganic materials O=[Fe]O[Fe]=O JEIPFZHSYJVQDO-UHFFFAOYSA-N 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000008832 photodamage Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000007480 spreading Effects 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
- G05D1/0253—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means extracting relative motion information from a plurality of images taken successively, e.g. visual odometry, optical flow
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0214—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0225—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving docking at a fixed facility, e.g. base station or loading bay
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0255—Control of position or course in two dimensions specially adapted to land vehicles using acoustic signals, e.g. ultra-sonic singals
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
- G05D1/0278—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle using satellite positioning signals, e.g. GPS
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Aviation & Aerospace Engineering (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Electromagnetism (AREA)
- Acoustics & Sound (AREA)
- Image Processing (AREA)
Abstract
Inspection evil intelligent carriage is gone on patrol automatically the invention discloses a kind of orchard and its working method, the trolley include mobile mechanism, for being moved according to instruction；Vision navigation system for acquiring and analyzing road condition information, and generates movement routine based on the analysis results, the control mobile mechanism operation when control centre sends enabling signal；Automatic camera examines evil system, generates corresponding warning information for acquiring and handling pomology information, and according to processing result, and carry out information exchange with control centre；Control centre examines evil system for starting mobile mechanism, vision navigation system and automatic camera, and receives the return signal of mobile mechanism and automatic camera inspection evil system；Power supply unit is that the control centre, mobile mechanism, vision navigation system and automatic camera examine evil system power supply.Trolley of the invention can carry out automatic patrol detection fruit tree diseases and pests, strong environmental adaptability, pest and disease damage Detection accuracy height in orchard.
Description
Technical Field
The invention relates to the technical field of intelligent agriculture, in particular to an orchard automatic patrol damage detection intelligent trolley and a working method thereof.
Background
In recent years, robotics has developed very rapidly and has been used in a wide range of applications, such as: fire scene rescue, explosion elimination, investigation, deep sea detection, patrol and the like are deeply loved and supported by people in various industries. Many research achievements of the robot are directly applied to various fields of national construction, national defense safety, life and the like, and play an important role in national major engineering construction, national prosperity and stability maintenance and stable and happy life guarantee of people.
At present, the inspection of the diseases and insect pests of the orchard is mainly carried out by a manual patrol method for inspection and prevention. The manual patrol inspection of the orchard diseases and insect pests not only takes time and labor, but also is difficult to realize timely and comprehensive inspection of the orchard diseases and insect pests due to the restriction of fatigue, limited visual field, plant leaf shielding, illumination, rugged terrain and the height of people. Therefore, a device capable of automatically patrolling and detecting diseases and insect pests of fruit trees in an orchard is urgently needed.
Disclosure of Invention
In view of the above, the invention provides an automatic patrol and pest detection intelligent trolley for an orchard and a working method thereof.
In order to achieve the above object, according to one aspect of the present invention, there is provided an orchard automatic patrol damage detection intelligent vehicle, comprising:
the moving mechanism is used for moving according to the instruction;
the visual navigation system is used for acquiring and analyzing road condition information, generating a moving path according to an analysis result, and controlling the moving mechanism to operate when the control center sends a starting signal;
the automatic photographing and pest detecting system is used for acquiring and processing information of the fruit trees, generating corresponding early warning information according to a processing result and carrying out information interaction with the control center;
the control center is used for starting the moving mechanism, the visual navigation system and the automatic photographing and pest detecting system and receiving return signals of the moving mechanism and the automatic photographing and pest detecting system;
and the power supply unit supplies power for the control center, the moving mechanism, the visual navigation system and the automatic photographing and pest detecting system.
Preferably, the moving mechanism comprises a vehicle body, a crawler mechanism, a driving mechanism and a start-stop control system; the crawler mechanisms are symmetrically arranged on two sides of the vehicle body; the start-stop control system is used for controlling the start-stop of the driving mechanism; the driving mechanism comprises a driving amplifying circuit and a stepping motor, the driving amplifying circuit amplifies a control signal of the visual navigation system, and the stepping motor is controlled to drive the track mechanism. The crawler-type chassis structure can realize the climbing capability of hilly lands, and improves the adaptability of the trolley under complex terrains such as hills.
Preferably, the visual navigation system comprises a distance detection device, a camera, a processor and a motion controller;
the camera is arranged at the front end of the vehicle body and used for collecting road image information;
the processor is used for processing road distance information acquired by the distance detection device and road image information acquired by the camera and planning a path, and particularly, the camera and the processor can be fixed at the front end of the vehicle body 9 by a support;
the motion controller is used for controlling the motion of the moving mechanism according to the planned path information;
the distance information output end of the distance detection device is connected with the distance information input end of the processor, the image information output end of the camera is connected with the image information input end of the processor, the path information output end of the processor is connected with the path information input end of the motion controller, and the control signal output end of the motion controller is connected with the signal input end of the driving amplification circuit. And the automatic patrol function is realized by taking the center line of the road as a navigation basis.
Further, the distance detection device is arranged at the front end and the rear end of the vehicle body. And feeding back road distance information in time.
Furthermore, the visual navigation system also comprises a photoelectric encoder, and the signal output end of the photoelectric encoder is connected with the angle signal input end of the motion controller. And feeding back the information of the running angle of the trolley in time.
Preferably, the automatic photographing and pest detection system comprises an automatic photographing mechanism and an information processing terminal, the automatic photographing mechanism photographs according to the instruction of the control center and sends the photographed picture to the information processing terminal, and the information processing terminal preprocesses the received picture and generates a processing result and corresponding early warning information. The detection of fruit tree diseases and insect pests is realized.
Further, the automatic photographing mechanism comprises a rotating device arranged on the vehicle body bottom plate and a telescopic device arranged on the rotating device, wherein the top of the telescopic device is provided with a photographing device, and a protection bag device used for protecting the photographing device is arranged. The important parts of the trolley are designed with the protective bag device, so that even if the trolley is carelessly turned over during walking, the trolley cannot be greatly lost. The trolley is kept balanced, the turning emergency stop function can be realized, and the secondary damage caused by continuous operation after turning can not occur.
Further, the automatic photographing mechanism further comprises a servo motor, a ball screw mechanism and a balance sensor, wherein the ball screw mechanism is used for adjusting the telescopic state of the telescopic device.
In order to achieve the above object, according to one aspect of the present invention, the present invention provides a working method of an intelligent car for automatic patrol inspection in an orchard, comprising the following steps:
s1, starting the trolley, and sending a starting instruction to the start-stop control system by the control center;
s2, the start-stop control system generates random time N and sends an operation instruction to the control center;
s3, the control center sends a starting instruction to the visual navigation system, and the visual navigation system starts road navigation work;
s4, after N seconds, the start-stop control system controls the trolley to stop and sends a stop instruction to the control center;
s5, the control center sends a stop instruction to the visual navigation system and simultaneously sends a start instruction to the automatic photographing damage detection system, and the automatic photographing damage detection system responds to the start instruction of the control center to start working;
s6, resetting after the photographing is finished, and sending a reset instruction to the control center;
steps S1-S6 are repeated.
Preferably, the method further comprises processing the picture taken by the automatic photographing mechanism, and the specific processing steps are as follows:
a1, adding corresponding photo information into the shot photo by the automatic photographing mechanism, and sending the photo added with the photo information to the information processing terminal, wherein the photo information comprises shooting time and shooting position;
a2, the information processing terminal extracts suspected pest information in the received picture by using a sliding window method;
a3, putting a characteristic diagram of suspected pest and disease information obtained through characteristic extraction into a trained convolutional neural network model to detect whether the characteristic diagram is a pest and disease characteristic, if so, executing the step A4, and if not, deleting the characteristic diagram containing the suspected pest and disease information;
and A4, performing characteristic marking on the picture containing the pest and disease characteristics, storing the picture, and giving an alarm, wherein the characteristic marking comprises the shooting time, the shooting place, the pest type information and the pest quantity information of the marked picture.
Due to the adoption of the technical scheme, the invention has the following beneficial effects:
1. the crawler-type chassis structure can realize the climbing capability of hilly lands, and improve the adaptability of the trolley in complex terrains such as hills;
2. the automatic patrol detection of the diseases and insect pests of the fruit trees can be carried out in the orchard, and the disease and insect pest detection accuracy is high;
3. disease information can be fed back to the garden owner, so that time guarantee is provided for timely finding the disease and taking corresponding measures, and loss caused by the disease is reduced;
4. the protective bag structure is designed for important parts of the trolley, so that even if the trolley carelessly turns over when walking, too large loss can not occur, and the device for turning over and stopping suddenly is equipped, so that the secondary damage caused by continuous operation after turning over can not occur.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present invention will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a schematic top view of a structure of an intelligent automatic patrol inspection vehicle for an orchard in a preferred embodiment of the invention;
FIG. 2 is a schematic left side view of the structure of an intelligent car for automatic patrol inspection in an orchard according to a preferred embodiment of the invention;
FIG. 3 is a flow chart of a working method of an intelligent orchard vehicle for automatic patrol inspection in accordance with a preferred embodiment of the present invention;
FIG. 4 is a flow chart of the work of the intelligent orchard vehicle for automatic patrol and damage detection in a preferred embodiment of the invention;
FIG. 5 is a schematic structural diagram of a moving mechanism of an intelligent car for automatic patrol and damage detection in an orchard according to a preferred embodiment of the invention;
FIG. 6 is a flowchart of the operation of the visual navigation system of the intelligent car for automatic patrol and damage detection in an orchard in accordance with a preferred embodiment of the present invention;
FIG. 7 is a frame diagram of an automatic photo-damage detection system of an intelligent car for automatic patrol detection in an orchard according to a preferred embodiment of the invention;
fig. 8 is a diagram illustrating an Alexnet network structure according to a preferred embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the description of the present invention, it is to be understood that the terms "longitudinal", "lateral", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like, indicate orientations or positional relationships based on those shown in the drawings, and are used merely for convenience of description and for simplicity of description, and do not indicate or imply that the referenced devices or elements must have a particular orientation, be constructed in a particular orientation, and be operated, and thus, are not to be construed as limiting the present invention.
In the description of the present invention, unless otherwise specified and limited, it is to be noted that the terms "mounted," "connected," and "connected" are to be interpreted broadly, and may be, for example, a mechanical connection or an electrical connection, a communication between two elements, a direct connection, or an indirect connection via an intermediate medium, and specific meanings of the terms may be understood by those skilled in the art according to specific situations.
Example 1
This embodiment provides an automatic intelligent vehicle of inspection of patrolling in orchard, as shown in fig. 1 and 2, include:
the moving mechanism is used for moving according to the instruction and carrying out information interaction with the control center;
the visual navigation system is used for acquiring and analyzing road condition information, generating a moving path according to an analysis result, and controlling the moving mechanism to operate when the control center sends a starting signal;
the automatic photographing and pest detecting system is used for acquiring and processing information of the fruit trees, generating corresponding early warning information according to a processing result and carrying out information interaction with the control center;
the control center is used for starting the moving mechanism, the visual navigation system and the automatic photographing and pest detecting system and receiving return signals of the moving mechanism and the automatic photographing and pest detecting system;
and the power supply unit 5 is used for supplying power to the control center, the moving mechanism, the visual navigation system and the automatic photographing and pest detecting system.
Specifically, because the intelligent trolley has large electric quantity loss, the simple assembled battery can not meet the use requirement, and therefore the storage battery is selected to provide electric energy.
Preferably, the moving mechanism comprises a vehicle body 9, a crawler mechanism, a driving mechanism and a start-stop control system; the crawler mechanisms are symmetrically arranged on two sides of the vehicle body 9; the start-stop control system is used for controlling the start-stop of the driving mechanism; the driving mechanism comprises a driving amplifying circuit and a stepping motor 2, the driving amplifying circuit amplifies a control signal of the visual navigation system, and the stepping motor 2 is controlled to drive the crawler mechanism.
Specifically, the bottom plate of the vehicle body 9 is made of hard plastic plates and is fixedly connected to the chassis frame of the vehicle body 9 through bolts, the structure is simple and light, the cost is low, the weight of the vehicle body is reduced, electric energy is saved, and the battery endurance time is prolonged.
The crawler mechanism includes a crawler 3, crawler wheels 13, and a shock-absorbing device 12. A stepping motor 2 in the driving mechanism is connected with a crawler wheel 13, and a start-stop control system controls the running and stopping of the stepping motor 2.
Preferably, the visual navigation system comprises a distance detection device, a camera, a processor and a motion controller;
the camera is fixed at the front end of the vehicle body by a bracket and is used for collecting road image information;
the processor is used for processing road distance information acquired by the distance detection device and road image information acquired by the camera and planning a path, and particularly, the camera and the processor can be fixed at the front end of the vehicle body 9 by a support;
the motion controller is used for controlling the motion of the moving mechanism according to the planned path information;
the distance information output end of the distance detection device is connected with the distance information input end of the processor, the image information output end of the camera is connected with the image information input end of the processor, the path information output end of the processor is connected with the path information input end of the motion controller, and the control signal output end of the motion controller is connected with the signal input end of the driving amplification circuit.
Further, the distance detection device adopts, but is not limited to, an ultrasonic distance sensor, and comprises a front distance detection device 8 and a rear distance detection device 1; the front distance detection devices 8 are symmetrically arranged on two sides of the front end of the vehicle body 9, and the rear distance detection devices 1 are arranged in the middle of the rear end of the vehicle body 9.
Furthermore, the visual navigation system also comprises a photoelectric encoder, and the signal output end of the photoelectric encoder is connected with the angle signal input end of the motion controller.
Preferably, the automatic photographing and damage detecting system comprises an automatic photographing mechanism and an information processing terminal, the automatic photographing mechanism photographs according to an instruction of the control center and sends the photographed picture to the information processing terminal, and the information processing terminal preprocesses the received picture and generates a processing result and corresponding early warning information.
Further, automatic mechanism of shooing is equipped with the device 15 of shooing including setting up rotary device on automobile body 9 bottom plate and the telescoping device 4 of setting on rotary device at telescoping device 4 top to be equipped with the protection bag device that is used for the protection device 15 of shooing. Specifically, the telescopic device can be a hydraulic telescopic rod, and the information processing terminal is embedded into the photographing device 15.
Further, the automatic photographing mechanism further comprises a servo motor, a ball screw mechanism 6 for adjusting the telescopic state of the telescopic device 4 and a balance sensor. The ball screw mechanism 6 is connected with the upper end of a rodless cavity of the hydraulic telescopic rod through the supporting rod 11, the control center controls the servo motor 7 to drive the ball screw mechanism 6 to enable the telescopic device 4 to be unfolded for shooting, and the telescopic device resets after the shooting is finished. The balance sensor is a level meter or an inclinometer, is arranged on a tray supporting the photographing device 15 and is used for assisting in adjusting the telescopic state of the telescopic device 4, and when the photographing device is detected to be horizontal, the telescopic state is stopped to be adjusted; the balance sensor is also used for realizing a rollover emergency stop function, namely when the detected angle signal is too large (the angle interval can be set to be 50-90 degrees, namely, emergency braking is realized when the angle is positioned in the interval), the control center cuts off a power supply circuit, and the emergency braking is realized.
The start-stop control system and the motion controller are both arranged in the control box 10.
In order to achieve the above object, according to one aspect of the present invention, the present invention provides a working method of an intelligent car for automatic patrol inspection in an orchard, as shown in fig. 3, including the following steps:
s1, starting the trolley, and sending a starting instruction to the start-stop control system by the control center;
s2, the start-stop control system generates random time N and sends an operation instruction to the control center;
s3, the control center sends a starting instruction to the visual navigation system, and the visual navigation system starts road navigation work;
s4, after N seconds, the start-stop control system controls the trolley to stop and sends a stop instruction to the control center;
s5, the control center sends a stop instruction to the visual navigation system and simultaneously sends a start instruction to the automatic photographing damage detection system, and the automatic photographing damage detection system responds to the start instruction of the control center to start working;
s6, resetting after the photographing is finished, and sending a reset instruction to the control center;
steps S1-S6 are repeated.
Preferably, the method further comprises processing the picture taken by the automatic photographing mechanism, and the specific processing steps are as follows:
a1, adding corresponding photo information into the shot photo by the automatic photographing mechanism, and sending the photo added with the photo information to the information processing terminal, wherein the photo information comprises shooting time and shooting position;
a2, the information processing terminal extracts suspected pest information in the received picture by using a sliding window method;
a3, putting the characteristic diagram of suspected pest and disease information obtained through characteristic extraction into a trained convolutional neural network model to detect whether the characteristic diagram is a pest and disease characteristic, if so, executing the step A4, and if not, deleting the characteristic diagram containing the suspected pest and disease information;
and A4, performing characteristic marking on the picture containing the pest and disease characteristics, storing the picture, and giving an alarm, wherein the characteristic marking comprises the shooting time, the shooting place, the pest type information and the pest quantity information of the marked picture.
Example 2
In order to further explain the system structure and the working process of the intelligent orchard patrol inspection trolley in detail, the intelligent orchard patrol inspection trolley is used for detecting citrus canker.
The method is applied to the aspect of disease detection in the field of citrus planting, the work flow is that the control center starts the start-stop control system of the moving mechanism after clicking starts, the start-stop control system controls the crawler mechanism to advance, the start-stop control system controls the crawler mechanism to stop after finishing command triggering, the control center controls the automatic photographing damage detection system to photograph, and the control center starts to circulate until clicking is flamed out, and is specifically shown as 4.
The citrus is generally planted in hilly areas, sloping fields are more, and citrus planting has certain scientific basis, the interval between two transverse citrus plants is generally 3 meters, the longitudinal interval is 2 meters, citrus trees are generally not more than 1.8 meters, a wider ridge is usually cleaned out to carry out daily management for fruit growers in order to facilitate management, the width is generally about 1 meter, conditions are provided for the successful implementation of the intelligent car for automatic patrol and damage detection in the orchard, and the intelligent car for automatic patrol and damage detection in the orchard mainly comprises the following parts: the system comprises a moving mechanism, a visual navigation system and an automatic photographing and pest detecting system.
1. Moving mechanism
Considering that most of citrus plants are hilly lands, most of terrains have certain gradients, but most of the gradients are not larger than 25 degrees, so that the climbing capacity of the trolley cannot be smaller than 25 degrees. The invention carries out comparative analysis on the advantages and the disadvantages of the crawler-type and wheel-type walking structures, and the analysis results are shown in the following table:
through the analysis of the characteristics of the upper table and the terrain, the walking mechanism of the trolley is designed by selecting the crawler-type chassis. Because the crawler type travelling mechanism has a lower chassis, the crawler type travelling mechanism has higher mass and the like, the gravity center of the crawler type vehicle is lower, and because the travelling speed of the crawler type vehicle is between 0.5 and 0.7 m/s (the crawler type is slower), the factor of turning over is basically not needed to be worried about in the climbing process. In addition, because the crawler belt, the sensor, the navigation system, the controller and the like of the trolley all need a power supply, the simple assembled battery pack can not meet the use requirement, and the accumulator jar is selected to provide energy.
The components of the moving mechanism are shown in fig. 5.
The moving mechanism drives the crawler belts to walk in a two-side driving mode, and turning, straight advancing and the like are realized through the speed difference of the crawler belts at the two sides; the wheel support department on dolly chassis both sides adopts steel construction joint support, and specific chassis plane adopts the stereoplasm plastic slab, connects with bolt and nut, can alleviate dolly heavy burden, energy resource consumption, rust etc. on the one hand like this, on the other hand uses the plastic slab structure to reduce the contact on power transmission line and ground, in addition in case seal can also keep inside clean.
1.1 starting and stopping control system
The start-stop control system is used for controlling the trolley to walk and stop. If the stop of each tree is selected to be taken during the patrol detection process, the working efficiency of the trolley is low, and the trolley is very wasteful. Therefore, the random number N (calculated according to the trolley moving speed of 0.5m/s, wherein N is taken from {4,8,12,16,20,24,28,32,36,40}) is designed to represent the time of each advance, so that the purpose of random sampling inspection can be achieved, and much time can be saved. The work flow is shown in fig. 4.
③ the ③ method ③ comprises ③ the ③ steps ③ of ③ starting ③, ③ clicking ③ a ③ start ③ button ③, ③ (③ 1 ③) ③ sending ③ a ③ start ③ instruction ③ ① ③ to ③ a ③ start ③ - ③ stop ③ control ③ system ③ by ③ a ③ control ③ center ③, ③ enabling ③ the ③ start ③ - ③ stop ③ control ③ system ③ ① ③ to ③ generate ③ random ③ time ③ and ③ send ③ a ③ running ③ instruction ③ ① ③ to ③ the ③ control ③ center ③, ③ sending ③ a ③ start ③ instruction ③ ① ③ to ③ a ③ control ③ center ③ ② ③ by ③ the ③ control ③ center ③, ③ enabling ③ the ③ visual ③ navigation ③ system ③ ① ③ to ③ start ③ working ③, ③ starting ③ and ③ stopping ③ the ③ control ③ system ③ ① ③ to ③ time ③, ③ controlling ③ a ③ trolley ③ ① ③ to ③ stop ③ and ③ send ③ a ③ stop ③ instruction ③ ① ③ to ③ the ③ control ③ center ③ by ③ the ③ start ③ - ③ stop ③ control ③ system ③ after ③ N ③ seconds ③, ③ then ③ sending ③ a ③ stop ③ instruction ③ ① ③ to ③ the ③ control ③ center ③ ② ③ by ③ the ③ control ③ center ③, ③ and ③ sending ③ a ③ start ③ instruction ③ ① ③ to ③ an ③ automatic ③ photographing ③ damage ③ detection ③ system ③, ③ (③ 2 ③) ③ starting ③ the ③ automatic ③ photographing ③ damage ③ detection ③ system ③ ① ③ to ③ work ③, ③ finishing ③ photographing ③ and ③ resetting ③, ③ (③ 3 ③) ③ feeding ③ a ③ reset ③ instruction ③ back ③ ① ③ to ③ the ③ control ③ center ③ after ③ the ③ automatic ③ photographing ③ mechanism ③ is ③ reset ③, ③ repeating ③ the ③ three ③ steps ③ ① ③ to ③ realize ③ the ③ automatic ③ patrol ③ function ③ of ③ the ③ trolley ③, ③ and ③ finally ③ clicking ③ a ③ flameout ③ button ③ ① ③ to ③ stop ③ the ③ vehicle ③. ③
2. Visual navigation system
The path navigation occupies a very important position in the design of the trolley, the cause and effect circle is complex, the road is narrow and soft, and the road and the surrounding background are not clearly distinguished, so that the design of a reasonable navigation system with strong adaptability is particularly important. The system comprises environment image acquisition, image processing, motion decision, motion control, motion information feedback, analysis, preprocessing, identification and detection of shot images and the like, wherein the trolley is controlled through sequential circulation, and the whole thought is clear and is not easy to be disordered, as shown in fig. 6.
In a navigation system, the quality of the obtained visual image directly affects the results of image processing and feature extraction, and because the trolley jolts and shakes due to the unevenness of the road surface when the trolley advances in a hilly area, the exposure time of a camera is ensured to be short enough to ensure that the image is clear, and meanwhile, the real-time performance of navigation can be improved. When the camera is selected, the speed and the quality of the image collected by the camera are ensured to be good, the exposure speed of the image is high, and the like. In addition, the lens of the video camera also requires a relatively large viewing angle range to ensure that the camera can shoot more road information under the condition of being relatively close to the ground, and meanwhile, the camera lens in the automatic shooting mechanism also needs to meet the standard.
The hardware part for acquiring the road information and the corresponding requirements are adopted, the software part for identifying the road is a core, and a road information acquisition and processing program is embedded into the camera, so that the camera can respond to and process image signals in time. Before extracting road edge information, corresponding preprocessing needs to be performed on an image, because the acquired image under natural conditions has the problems of light spots, pocks, motion blur, unobvious edge effect and the like due to the influence of factors such as illumination, jitter, dust and the like, the image needs to be subjected to processing such as enhancement processing, blurred image recovery, image size compression, denoising and the like. The edge information in the preprocessed image is segmented by a fixed threshold segmentation method based on H components, and many other irrelevant or interference boundaries exist in the segmented result, so that the boundary of the road is extracted by adopting a maximum connected boundary method, discrete points are marked and fitted to obtain a road center line (which may be a straight line or a curve depending on the road condition). The navigation system plans the advancing direction and path of the trolley by combining the distance information detected by the distance detection device and the recognition result of the road center line, and realizes the advancing, turning and the like of the trolley by controlling the moving mechanism through the motion controller, thereby realizing the purpose of automatic patrol of the trolley.
The actual working process of the navigation system is as follows:
(1) after the start-stop control system is started, feeding back an operation instruction to a control center, sending a start instruction to the visual navigation system by the control center, and starting the visual navigation system;
(2) and the start-stop control system starts ⑤ to time, the start-stop control system controls the trolley ⑤ to stop after N seconds and sends a stop instruction ⑤ to the control center, then the control center sends a stop instruction ⑤ to the visual navigation system, and the visual navigation system stops running.
3. Automatic photographing and pest detecting system
The information terminal extracts disease spot information possibly existing in the picture by using a sliding window method, puts an obtained feature map which is suspected of disease spots into a trained convolutional neural network model to detect whether the disease spots are citrus canker disease spots or not, and sends out an alarm if the disease spots are detected, and records position information and time information for reference of fruit growers in handling dangerous cases. The automatic photographing damage detection system comprises an information processing terminal and an automatic photographing mechanism, wherein the automatic photographing mechanism comprises a protective bag device, a telescopic device 11, a rotating device and a photographing device, and the photographing device comprises a rotary photographing structure and a camera, as shown in fig. 7.
The telescopic device adopts a hydraulic telescopic rod structure, and has better stability and compactness compared with a folding structure and a lead screw nut structure. The bottom is ball screw structure, is mainly used for adjusting hydraulic stem structure during operation angle, avoids making the dolly dangerous because of unnecessary overturning moment. The automatic photographing and pest detecting system comprises the following working processes:
the automatic photographing pest detection system receives a starting instruction ⑥ of the control center, and firstly opens the protection bag device;
after the protective bag device is completely opened, the ball screw structure and the balance sensor jointly adjust the hydraulic telescopic rod structure to reach a vertical state, the hydraulic telescopic rod extends out to reach a specified position (namely the balance sensor detects that the hydraulic telescopic rod structure is in a horizontal state), and information is fed back to the photographing device;
triggering a circuit switch trigger of the photographing device, photographing pictures of surrounding fruit trees by rotating the photographing structure, rotating 180 degrees to photograph a right picture after the left photographing is finished, sending picture information to the information processing terminal and resetting after the photographing is finished, and switching off the circuit switch trigger of the photographing device after the sending is finished and feeding back the information to the hydraulic telescopic rod structure;
the hydraulic telescopic structure retracts to reach a specified position and feeds information back to the ball screw structure;
the ball screw structure returns to the original state again under the driving of the stepping motor and feeds back information to the protective capsule device;
the protective bag device is closed, and a reset command ⑦ is fed back to the control center.
3.1 information processing part of automatic photographing and pest detecting system
The invention has the primary purpose of automatically detecting the citrus plant diseases, reduces the time and capital expenditure of the link of examining and verifying by experts in the traditional detection method, can feed back the disease information to fruit growers in time, and has particularly obvious effects on preventing the disease from spreading and treating dangerous cases in time; and secondly, realizing the automatic patrol function in the orchard. In the aspect of identifying and detecting diseases of citrus crops, a lot of experiments are already carried out, preprocessing such as histogram equalization, normalization, drying removal processing and fuzzy image screening is carried out on pictures, a network model suitable for detecting citrus canker is successfully trained by utilizing an improved convolutional neural network, the detection speed of the network is optimized, and the detection speed of a single picture is less than 0.1s, so that the network model is embedded into an information terminal in the design, and auxiliary functions (information alarm, time recording, position information and the like are added, and the acquisition of the position information is realized by adopting a GPS (global positioning system) positioning mode at present). The specific workflow is as follows:
after the photo taking mechanism finishes photo collection, the two collected photos and the information such as the time and the position of the collection are transmitted back to the information terminal;
processing the motion blur in the returned image data, enhancing the image, and detecting, identifying and storing the target area by using a sliding window method;
because the shooting area is large and the number of the leaves is large, the area suspected of the disease spots in the picture is detected and marked by a sliding window method, the detected target picture is placed in a network for identification, warning information is sent out as long as the citrus canker disease spots exist, and information such as time and position for shooting the picture is recorded (for the fruit growers to use for reference in processing the disease plants).
3.2 introduction of orange ulcer disease speckle detection method
The existing methods for identifying and detecting the citrus canker lesion spots mainly comprise two methods: the method comprises the steps of identification detection based on a traditional machine learning method and identification detection based on a deep learning method. The invention adopts a deep learning method based on a convolutional neural network, and based on the existing Alexnet network structure, the parameters of the network are modified, and a network model capable of successfully detecting citrus canker lesions is trained.
3.3 Alexnet network model introduction
Alexnet contains convolutional layers, pooling layers, fully-connected layers, softmax loss layers, each of which is described below:
3.3.1 convolution layer
The biggest feature of convolutional layers is local sensing and weight sharing. The local perception means that the pixel value of a local image is input into a neuron, so that the number of connections is reduced, and the number of connections is also reducedThe parameters of the network are reduced. Weight sharing means that several neurons share the same weight. Convolutional layer convolution operations, which are used to extract features. One convolution kernel can extract one feature map, and one convolution layer is generally composed of a plurality of convolution kernels and can extract a group of feature maps. Assuming the layer I is a convolutional layer, the layer I has the jth feature mapThe calculation formula of (a) is as follows:
wherein,the jth feature map in the l layer; mjRepresenting a set of input feature maps; the number indicates the convolution operation of the convolution kernel k and the feature graph x; b is a bias parameter; f (-) is the activation function.
3.3.2 layers of pooling
The pooling operation reduces the scale sensitivity of the network to the image. Pooling is typically done after the convolutional layer to reduce the size of the feature map. Assuming that the size of the pooling layer convolution kernel is n × n and the step size is n, the pooling operation is to map the region of the feature map with the size of n × n into a pixel point on the new feature map by taking the maximum value or the average value. After pooling, the length or width of the feature map becomes 1/n of the original. If the step size of the pooling layer is not smaller than the size of the convolution kernel, it is called non-overlapping pooling.
3.3.3 full connection layer
The fully connected layer is used to classify the input features as in neural networks. To reduce the overfitting of the fully connected layers, the Dropout method may be used. The Dropout method is used to control whether the neuron appears or hides, which means ignoring all connections of the neuron. Thus the structure of the fully-connected layer is not fixed during training. This is as if multiple neural networks were learned, and the output result is decided by the multiple neural networks in common.
3.3.4 SoftmaxLoss layer
The Softmax function is used in the multi-classification problem and is extended by a binary logistic regression. The output of the logistic regression function is the score probability of each class, and the class with the highest score probability is selected as the classification result.
Let the training data set be D { (x)1,y1),(x2,y2),...,(xm,ym)}，xi∈Rn+1,xi0＝1， yiE.g., {0, 1., k }. For the training sample x, the k-dimensional vector output by the Softmax target function is represented as hθ(x)＝(hθ1(x),hθ2(x),...,hθk(x))
WhereinIs represented as follows:
the loss function of the Softmax model is defined as:
for solving the theta value by the minimization loss function, a gradient descent method can be used for solving.
3.4 model training
The sample set of the experiment is 876 samples of citrus canker scab specimens (after removing the quite fuzzy picture) identified by professional departments, and 6000 samples without scabs and similar scabs (after removing the quite fuzzy picture). Considering the problem of data imbalance of positive and negative samples (the positive sample is a scab sample picture, and the negative sample is a scab-free and similar scab sample picture), expanding scab sample data by methods of horizontal turning, picture resolution changing, noise data adding, nearest neighbor interpolation and the like, and obtaining 2301 scab pictures. Because the problems of uneven illumination, pockmarks, dust and the like exist in the positive sample, the data set is preprocessed by illumination homogenization, histogram equalization, denoising and the like, and the data is divided according to the following table:
and under a caffe framework, training the sample data by using a GPU, and reflecting the identification accuracy rate of the network by using the true rate, the false positive rate and the error rate. The results of network prediction are represented by TP, FP, FN, TN, as shown in the following table:
the calculation formulas of True Rate (TPR), False Positive Rate (False Positive Rate), and Error Rate (ER) are as follows:
the values of TPR, FPR and ER are all within the interval [0,1 ]. The true rate is used for measuring the classifying ability of the classifier on the positive sample, and the higher the true rate is, the higher the recognition rate of the classifier on the positive sample is. The false positive rate is used for measuring the classification capability of the classifier on the negative samples, and the lower the false positive rate is, the higher the recognition rate of the classifier on the negative samples is. The classifier is a good classifier only when the true rate is high and the false positive rate is low, the recognition effect on positive samples and negative samples is good, and the error rate is low at the moment.
The neural network algorithm adopted by the design uses 5 convolutional layers, 3 pooling layers and 3 full-connection layers, and the specific connection relation is shown in fig. 8.
Inputting the preprocessed picture into a first convolution layer, performing convolution operation on the preprocessed picture and a convolution kernel, using Relu as an activation function, processing data by using local response normalization, and outputting a characteristic diagram; inputting the feature graph output by the first convolution layer into a first pooling layer, pooling by using a maximum pooling method, and outputting the feature graph; inputting the feature map output by the first pooling layer into a second convolution layer, performing convolution operation with a convolution kernel, processing data by using Relu as an activation function and using local response normalization, and outputting the feature map; inputting the feature map output by the second convolution layer into a second pooling layer, pooling by using a maximum pooling method, and outputting the feature map; inputting the feature map output by the second pooling layer into a third convolution layer, performing convolution operation with a convolution kernel, processing data by using Relu as an activation function and using local response normalization, and outputting the feature map; inputting the feature graph output by the third convolution layer into a fourth convolution layer, performing convolution operation with a convolution kernel, processing data by using Relu as an activation function and using local response normalization, and outputting the feature graph; inputting the feature graph output by the fourth convolution layer into the fifth convolution layer, performing convolution operation on the feature graph and a convolution kernel, using Relu as an activation function, using local response normalization to process data, and outputting the feature graph; inputting the feature map output by the fifth convolutional layer into a third pooling layer, pooling by using a maximum pooling method, and outputting the feature map; inputting a feature map output by the third pooling layer into the first full-connection layer, using Relu as an activation function, reducing overfitting of the full-connection layer by using a Dropout method, and classifying the input features; inputting the output result of the first full-connection layer into a second full-connection layer, using Relu as an activation function, reducing the overfitting of the full-connection layer by using a Dropout method, and classifying the input features; inputting a feature graph output by the second full-connection layer into a third full-connection layer, using Relu as an activation function, reducing overfitting of the full-connection layer by using a Dropout method, and classifying input features; and the output result of the third fully-connected layer is input into a Softmax layer to classify or calculate errors.
By experiment TPR ═ 0.9804, FPR ═ 0.0129, and ER ═ 0.0141. In addition, the design also improves the network structure, the network speed is improved under the condition of ensuring the accuracy, and the picture detection speed is improved from 0.068965 seconds per picture to 0.050668 seconds per picture.
3.5 implementation of lesion detection in the invention
Because the industrial camera collects high-definition pictures, serious hysteresis can be generated if the high-definition pictures are directly transmitted to the remote information terminal in a wireless mode, and the influence of signal quality is large, so that the information terminal is transplanted to the mobile device by the design choice. The automatic photographing device collects the photos and then transmits the photos back to the information terminal, the transmitted photos need to be preprocessed such as enhancement, illumination homogenization and denoising of the photos due to the wide photographing range, the limitation of photographing conditions and the like, and the sliding window is used for segmenting and extracting similar scab areas.
And transmitting the picture extracted by the sliding window to an information terminal for testing, marking and storing the picture with the scab in the test result, and giving an alarm. The mark storage content comprises shooting time information, shooting position information and lesion number information.
It should be noted that the system structures or method flows shown in fig. 1 to fig. 8 of the present invention are only some preferred embodiments of the present invention, and the illustration is only for the convenience of understanding the present invention and is not to be construed as a limitation of the present invention.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
While embodiments of the invention have been shown and described, it will be understood by those of ordinary skill in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.
Claims (10)
1. The utility model provides an automatic intelligent vehicle of inspection of patrolling in orchard which characterized in that includes:
the moving mechanism is used for moving according to the instruction;
the visual navigation system is used for acquiring and analyzing road condition information, generating a moving path according to an analysis result, and controlling the moving mechanism to operate when the control center sends a starting signal;
the automatic photographing and pest detecting system is used for acquiring and processing information of the fruit trees, generating corresponding early warning information according to a processing result and carrying out information interaction with the control center;
the control center is used for starting the moving mechanism, the visual navigation system and the automatic photographing and pest detecting system and receiving return signals of the moving mechanism and the automatic photographing and pest detecting system;
and the power supply unit supplies power for the control center, the moving mechanism, the visual navigation system and the automatic photographing and pest detecting system.
2. An orchard automatic patrol inspection intelligent trolley according to claim 1, wherein the moving mechanism comprises a trolley body, a crawler mechanism, a driving mechanism and a start-stop control system; the crawler mechanisms are symmetrically arranged on two sides of the vehicle body; the start-stop control system is used for controlling the start-stop of the driving mechanism; the driving mechanism comprises a driving amplifying circuit and a stepping motor, the driving amplifying circuit amplifies a control signal of the visual navigation system, and the stepping motor is controlled to drive the track mechanism.
3. An orchard automatic patrol inspection intelligent trolley according to claim 1, wherein the visual navigation system comprises a distance detection device, a camera, a processor and a motion controller;
the camera is arranged at the front end of the vehicle body and used for collecting road image information;
the processor is used for processing road distance information acquired by the distance detection device and road image information acquired by the camera and planning a path;
the motion controller is used for controlling the motion of the moving mechanism according to the planned path information;
the distance information output end of the distance detection device is connected with the distance information input end of the processor, the image information output end of the camera is connected with the image information input end of the processor, the path information output end of the processor is connected with the path information input end of the motion controller, and the control signal output end of the motion controller is connected with the signal input end of the driving amplification circuit.
4. An orchard automatic patrol detection intelligent trolley according to claim 3, wherein the distance detection device is arranged at the front end and the rear end of the trolley body.
5. An orchard automatic patrol inspection intelligent trolley according to claim 4, wherein the visual navigation system further comprises a photoelectric encoder, and a signal output end of the photoelectric encoder is connected with an angle signal input end of the motion controller.
6. The intelligent orchard patrol inspection intelligent trolley according to claim 1, wherein the automatic photographing inspection system comprises an automatic photographing mechanism and an information processing terminal, the automatic photographing mechanism photographs according to instructions of the control center and sends the photographed photos to the information processing terminal, and the information processing terminal preprocesses the received photos and generates processing results and corresponding early warning information.
7. The intelligent orchard patrol inspection trolley according to claim 6, wherein the automatic photographing mechanism comprises a rotating device arranged on a bottom plate of a trolley body and a telescopic device arranged on the rotating device, the top of the telescopic device is provided with a photographing device, and a protective bag device for protecting the photographing device is arranged.
8. The intelligent orchard patrol inspection trolley according to claim 7, wherein the automatic photographing mechanism further comprises a servo motor, a ball screw mechanism for adjusting the telescopic state of the telescopic device and a balance sensor.
9. An operating method of an automatic patrol damage detection intelligent vehicle for an orchard is characterized by comprising the following steps:
s1, starting the trolley, and sending a starting instruction to the start-stop control system by the control center;
s2, the start-stop control system generates random time N and sends an operation instruction to the control center;
s3, the control center sends a starting instruction to the visual navigation system, and the visual navigation system starts road navigation work;
s4, after N seconds, the start-stop control system controls the trolley to stop and sends a stop instruction to the control center;
s5, the control center sends a stop instruction to the visual navigation system and simultaneously sends a start instruction to the automatic photographing damage detection system, and the automatic photographing damage detection system responds to the start instruction of the control center to start working;
s6, resetting after the photographing is finished, and sending a reset instruction to the control center;
steps S1-S6 are repeated.
10. The working method of the intelligent orchard patrol inspection trolley according to claim 9, further comprising the step of processing the pictures taken by the automatic photographing mechanism, wherein the specific processing steps are as follows:
a1, adding corresponding photo information into the shot photo by the automatic photographing mechanism, and sending the photo added with the photo information to the information processing terminal, wherein the photo information comprises shooting time and shooting position;
a2, the information processing terminal extracts suspected pest information in the received picture by using a sliding window method;
a3, putting a characteristic diagram of suspected pest and disease information obtained through characteristic extraction into a trained convolutional neural network model to detect whether the characteristic diagram is a pest and disease characteristic, if so, executing the step A4, and if not, deleting the characteristic diagram containing the suspected pest and disease information;
and A4, carrying out characteristic marking on the picture containing the pest and disease characteristics, storing the picture, and giving out an alarm, wherein the characteristic marking comprises the shooting time, the shooting place, the pest type information and the pest quantity information of the marked picture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811018739.XA CN109101029A (en) | 2018-09-03 | 2018-09-03 | Inspection evil intelligent carriage and its working method are gone on patrol in a kind of orchard automatically |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811018739.XA CN109101029A (en) | 2018-09-03 | 2018-09-03 | Inspection evil intelligent carriage and its working method are gone on patrol in a kind of orchard automatically |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109101029A true CN109101029A (en) | 2018-12-28 |
Family
ID=64864822
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811018739.XA Pending CN109101029A (en) | 2018-09-03 | 2018-09-03 | Inspection evil intelligent carriage and its working method are gone on patrol in a kind of orchard automatically |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109101029A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113812334A (en) * | 2021-09-16 | 2021-12-21 | 葛承暄 | Artificial intelligence robot device in agricultural production field |
| CN114415695A (en) * | 2022-03-28 | 2022-04-29 | 南京农业大学 | Tea garden inspection system based on vision technology and inspection robot |
| CN117671329A (en) * | 2023-11-14 | 2024-03-08 | 平安科技(上海)有限公司 | Vehicle damage analysis method, device, equipment and medium based on artificial intelligence |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101774170A (en) * | 2010-01-29 | 2010-07-14 | 华北电力大学 | Nuclear power plant working robot and control system thereof |
| EP2336719A2 (en) * | 2009-12-17 | 2011-06-22 | Deere & Company | Automated tagging for landmark identification |
| CN103918636A (en) * | 2014-04-29 | 2014-07-16 | 青岛农业大学 | Intelligent spraying method based on image processing and spraying robot based on image processing |
| CN105182976A (en) * | 2015-09-17 | 2015-12-23 | 西北农林科技大学 | Visual navigation strategy of agricultural robot |
| CN105894003A (en) * | 2016-04-29 | 2016-08-24 | 无锡中科智能农业发展有限责任公司 | Large-field fruit tree disease monitoring early-warning system based on machine vision |
| CN106017477A (en) * | 2016-07-07 | 2016-10-12 | 西北农林科技大学 | Visual navigation system of orchard robot |
| CN106250899A (en) * | 2016-07-29 | 2016-12-21 | 华东交通大学 | A kind of orange disease and insect pests monitoring and pre-alarming method based on distributed compression perception WSN |
| CN205843680U (en) * | 2016-07-07 | 2016-12-28 | 西北农林科技大学 | A kind of orchard robotic vision navigation system |
| CN107067043A (en) * | 2017-05-25 | 2017-08-18 | 哈尔滨工业大学 | A kind of diseases and pests of agronomic crop detection method |
-
2018
- 2018-09-03 CN CN201811018739.XA patent/CN109101029A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2336719A2 (en) * | 2009-12-17 | 2011-06-22 | Deere & Company | Automated tagging for landmark identification |
| CN101774170A (en) * | 2010-01-29 | 2010-07-14 | 华北电力大学 | Nuclear power plant working robot and control system thereof |
| CN103918636A (en) * | 2014-04-29 | 2014-07-16 | 青岛农业大学 | Intelligent spraying method based on image processing and spraying robot based on image processing |
| CN105182976A (en) * | 2015-09-17 | 2015-12-23 | 西北农林科技大学 | Visual navigation strategy of agricultural robot |
| CN105894003A (en) * | 2016-04-29 | 2016-08-24 | 无锡中科智能农业发展有限责任公司 | Large-field fruit tree disease monitoring early-warning system based on machine vision |
| CN106017477A (en) * | 2016-07-07 | 2016-10-12 | 西北农林科技大学 | Visual navigation system of orchard robot |
| CN205843680U (en) * | 2016-07-07 | 2016-12-28 | 西北农林科技大学 | A kind of orchard robotic vision navigation system |
| CN106250899A (en) * | 2016-07-29 | 2016-12-21 | 华东交通大学 | A kind of orange disease and insect pests monitoring and pre-alarming method based on distributed compression perception WSN |
| CN107067043A (en) * | 2017-05-25 | 2017-08-18 | 哈尔滨工业大学 | A kind of diseases and pests of agronomic crop detection method |
Non-Patent Citations (1)
| Title |
|---|
| 张敏等: "基于卷积神经网络的柑橘溃疡病识别方法", 《计算机应用》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113812334A (en) * | 2021-09-16 | 2021-12-21 | 葛承暄 | Artificial intelligence robot device in agricultural production field |
| CN114415695A (en) * | 2022-03-28 | 2022-04-29 | 南京农业大学 | Tea garden inspection system based on vision technology and inspection robot |
| CN117671329A (en) * | 2023-11-14 | 2024-03-08 | 平安科技(上海)有限公司 | Vehicle damage analysis method, device, equipment and medium based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109101029A (en) | Inspection evil intelligent carriage and its working method are gone on patrol in a kind of orchard automatically | |
| Bender et al. | A high‐resolution, multimodal data set for agricultural robotics: A Ladybird's‐eye view of Brassica | |
| US20110026770A1 (en) | Person Following Using Histograms of Oriented Gradients | |
| CN110728236B (en) | Vehicle loss assessment method and special equipment thereof | |
| CN109785363A (en) | A kind of unmanned plane video motion Small object real-time detection and tracking | |
| Li et al. | Fast detection and location of longan fruits using UAV images | |
| CN111460903B (en) | System and method for monitoring growth of field broccoli based on deep learning | |
| Ali et al. | Visual tree detection for autonomous navigation in forest environment | |
| CN114005021B (en) | Laser vision fusion based unmanned inspection system and method for aquaculture workshop | |
| CN108500992A (en) | A kind of multi-functional mobile security robot | |
| CN109816698A (en) | Unmanned plane visual target tracking method based on dimension self-adaption core correlation filtering | |
| Sambolek et al. | Person detection in drone imagery | |
| CN105913011B (en) | Human body anomaly detection method based on parameter self-regulation neural network | |
| CN101763117B (en) | Reconnaissance robot | |
| Sun et al. | Beet seedling and weed recognition based on convolutional neural network and multi-modality images | |
| Fahn et al. | A real-time pedestrian legs detection and tracking system used for autonomous mobile robots | |
| CN107333043B (en) | Image Acquisition and identifying system | |
| CN116307357A (en) | Animal monitoring method, device, terminal equipment and storage medium | |
| CN116466724A (en) | Mobile positioning method and device of robot and robot | |
| CN113763432B (en) | Target detection tracking method based on image definition and tracking stability conditions | |
| CN213932635U (en) | Solid waste remote sensing monitoring system based on unmanned aerial vehicle platform | |
| CN209274766U (en) | A kind of mobile platform | |
| Zhao et al. | An RGBD data based vehicle detection algorithm for vehicle following systems | |
| CN113718732A (en) | Unmanned ship on water with active rubbish recovery function | |
| TWI793051B (en) | Monitoring System and Method of Internet of Things for Mushroom Cultivation |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181228 |