CN110162102A - Unmanned plane automatic identification tracking and system based on cloud platform and machine vision - Google Patents
Unmanned plane automatic identification tracking and system based on cloud platform and machine vision Download PDFInfo
- Publication number
- CN110162102A CN110162102A CN201910412746.6A CN201910412746A CN110162102A CN 110162102 A CN110162102 A CN 110162102A CN 201910412746 A CN201910412746 A CN 201910412746A CN 110162102 A CN110162102 A CN 110162102A
- Authority
- CN
- China
- Prior art keywords
- unmanned plane
- tracking
- short distance
- information
- cloud server
- 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
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/12—Target-seeking control
Abstract
The invention discloses unmanned plane automatic identification trackings and system based on cloud platform and machine vision, wherein this method comprises: user terminal APP receives the unmanned plane tracing task of user's input, and is sent to Cloud Server；The flight information of data acquisition device acquisition unmanned plane is simultaneously transmitted to main control chip；Flight information is sent to Cloud Server and user terminal APP by main control chip；Cloud Server handles flight information, obtains avoidance motion control signal and tracks the recognition and tracking signal of target；The recognition and tracking signal of avoidance motion control signal and tracking target is sent to main control chip by Cloud Server；The recognition and tracking signal of avoidance motion control signal and tracking target is transmitted to the flight control panel of unmanned plane by main control chip, is tracked with controlling unmanned plane to target object.Through the invention, it the intelligence and integrated level of unmanned plane automatic identification tracking system are improved, and can be transplanted on common unmanned plane, the widely develop intelligence of UAV system.
Description
Technical field
The present invention relates to unmanned plane automatic identification tracking technique fields, in particular to a kind of to be based on cloud platform and machine vision
Unmanned plane automatic identification tracking and system.
Background technique
Existing unmanned plane is combined by rack fuselage, dynamical system, flight control system, this four part of remote control system.
Wherein second core technology of the no-manned machine distant control system as unmanned plane can be used for agricultural plant protection, electric inspection process, police hold
Method, geological prospecting, environmental monitoring, forest fire protection and video display are taken photo by plane, and cover the fields such as agricultural, security protection, geography, and it is applicable in
Field is also being expanded rapidly.In no-manned machine distant control system, automatic identification tracking system is that it realizes entirely autonomous intelligent control
Important composition.
The autonomous control technology of existing unmanned plane is quite mature perfect, but intelligent control technology is still in research and development
In the stage, in automatic identification tracking system, environment information acquisition, information transmission, information processing etc. all decide the intelligence of unmanned plane
It can property.Minority Intelligent unattended owner will use valiant dragon Flight, Intel's atom processor, Samsung Artik chip and moral at present
State instrument OMAP3630 is as main control chip.These chips all have stronger processing performance, can be used for realizing the intelligence of unmanned plane
Control, but UAV Intelligent is limited to hardware performance, and development cost is high, not convenient for safeguarding.In order to expand each manufacturer's unmanned plane
Intelligence, urgent need design a kind of transplantable unmanned plane automatic identification tracking system based on cloud platform and machine vision, are one
As unmanned plane intelligence provide portable platform.
Summary of the invention
The present invention provides a kind of unmanned plane automatic identification tracking and system based on cloud platform and machine vision, intelligence
Property and integrated level it is high, can also be transplanted on common unmanned plane, the widely develop intelligence of UAV system.
According to an aspect of the invention, there is provided a kind of unmanned plane automatic identification based on cloud platform and machine vision with
Track method, comprising the following steps: user terminal APP receives the unmanned plane tracing task of user's input, and the information that will track target
It is sent to Cloud Server；
The flight information is simultaneously transmitted to main control chip by the flight information of data acquisition device acquisition unmanned plane；The nothing
Man-machine flight information includes around the range information and unmanned plane of the location information of unmanned plane, unmanned plane and surrounding objects
Video image information；
The flight information is sent to the Cloud Server and user terminal APP by the main control chip；
The Cloud Server handles the flight information, obtains avoidance motion control signal and tracks the knowledge of target
It Gen Zong not signal；
The recognition and tracking signal of the avoidance motion control signal and tracking target is sent to described by the Cloud Server
Main control chip；
The recognition and tracking signal of the avoidance motion control signal and tracking target is transmitted to nobody by the main control chip
The flight control panel of machine, is tracked with controlling unmanned plane to target object.
Preferably, the data acquisition device includes laser obstacle avoidance sensor, GPS positioning device and camera, described
Laser obstacle avoidance sensor is for measuring the distance between unmanned plane and periphery object, and the GPS positioning device is for positioning nobody
The position of machine, the camera are used to shoot the video image around unmanned plane.
Preferably, the Cloud Server handles the flight information, obtains avoidance motion control signal and tracking
The recognition and tracking signal of target, comprising the following steps: the Cloud Server analyzes the tracking target that the user terminal APP is sent
Whether information, recognition and tracking target are people；
If it is, the Cloud Server calls Adaboost algorithm to identify people from the video image around unmanned plane
Face；
The Cloud Server will identify that the face come inputs effective faceform that the information of the tracking target includes
In matched, by the highest face of matching degree be determined as track target；
Location information and unmanned plane and surrounding objects the distance between information of the Cloud Server according to unmanned plane, meter
Calculate the location information of the tracking target；
The Cloud Server obtains avoidance movement according to the location information of unmanned plane and the location information of the tracking target
It controls signal and tracks the recognition and tracking signal of target.
Preferably, the Cloud Server sends the recognition and tracking signal of the avoidance motion control signal and tracking target
After to the main control chip, this method is further comprising the steps of: location information and institute of the Cloud Server according to unmanned plane
The location information of tracking target is stated, the tracking section between unmanned plane position and the tracking position of object is calculated；
The tracking section is divided into multiple short distance sections by the Cloud Server；
Cloud Server search in the database of its storage is recorded with the presence or absence of the map in the multiple short distance section；
If it is, the Cloud Server calls the map in the multiple short distance section to record, and by the multiple short distance
The flying height in section is adjusted to the height in the corresponding map record in the multiple short distance section；
The flying height information in each short distance section in the multiple short distance section is sent to institute by the Cloud Server
State main control chip；
The recognition and tracking signal of the avoidance motion control signal and tracking target is transmitted to nobody by the main control chip
The flight control panel of machine, is tracked with controlling unmanned plane to target object, specifically:
The main control chip by the avoidance motion control signal, track the recognition and tracking signal of target and the multiple
The flying height information in each short distance section in short distance section is transmitted to the flight control panel of unmanned plane, to control unmanned plane pair
Target object is tracked.
Preferably, the map in the multiple short distance section is not present in Cloud Server search in the database of its storage
When record, this method is further comprising the steps of: the Cloud Server according to around unmanned plane video image information and nobody
The information of distance between machine and surrounding objects calls SLAM algorithm to establish each short distance section in the multiple short distance section
Two-dimensional map；
It is corresponding that the flying height in the multiple short distance section is adjusted to the multiple short distance section by the Cloud Server
Height in two-dimensional map；
The flying height information in each short distance section in the multiple short distance section is sent to institute by the Cloud Server
State main control chip；
The recognition and tracking signal of the avoidance motion control signal and tracking target is transmitted to nobody by the main control chip
The flight control panel of machine, is tracked with controlling unmanned plane to target object, specifically:
The main control chip by the avoidance motion control signal, track the recognition and tracking signal of target and the multiple
The flying height information in each short distance section in short distance section is transmitted to the flight control panel of unmanned plane, to control unmanned plane pair
Target object is tracked.
According to another aspect of the present invention, a kind of unmanned plane based on cloud platform and machine vision is additionally provided to know automatically
Other tracking system, including user terminal APP, data acquisition device, main control chip and Cloud Server:
Wherein, the user terminal APP includes: tracing task transmission unit, for receiving the unmanned plane tracking of user's input
Task, and the information for tracking target is sent to Cloud Server；
Wherein, the data acquisition device includes: information acquisition unit, for acquiring the flight information of unmanned plane and by institute
It states flight information and is transmitted to main control chip；The flight information of the unmanned plane includes the location information of unmanned plane, unmanned plane and week
Enclose the video image information around the range information and unmanned plane of object；
Wherein, the main control chip includes: flight information transmission unit, for the flight information to be sent to the cloud
Server and user terminal APP；
Wherein, the Cloud Server includes: processing unit, for handling the flight information, obtains avoidance fortune
Dynamic control signal and the recognition and tracking signal for tracking target；Signal transmitting unit, for by the avoidance motion control signal and
The recognition and tracking signal of tracking target is sent to the main control chip；
The main control chip further include: signal transmission unit is used for the avoidance motion control signal and tracking target
Recognition and tracking signal be transmitted to the flight control panel of unmanned plane, tracked with controlling unmanned plane to target object.
Preferably, the data acquisition device includes laser obstacle avoidance sensor, GPS positioning device and camera, described
Laser obstacle avoidance sensor is for measuring the distance between unmanned plane and periphery object, and the GPS positioning device is for positioning nobody
The position of machine, the camera are used to shoot the video image around unmanned plane.
Preferably, the processing unit includes: analysis module, the tracking target sent for analyzing the user terminal APP
Information, whether recognition and tracking target is people；
Face recognition module, for analyzing the information for the tracking target that the user terminal APP is sent when the analysis module,
When recognition and tracking target is people, Adaboost algorithm is called to identify face from the video image around unmanned plane；
Face matching module, for will identify that the face come inputs effective face that the information of the tracking target includes
It is matched in model, the highest face of matching degree is determined as to track target；
Computing module, for the location information and the distance between unmanned plane and surrounding objects information according to unmanned plane,
Calculate the location information of the tracking target；
Signal acquisition module, for being kept away according to the location information of unmanned plane and the location information of the tracking target
Hinder motion control signal and tracks the recognition and tracking signal of target.
Preferably, the Cloud Server further include: computing unit, for the avoidance to be transported in the signal transmitting unit
Dynamic control signal and the recognition and tracking signal for tracking target are sent to after the main control chip, according to the location information of unmanned plane
With the location information of the tracking target, the tracking section between unmanned plane position and the tracking position of object is calculated；
Section cutting unit, for the tracking section to be divided into multiple short distance sections；
Map search unit whether there is the multiple short distance for searching in the database that the Cloud Server stores
The map in section records；
The first adjustment unit, for being deposited when the map search unit is searched in the database that the Cloud Server stores
In the map record in the multiple short distance section, the map in the multiple short distance section is called to record, and will be the multiple short
The flying height in journey section is adjusted to the height in the corresponding map record in the multiple short distance section；
Transmission unit, for the flying height information in each short distance section in the multiple short distance section to be sent to institute
State main control chip；
The signal transmission unit, specifically for believing the recognition and tracking of the avoidance motion control signal, tracking target
Number and the flying height information in each short distance section in the multiple short distance section be transmitted to the flight control panel of unmanned plane,
It is tracked with controlling unmanned plane to target object.
Preferably, the Cloud Server further include: map establishes unit, for when the map search unit is in the cloud
When search is recorded there is no the map in the multiple short distance section in the database of server storage, according to the view around unmanned plane
The information of distance between frequency image information and unmanned plane and surrounding objects calls SLAM algorithm to the multiple short distance section
Two-dimensional map is established in each short distance section；
Second adjustment unit, for the flying height in the multiple short distance section to be adjusted to the multiple short distance section pair
The height in two-dimensional map answered；
The transmission unit, for sending the flying height information in each short distance section in the multiple short distance section
To the main control chip；
The signal transmission unit, specifically for believing the recognition and tracking of the avoidance motion control signal, tracking target
Number and the flying height information in each short distance section in the multiple short distance section be transmitted to the flight control panel of unmanned plane,
It is tracked with controlling unmanned plane to target object.
Compared with prior art, beneficial effects of the present invention are as follows:
Through the invention, the flight information of unmanned plane is acquired using data acquisition device and flight information is transmitted to cloud clothes
Business device is handled, and then obtains tracking signal, is realized and is tracked to the automatic identification of target, data acquisition device of the invention can
Data acquisition is carried out to be transplanted on general unmanned plane, the data processing software in Cloud Server can also be transplanted to general nothing
Data processing is carried out in man-machine server, is not only increased portability and integrated level of the invention in this way, is also greatly mentioned
The high intelligence of general unmanned plane, reduces the production cost of high intelligent unmanned plane, is more convenient to use.
Detailed description of the invention
The drawings described herein are used to provide a further understanding of the present invention, constitutes part of this application, this hair
Bright illustrative embodiments and their description are used to explain the present invention, and are not constituted improper limitations of the present invention.In attached drawing:
Fig. 1 is the unmanned plane automatic identification tracking according to an embodiment of the present invention based on cloud platform and machine vision
Flow chart；
Fig. 2 is the unmanned plane automatic identification tracking system according to an embodiment of the present invention based on cloud platform and machine vision
Structural block diagram；
Fig. 3 be according to embodiments of the present invention one another unmanned plane automatic identification based on cloud platform and machine vision with
The flow chart of track method；
Fig. 4 is the inside principle assumption diagram of the main control chip of the embodiment of the present invention；
Fig. 5 is the circuit diagram of the power boost circuit of the embodiment of the present invention；
Fig. 6 is the circuit diagram of the power voltage step down circuit of the embodiment of the present invention；
Fig. 7 is the flow chart of image recognition processing algorithm in the Cloud Server of the embodiment of the present invention.
Specific embodiment
Below in conjunction with attached drawing of the present invention, technical solution of the present invention is described, but described embodiment is only
A part of the embodiment of the present invention, based on the embodiments of the present invention, those of ordinary skill in the art are not making creative labor
Every other embodiment obtained under the premise of dynamic, shall fall within the protection scope of the present invention.
The embodiment of the invention provides a kind of unmanned plane automatic identification tracking based on cloud platform and machine vision, figure
1 is the flow chart of the unmanned plane automatic identification tracking according to an embodiment of the present invention based on cloud platform and machine vision, such as
Shown in Fig. 1, comprising the following steps:
Step S101: user terminal APP receives the unmanned plane tracing task of user's input, and the information for tracking target is sent
To Cloud Server；
Step S102: flight information is simultaneously transmitted to main control chip by the flight information of data acquisition device acquisition unmanned plane；
Step S103: flight information is sent to Cloud Server and user terminal APP by main control chip；
Step S104: Cloud Server handles flight information, obtains avoidance motion control signal and tracks target
Recognition and tracking signal；
Step S105: the recognition and tracking signal of the avoidance motion control signal and tracking target is sent to by Cloud Server
Main control chip；
Step S106: the recognition and tracking signal of avoidance motion control signal and tracking target is transmitted to nobody by main control chip
The flight control panel of machine, is tracked with controlling unmanned plane to target object.
In implementation process, in step s 102, data acquisition device includes laser obstacle avoidance sensor, GPS positioning device
And camera, laser obstacle avoidance sensor is for measuring the distance between unmanned plane and periphery object, and GPS positioning device is for fixed
The position of position unmanned plane, camera are used to shoot the video image around unmanned plane.
In step S104, the information for the tracking target that the Cloud Server analysis user terminal APP is sent, recognition and tracking mesh
Whether mark is people；If it is, Cloud Server calls Adaboost algorithm to identify face from the video image around unmanned plane；
It matches, will match in effective faceform that the information for the face input tracking target that Cloud Server will identify that includes
Highest face is spent to be determined as tracking target；Cloud Server according to the location information and unmanned plane of unmanned plane and surrounding objects it
Between range information, calculate tracking target location information；Cloud Server is according to the location information of unmanned plane and tracking target
Location information, obtain avoidance motion control signal and track target recognition and tracking signal.
After step S105, Cloud Server is calculated according to the location information of unmanned plane and the location information of tracking target
Tracking section between unmanned plane position and tracking position of object；The tracking section is divided into multiple short distance roads by Cloud Server
Section；Cloud Server search in the database of its storage is recorded with the presence or absence of the map in multiple short distance sections；If it is, cloud takes
Business device calls the map in multiple short distance sections to record, and the flying height in multiple short distance sections is adjusted to multiple short distance sections pair
The height in map record answered；Cloud Server sends the flying height information in each short distance section in multiple short distance sections
To main control chip；
The recognition and tracking signal of avoidance motion control signal and tracking target is transmitted to the flight of unmanned plane by main control chip
Control panel is tracked with controlling unmanned plane to target object, specifically: main control chip by avoidance motion control signal, tracking
The flying height information in each short distance section in the recognition and tracking signal of target and multiple short distance sections is transmitted to unmanned plane
Flight control panel, tracked with controlling unmanned plane to target object.
Further, there is no the maps in multiple short distance sections to record for search in the database of its storage for Cloud Server
When, Cloud Server is adjusted according to the information of distance between the video image information and unmanned plane and surrounding objects around unmanned plane
Two-dimensional map is established with each the short distance section of SLAM algorithm to multiple short distance sections；Cloud Server is by multiple short distance sections
Flying height is adjusted to the height in the corresponding two-dimensional map in multiple short distance sections；Cloud Server is each by multiple short distance sections
The flying height information in a short distance section is sent to main control chip；
The recognition and tracking signal of avoidance motion control signal and tracking target is transmitted to the flight of unmanned plane by main control chip
Control panel is tracked with controlling unmanned plane to target object, specifically: main control chip by avoidance motion control signal, tracking
The flying height information in each short distance section in the recognition and tracking signal of target and multiple short distance sections is transmitted to unmanned plane
Flight control panel, tracked with controlling unmanned plane to target object.
Through the above steps, the intelligence and integrated level of unmanned plane automatic identification tracking system are improved, and can be moved
It plants on common unmanned plane, the widely develop intelligence of UAV system.
The embodiment of the invention also provides the unmanned plane automatic identification tracking systems based on cloud platform and machine vision, are used for
Realize the above-mentioned unmanned plane automatic identification tracking based on cloud platform and machine vision.
Fig. 2 is the unmanned plane automatic identification tracking system according to an embodiment of the present invention based on cloud platform and machine vision
Structural block diagram, as shown in Fig. 2, the system includes user terminal APP10, data acquisition device 20, main control chip 30 and cloud service
Device 40:
Wherein, user terminal APP10 includes: tracing task transmission unit 101, for receiving the unmanned plane tracking of user's input
Task, and the information for tracking target is sent to Cloud Server 40；
Wherein, data acquisition device 20 includes: information acquisition unit 201, for acquiring the flight information of unmanned plane and inciting somebody to action
Flight information is transmitted to main control chip 30；The flight information of unmanned plane includes the location information, unmanned plane and ambient of unmanned plane
The range information and the video image information around unmanned plane of body；
Wherein, main control chip 30 includes: flight information transmission unit 301, for flight information to be sent to Cloud Server
40 and user terminal APP10；
Wherein, Cloud Server 40 includes: processing unit 401, for handling flight information, obtains avoidance movement control
The recognition and tracking signal of signal processed and tracking target；Signal transmitting unit 402 is used for avoidance motion control signal and tracking mesh
Target recognition and tracking signal is sent to main control chip 30；
Main control chip 30 further include: signal transmission unit 302, for by avoidance motion control signal and tracking target knowledge
Not Gen Zong signal be transmitted to the flight control panel of unmanned plane, tracked with controlling unmanned plane to target object.
For the unmanned plane automatic identification tracking system based on cloud platform and machine vision, data acquisition device includes laser
Obstacle avoidance sensor, GPS positioning device and camera, laser obstacle avoidance sensor is for measuring between unmanned plane and periphery object
Distance, GPS positioning device are used to position the position of unmanned plane, and camera is used to shoot the video image around unmanned plane.
For the unmanned plane automatic identification tracking system based on cloud platform and machine vision, processing unit 401 includes: analysis
Module 4011, the information of the tracking target for analyzing user terminal APP10 transmission, whether recognition and tracking target is people；Face is known
Other module 4012, for analyzing the information for the tracking target that user terminal APP10 is sent, recognition and tracking target when analysis module 4011
When for people, Adaboost algorithm is called to identify face from the video image around unmanned plane；Face matching module 4013, is used for
It is matched in effective faceform that the information for the face input tracking target that will identify that includes, matching degree is highest
Face is determined as tracking target；Computing module 4014, for the location information and unmanned plane and surrounding objects according to unmanned plane
The distance between information, calculate tracking target location information；Signal acquisition module 4015, for the position according to unmanned plane
The location information of information and tracking target obtains avoidance motion control signal and tracks the recognition and tracking signal of target.
For the unmanned plane automatic identification tracking system based on cloud platform and machine vision, Cloud Server 40 further include: meter
Unit 403 is calculated, for sending the recognition and tracking signal of avoidance motion control signal and tracking target in signal transmitting unit 402
To main control chip, according to the location information of the location information of unmanned plane and tracking target, unmanned plane position and tracking are calculated
Tracking section between target position；Section cutting unit 404 is divided into multiple short distance sections for that will track section；Map
Search unit 405 is recorded for the map that search whether there is multiple short distance sections in the database that Cloud Server 40 stores；
The first adjustment unit 406, for there are multiple short when map search unit 405 is searched in the database that Cloud Server 40 stores
When the map record in journey section, the map in multiple short distance sections is called to record, and the flying height in multiple short distance sections is adjusted
To the height in the corresponding map record in multiple short distance sections；Transmission unit 407, for by multiple short distance sections each is short
The flying height information in journey section is sent to main control chip 30；Signal transmission unit 302, specifically for believing avoidance motion control
Number, the flying height information in each short distance section in the recognition and tracking signal of tracking target and multiple short distance sections is transmitted to
The flight control panel of unmanned plane, is tracked with controlling unmanned plane to target object.
For the unmanned plane automatic identification tracking system based on cloud platform and machine vision, Cloud Server 40 further include: ground
Figure establishes unit 408, for when map search unit 405 is searched in the database that Cloud Server 40 stores, there is no multiple short
When the map record in journey section, according to distance between the video image information and unmanned plane and surrounding objects around unmanned plane
Information calls SLAM algorithm to establish two-dimensional map to each short distance section in multiple short distance sections；Second adjustment unit 409,
Height for being adjusted to the flying height in multiple short distance sections in the corresponding two-dimensional map in multiple short distance sections；Transmission unit
407, for the flying height information in each short distance section in multiple short distance sections to be sent to main control chip 30；Signal transmission
Unit 302, specifically for by avoidance motion control signal, track target recognition and tracking signal and multiple short distance sections it is every
The flying height information in one short distance section is transmitted to the flight control panel of unmanned plane, is carried out with controlling unmanned plane to target object
Tracking.
It should be noted that unmanned plane automatic identification based on cloud platform and machine vision described in Installation practice with
Track system corresponds to above-mentioned embodiment of the method, and concrete implementation process had carried out specifically in embodiment of the method
Bright, details are not described herein.
In order to keep technical solution of the present invention and implementation method clearer, below in conjunction with preferred embodiment in fact
Existing process is described in detail.
Embodiment one
The present embodiment provides the another kinds unmanned plane automatic identification tracking based on cloud platform and machine vision, such as Fig. 3
Shown, Fig. 3 is that according to embodiments of the present invention one another is tracked based on the unmanned plane automatic identification of cloud platform and machine vision
The flow chart of method, comprising the following steps:
Step S301: user terminal APP receives the unmanned plane tracing task of user's input, and the information for tracking target is sent
To Cloud Server；
In the embodiment of the present invention, before carrying out automatic identification tracking using unmanned plane, user can be first in user terminal APP
Upper input tracing task, and tracing task is sent to Cloud Server, Cloud Server can be analyzed and be adjusted to tracing task
Task is completed with relevant program；
As an alternative embodiment, the mission mode that user inputs on user terminal APP can be personal identification
Tracing mode, in such a mode, after typing target person face information, when unmanned plane detects the target person, Jiu Huixiang
User terminal APP sends the location message of target person, and automatically controls unmanned plane and track the target person；User is in user terminal
The mission mode inputted on APP is also possible to object searching modes, in such a mode, needs the letter of the object to be identified of typing in advance
Breath, during unmanned plane during flying, Cloud Server can carry out identifying processing to collected video image, when there is target object
When, the location information that can send unmanned plane at this time is shown into user terminal APP, while being controlled unmanned plane and being hovered over the object
Position, or carry out the searching of next target object；The mission mode that user inputs on user terminal APP can also be patrol mould
Formula, in such a mode, unmanned plane can be spaced fixed duration and carry out patrol aviation within the specified range, in flight course, cloud clothes
Being engaged in device can be to collected video image progress identifying processing, when detecting dangerous person or hazardous act, Real-time Feedback
To user terminal APP to prompt user to alarm；
Step S302: flight information is simultaneously transmitted to main control chip by the flight information of data acquisition device acquisition unmanned plane；
In the embodiment of the present invention, the flight information of unmanned plane includes the location information of unmanned plane, unmanned plane and surrounding objects
Range information and unmanned plane around video image information；
As an alternative embodiment, above-mentioned data acquisition device includes laser obstacle avoidance sensor, GPS positioning device
And camera, laser obstacle avoidance sensor is for measuring the distance between unmanned plane and periphery object, and GPS positioning device is for fixed
The position of position unmanned plane, camera are used to shoot the video image around unmanned plane；
In the embodiment of the present invention, for above-mentioned main control chip using ESP32-CAM module as core, ESP32-CAM is that peace letter can be public
The camera module of the small size of department supports OV2640 and OV7670 camera, built-in flash that picture WiFI is supported to upload,
UART, SPI, I2C, PWM interface is supported to be extended peripheral hardware, ESP32 integrated chip conventional Bluetooth, low-power consumption bluetooth and Wi-
Fi function directly may be coupled to mutually using ESP32 onboard Wi-Fi configuration IP address and network port number access hotspot
Networking Cloud Server, the onboard bluetooth for directly configuring ESP32 can connect user mobile phone；The inside principle knot of above-mentioned main control chip
Composition is as shown in Figure 4；Above-mentioned main control chip is connected with 4G communication module by serial ports, and 4G communication module selects do4G-
ME3630M02 module, the module use ME3630-C3C chip, support serial communication and 4G the whole network logical, and support GPS and Beidou
The range information of image and laser range sensor is passed to the 4G communication module by serial ports by positioning function, ESP32 module,
Upload to Cloud Server.And the included GPS and Beidou positioning information of the module passes through the module as unmanned plane location information
Be uploaded to Cloud Server, between the module and ESP32 serial ports configuration parameter be 1Mbps, 1 stop position, 8 data bit, without odd even
Check bit；
Laser obstacle avoidance sensor uses VL53L laser range sensor, and VL53L0X is ToF laser ranging module of new generation,
Using the encapsulation of smallest currently on the market, ranging, precision ranging 30- can be carried out based on various object colors and reflection characteristic
2000mm, range accuracy reach ± 3%, and ranging time≤30ms supports I2C bus, ESP32-CAM module to drive using I2C bus
Multiple laser range sensors are moved, unmanned plane surrounding obstacles range information is obtained, the I2C of ESP32 is configured to transmission rate
100kbit/s, 8 data bit；
Camera selects OV2640 camera module, using 1/4 cun of 2,000,000 high definition cmos sensors, has highly sensitive
Degree, high flexibility support the output of JPEG/RGB565 format, and can support exposure, white balance, coloration, saturation degree, comparison
Numerous parameter settings such as degree, ESP32 are connect by DVP interface with OV2640；
Step S303: flight information is sent to Cloud Server and user terminal APP by main control chip；
In the embodiment of the present invention, the range data for the laser range sensor that ESP32-CAM module will acquire, GPS mould
Location data, the image data of camera module of block are uploaded at Cloud Server by onboard WIFI module or 4G module
Reason, the winged control instruction obtained after processing are retransmited to ESP32-CAM module, and winged control instruction is sent to by ESP32-CAM module again
The flight of the winged control plate control unmanned plane of unmanned plane；Specifically, ESP32 main control module obtains figure by the camera module being equipped with
As information, by SCCB protocol-driven camera module, and it is packaged into ros format transmission and stores to Cloud Server to rosbag number
According in packet；ESP32 main control module reads the geographic position data in GPS module by serial ports, and passes through WIFI 4G network
It is uploaded to server, calls Amap API to parse GPS in the server and encodes corresponding unmanned plane geographical location；ESP32
Main control module reads the ranging information of L53L0X laser range sensor by I2C interface, after being uploaded to Cloud Server, for keeping away
Barrier calculates；
As an alternative embodiment, use of the embodiment of the present invention and the independent 3.7V poly-lithium battery of unmanned plane
Power supply exports 5V voltage to ESP32-CAM module and do4G-ME3630M02 module for power supply by booster circuit, and 5V is depressured again
3.3V powers to laser range sensor, and boost chip selects TPS61088RHLR, and decompression chip is selected ASM1117-3.3, risen
Volt circuit schematic diagram is as shown in figure 5, reduction voltage circuit schematic diagram is as shown in Figure 6；
Cell phone application can check system mode by accessing public network IP address and the port of Cloud Server and carry out long-range
Operation；
Step S304: Cloud Server analyze user terminal APP send tracking target information, recognition and tracking target whether be
People；If so, thening follow the steps S305~step S312；
Step S305: Cloud Server calls Adaboost algorithm to identify face from the video image around unmanned plane；
As an alternative embodiment, Cloud Server needs to the video image around the unmanned plane of acquisition
Reason, it is necessary first to solve problem of image blurring, the present embodiment is reduced using least square moving window moving-polynomial smoother method
Problem of image blurring caused by attitude jitter, this method parameter is few, and calculation amount is small, and speed is fast, has good smooth effect,
Then the image processing function in opencv is called to carry out target identification according to user demand pretreated video image again
With Tracing Control；Specific image recognition processing algorithm flow chart is as shown in Figure 7；
Step S306: Cloud Server will identify that effective faceform that the information of the face come input tracking target includes
In matched, by the highest face of matching degree be determined as track target；
Step S307: Cloud Server is believed according to the location information and the distance between unmanned plane and surrounding objects of unmanned plane
Breath calculates the location information of tracking target；
Step S308: Cloud Server obtains avoidance fortune according to the location information of unmanned plane and the location information of tracking target
Dynamic control signal and the recognition and tracking signal for tracking target；
Step S309: the recognition and tracking signal of avoidance motion control signal and tracking target is sent to master control by Cloud Server
Chip；
Step S310: Cloud Server calculates unmanned plane according to the location information of unmanned plane and the location information of tracking target
Tracking section between position and tracking position of object；
Step S311: Cloud Server is divided into multiple short distance sections for section is tracked；
Step S312: Cloud Server search in the database of its storage whether there is the ground seal in multiple short distance sections
Record；If so, thening follow the steps S313, step S316~step S317；If not, thening follow the steps S314~step S317；
Step S313: Cloud Server calls the map in multiple short distance sections to record, and the flight in multiple short distance sections is high
Degree is adjusted to the height in the corresponding map record in multiple short distance sections；
As an alternative embodiment, Cloud Server can adjust the flying height in short distance section one by one, specifically, from
First short distance section starts, and whether Cloud Server is first searched for has the map in this short distance section to record, if so, then by nobody
The flying height of machine is adjusted to the height of map record, records unmanned plane according to the map and flies over this short distance road with flying height
Section is further continued for the map record in second short distance section of search when unmanned plane flies to the terminal in this short distance section, if
Have, then adjusts the flying height of unmanned plane to the height of the map record in this short distance section, and so on, until that will own
Short distance section fly over, the purpose of the tracking target thus reached, but also can accurately track；
Step S314: Cloud Server is according between the video image information and unmanned plane and surrounding objects around unmanned plane
The information of distance calls SLAM algorithm to establish two-dimensional map to each short distance section in multiple short distance sections；
Step S315: the flying height in multiple short distance sections is adjusted to the corresponding two dimension in multiple short distance sections by Cloud Server
Height in map；
Step S316: the flying height information in each short distance section in multiple short distance sections is sent to master by Cloud Server
Control chip；
Step S317: main control chip by avoidance motion control signal, track the recognition and tracking signal of target and multiple short
The flying height information in each short distance section in journey section is transmitted to the flight control panel of unmanned plane, to control unmanned plane to mesh
Mark object is tracked.
In summary, through the foregoing embodiment, the flight information of unmanned plane is acquired using data acquisition device and will flown
Information is transmitted to Cloud Server and is handled, and then obtains tracking signal, realizes and tracks to the automatic identification of target, of the invention
Data acquisition device can be transplanted to progress data acquisition on general unmanned plane, and the data processing software in Cloud Server can also
Data processing is carried out in server to be transplanted to general unmanned plane, not only increases portability of the invention and integrated in this way
Degree, also greatly improves the intelligence of general unmanned plane, reduces the production cost of high intelligent unmanned plane, more convenient to make
With.
Claims (10)
1. a kind of unmanned plane automatic identification tracking based on cloud platform and machine vision, which is characterized in that including following step
It is rapid:
User terminal APP receives the unmanned plane tracing task of user's input, and the information for tracking target is sent to Cloud Server；
The flight information is simultaneously transmitted to main control chip by the flight information of data acquisition device acquisition unmanned plane；The unmanned plane
Flight information include the location information of unmanned plane, the range information of unmanned plane and surrounding objects and the video around unmanned plane
Image information；
The flight information is sent to the Cloud Server and user terminal APP by the main control chip；
The Cloud Server handles the flight information, obtain avoidance motion control signal and track target identification with
Track signal；
The recognition and tracking signal of the avoidance motion control signal and tracking target is sent to the master control by the Cloud Server
Chip；
The recognition and tracking signal of the avoidance motion control signal and tracking target is transmitted to unmanned plane by the main control chip
Flight control panel is tracked with controlling unmanned plane to target object.
2. the method according to claim 1, wherein the data acquisition device include laser obstacle avoidance sensor,
GPS positioning device and camera, the laser obstacle avoidance sensor is for measuring the distance between unmanned plane and periphery object, institute
It states GPS positioning device and is used to shoot the video image around unmanned plane for positioning the position of unmanned plane, the camera.
3. according to the method described in claim 2, it is characterized in that, the Cloud Server handles the flight information,
It obtains avoidance motion control signal and tracks the recognition and tracking signal of target, comprising the following steps:
The Cloud Server analyzes the information for the tracking target that the user terminal APP is sent, and whether recognition and tracking target is people；
If it is, the Cloud Server calls Adaboost algorithm to identify face from the video image around unmanned plane；
The Cloud Server will identify that come face input it is described tracking target information include effective faceform in into
The highest face of matching degree is determined as tracking target by row matching；
The Cloud Server is calculated according to the location information and the distance between unmanned plane and surrounding objects information of unmanned plane
The location information of the tracking target；
The Cloud Server obtains avoidance motion control according to the location information of unmanned plane and the location information of the tracking target
The recognition and tracking signal of signal and tracking target.
4. according to the method described in claim 3, it is characterized in that, the Cloud Server by the avoidance motion control signal and
The recognition and tracking signal of tracking target is sent to after the main control chip, further comprising the steps of:
The Cloud Server according to the location information of the location information of unmanned plane and the tracking target, calculate unmanned plane position with
Tracking section between the tracking position of object；
The tracking section is divided into multiple short distance sections by the Cloud Server；
Cloud Server search in the database of its storage is recorded with the presence or absence of the map in the multiple short distance section；
If it is, the Cloud Server calls the map in the multiple short distance section to record, and by the multiple short distance section
Flying height be adjusted to the height in the corresponding map record of the multiple short distance section；
The flying height information in each short distance section in the multiple short distance section is sent to the master by the Cloud Server
Control chip；
The recognition and tracking signal of the avoidance motion control signal and tracking target is transmitted to unmanned plane by the main control chip
Flight control panel is tracked with controlling unmanned plane to target object, specifically:
The main control chip by the avoidance motion control signal, track target recognition and tracking signal and the multiple short distance
The flying height information in each short distance section in section is transmitted to the flight control panel of unmanned plane, to control unmanned plane to target
Object is tracked.
5. according to the method described in claim 4, it is characterized in that, the Cloud Server is searched for not in the database of its storage
It is further comprising the steps of when being recorded there are the map in the multiple short distance section:
The Cloud Server is according to the letter of distance between the video image information and unmanned plane and surrounding objects around unmanned plane
Breath calls SLAM algorithm to establish two-dimensional map to each short distance section in the multiple short distance section；
The flying height in the multiple short distance section is adjusted to the corresponding two dimension in the multiple short distance section by the Cloud Server
Height in map；
The flying height information in each short distance section in the multiple short distance section is sent to the master by the Cloud Server
Control chip；
The recognition and tracking signal of the avoidance motion control signal and tracking target is transmitted to unmanned plane by the main control chip
Flight control panel is tracked with controlling unmanned plane to target object, specifically:
The main control chip by the avoidance motion control signal, track target recognition and tracking signal and the multiple short distance
The flying height information in each short distance section in section is transmitted to the flight control panel of unmanned plane, to control unmanned plane to target
Object is tracked.
6. a kind of unmanned plane automatic identification tracking system based on cloud platform and machine vision, which is characterized in that including user terminal
APP, data acquisition device, main control chip and Cloud Server:
Wherein, the user terminal APP includes:
Tracing task transmission unit is sent for receiving the unmanned plane tracing task of user's input, and by the information for tracking target
To Cloud Server；
Wherein, the data acquisition device includes:
Information acquisition unit, for acquiring the flight information of unmanned plane and the flight information being transmitted to main control chip；It is described
The flight information of unmanned plane includes around the range information and unmanned plane of the location information of unmanned plane, unmanned plane and surrounding objects
Video image information；
Wherein, the main control chip includes:
Flight information transmission unit, for the flight information to be sent to the Cloud Server and user terminal APP；
Wherein, the Cloud Server includes:
Processing unit, for handling the flight information, obtaining avoidance motion control signal and tracking the identification of target
Track signal；
Signal transmitting unit, it is described for the recognition and tracking signal of the avoidance motion control signal and tracking target to be sent to
Main control chip；
The main control chip further include:
Signal transmission unit, for the recognition and tracking signal of the avoidance motion control signal and tracking target to be transmitted to nobody
The flight control panel of machine, is tracked with controlling unmanned plane to target object.
7. system according to claim 6, which is characterized in that the data acquisition device include laser obstacle avoidance sensor,
GPS positioning device and camera, the laser obstacle avoidance sensor is for measuring the distance between unmanned plane and periphery object, institute
It states GPS positioning device and is used to shoot the video image around unmanned plane for positioning the position of unmanned plane, the camera.
8. system according to claim 7, which is characterized in that the processing unit includes:
Analysis module, for analyzing the information for the tracking target that the user terminal APP is sent, whether recognition and tracking target is people；
Face recognition module, for analyzing the information for the tracking target that the user terminal APP is sent, identification when the analysis module
When tracking target is people, Adaboost algorithm is called to identify face from the video image around unmanned plane；
Face matching module, for will identify that the face come inputs effective faceform that the information of the tracking target includes
In matched, by the highest face of matching degree be determined as track target；
Computing module is calculated for the location information and the distance between unmanned plane and surrounding objects information according to unmanned plane
The location information of the tracking target out；
Signal acquisition module, for obtaining avoidance fortune according to the location information of unmanned plane and the location information of the tracking target
Dynamic control signal and the recognition and tracking signal for tracking target.
9. system according to claim 8, which is characterized in that the Cloud Server further include:
Computing unit, for the signal transmitting unit by the avoidance motion control signal and tracking target recognition and tracking
Signal is sent to after the main control chip, according to the location information of the location information of unmanned plane and the tracking target, is calculated
Tracking section between unmanned plane position and the tracking position of object；
Section cutting unit, for the tracking section to be divided into multiple short distance sections；
Map search unit whether there is the multiple short distance section for searching in the database that the Cloud Server stores
Map record；
The first adjustment unit, for there are institutes when the map search unit is searched in the database that the Cloud Server stores
When stating the map record in multiple short distance sections, the map in the multiple short distance section is called to record, and by the multiple short distance road
The flying height of section is adjusted to the height in the corresponding map record in the multiple short distance section；
Transmission unit, for the flying height information in each short distance section in the multiple short distance section to be sent to the master
Control chip；
The signal transmission unit, specifically for by the avoidance motion control signal, track target recognition and tracking signal with
And the flying height information in each short distance section in the multiple short distance section is transmitted to the flight control panel of unmanned plane, with control
Unmanned plane processed tracks target object.
10. system according to claim 9, which is characterized in that the Cloud Server further include:
Map establishes unit, for being not present when the map search unit is searched in the database that the Cloud Server stores
When the map record in the multiple short distance section, according to the video image information and unmanned plane and surrounding objects around unmanned plane
Between distance information, call SLAM algorithm two-dimensional map is established to each short distance section in the multiple short distance section；
Second adjustment unit, it is corresponding for the flying height in the multiple short distance section to be adjusted to the multiple short distance section
Height in two-dimensional map；
The transmission unit, for the flying height information in each short distance section in the multiple short distance section to be sent to institute
State main control chip；
The signal transmission unit, specifically for by the avoidance motion control signal, track target recognition and tracking signal with
And the flying height information in each short distance section in the multiple short distance section is transmitted to the flight control panel of unmanned plane, with control
Unmanned plane processed tracks target object.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910412746.6A CN110162102A (en) | 2019-05-17 | 2019-05-17 | Unmanned plane automatic identification tracking and system based on cloud platform and machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910412746.6A CN110162102A (en) | 2019-05-17 | 2019-05-17 | Unmanned plane automatic identification tracking and system based on cloud platform and machine vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110162102A true CN110162102A (en) | 2019-08-23 |
Family
ID=67631268
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910412746.6A Pending CN110162102A (en) | 2019-05-17 | 2019-05-17 | Unmanned plane automatic identification tracking and system based on cloud platform and machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110162102A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111103891A (en) * | 2019-12-30 | 2020-05-05 | 西安交通大学 | Unmanned aerial vehicle rapid posture control system and method based on skeleton point detection |
| CN112215091A (en) * | 2020-09-23 | 2021-01-12 | 山东众志军创电子科技有限公司 | Unmanned aerial vehicle based on buffer cooling and face recognition system thereof |
| CN112514374A (en) * | 2020-03-09 | 2021-03-16 | 深圳市大疆创新科技有限公司 | Monitoring system, monitoring method, mobile platform and remote equipment |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105447459A (en) * | 2015-11-18 | 2016-03-30 | 上海海事大学 | Unmanned plane automation detection target and tracking method |
| CN106094875A (en) * | 2016-06-27 | 2016-11-09 | 南京邮电大学 | A kind of target follow-up control method of mobile robot |
| CN206133317U (en) * | 2016-08-11 | 2017-04-26 | 深圳职业技术学院 | Unmanned aerial vehicle automatic tracking target system |
| CN106781707A (en) * | 2016-12-21 | 2017-05-31 | 华北计算技术研究所（中国电子科技集团公司第十五研究所） | A kind of path planning method for low latitude middle and long distance ferry flight |
| CN106972948A (en) * | 2017-02-15 | 2017-07-21 | 广东技术师范学院 | A kind of perception data acquisition methods |
| CN107105051A (en) * | 2017-04-13 | 2017-08-29 | 黄如 | A kind of off-line type aircraft monitoring method and supervising device based on cloud service |
| CN107291100A (en) * | 2017-07-07 | 2017-10-24 | 深圳市华琥技术有限公司 | A kind of monitoring method based on unmanned plane |
| CN107479569A (en) * | 2017-07-05 | 2017-12-15 | 南宁学院 | A kind of unmanned plane automatic flight control system of adjustable spiral wing posture |
| CN108107916A (en) * | 2018-01-31 | 2018-06-01 | 成都草寓科技有限公司 | A kind of Intelligent unattended machine hunting system |
| CN108351649A (en) * | 2015-09-15 | 2018-07-31 | 深圳市大疆创新科技有限公司 | System and method for UAV interactive instructions and control |
| CN108803656A (en) * | 2018-06-12 | 2018-11-13 | 南京航空航天大学 | A kind of flight control method and system based on complicated low latitude |
| CN109445465A (en) * | 2018-10-17 | 2019-03-08 | 深圳市道通智能航空技术有限公司 | Method for tracing, system, unmanned plane and terminal based on unmanned plane |
| CN208760890U (en) * | 2018-08-03 | 2019-04-19 | 广东技术师范学院 | Unmanned plane and rescue aid |
-
2019
- 2019-05-17 CN CN201910412746.6A patent/CN110162102A/en active Pending
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108351649A (en) * | 2015-09-15 | 2018-07-31 | 深圳市大疆创新科技有限公司 | System and method for UAV interactive instructions and control |
| CN105447459A (en) * | 2015-11-18 | 2016-03-30 | 上海海事大学 | Unmanned plane automation detection target and tracking method |
| CN106094875A (en) * | 2016-06-27 | 2016-11-09 | 南京邮电大学 | A kind of target follow-up control method of mobile robot |
| CN206133317U (en) * | 2016-08-11 | 2017-04-26 | 深圳职业技术学院 | Unmanned aerial vehicle automatic tracking target system |
| CN106781707A (en) * | 2016-12-21 | 2017-05-31 | 华北计算技术研究所（中国电子科技集团公司第十五研究所） | A kind of path planning method for low latitude middle and long distance ferry flight |
| CN106972948A (en) * | 2017-02-15 | 2017-07-21 | 广东技术师范学院 | A kind of perception data acquisition methods |
| CN107105051A (en) * | 2017-04-13 | 2017-08-29 | 黄如 | A kind of off-line type aircraft monitoring method and supervising device based on cloud service |
| CN107479569A (en) * | 2017-07-05 | 2017-12-15 | 南宁学院 | A kind of unmanned plane automatic flight control system of adjustable spiral wing posture |
| CN107291100A (en) * | 2017-07-07 | 2017-10-24 | 深圳市华琥技术有限公司 | A kind of monitoring method based on unmanned plane |
| CN108107916A (en) * | 2018-01-31 | 2018-06-01 | 成都草寓科技有限公司 | A kind of Intelligent unattended machine hunting system |
| CN108803656A (en) * | 2018-06-12 | 2018-11-13 | 南京航空航天大学 | A kind of flight control method and system based on complicated low latitude |
| CN208760890U (en) * | 2018-08-03 | 2019-04-19 | 广东技术师范学院 | Unmanned plane and rescue aid |
| CN109445465A (en) * | 2018-10-17 | 2019-03-08 | 深圳市道通智能航空技术有限公司 | Method for tracing, system, unmanned plane and terminal based on unmanned plane |
Non-Patent Citations (2)
| Title |
|---|
| 潘白齐，等: "《泰苑学术 2010年卷》", 30 September 2011 * |
| 陆耀宾等: "反辐射无人机导航路径规划模型研究", 《指挥控制与仿真》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111103891A (en) * | 2019-12-30 | 2020-05-05 | 西安交通大学 | Unmanned aerial vehicle rapid posture control system and method based on skeleton point detection |
| CN111103891B (en) * | 2019-12-30 | 2021-03-16 | 西安交通大学 | Unmanned aerial vehicle rapid posture control system and method based on skeleton point detection |
| CN112514374A (en) * | 2020-03-09 | 2021-03-16 | 深圳市大疆创新科技有限公司 | Monitoring system, monitoring method, mobile platform and remote equipment |
| CN112215091A (en) * | 2020-09-23 | 2021-01-12 | 山东众志军创电子科技有限公司 | Unmanned aerial vehicle based on buffer cooling and face recognition system thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105425815B (en) | A kind of pasture intelligent management system and method using unmanned vehicle | |
| US11151864B2 (en) | System and method for monitoring a property using drone beacons | |
| CN110162102A (en) | Unmanned plane automatic identification tracking and system based on cloud platform and machine vision | |
| WO2020187082A1 (en) | Method for reporting position information by device, and related apparatus | |
| CN106444843B (en) | Unmanned plane relative bearing control method and device | |
| US10514708B2 (en) | Method, apparatus and system for controlling unmanned aerial vehicle | |
| CN112326686B (en) | Unmanned aerial vehicle intelligent cruising pavement disease detection method, unmanned aerial vehicle and detection system | |
| CN205229809U (en) | Utilize unmanned vehicles's pasture intelligent management system and unmanned vehicles thereof | |
| CN106054917A (en) | Unmanned aerial vehicle flight control method and device, and remote controller | |
| JP7179382B2 (en) | Phenotypic information collection system for field plants | |
| CN110370273B (en) | Robot obstacle avoidance method, device and system | |
| CN106210613A (en) | A kind of video frequency monitoring method, Apparatus and system | |
| US11763566B2 (en) | Target association using occlusion analysis, clustering, or both | |
| CN103438869A (en) | Aerial dynamic large-scale vegetation coverage acquisition system | |
| CN107861436A (en) | A kind of multi-rotor unmanned aerial vehicle high altitude environment detecting system | |
| KR101916391B1 (en) | Agricultural Drone | |
| Nahar et al. | Autonomous UAV forced graffiti detection and removal system based on machine learning | |
| WO2018177269A1 (en) | Device and method for performing long-distance information interaction with unmanned aerial vehicle by means of mobile network | |
| CN205942440U (en) | Intelligence business office robot | |
| CN204046707U (en) | A kind of Portable scene camera arrangement | |
| CN111145065A (en) | Artificial intelligence community internet of things service terminal and system | |
| CN207560210U (en) | Home farm's Environmental security prior-warning device | |
| CN203385417U (en) | Large-scale aviation dynamic acquisition system for vegetation coverage | |
| US11978247B2 (en) | Adversarial masks for scene-customized false detection removal | |
| CN110044348A (en) | A kind of three-dimensional indoor navigation system and its implementation |
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 |
Application publication date: 20190823 |
|
| RJ01 | Rejection of invention patent application after publication |