CN108777071A - A kind of highway patrol robot - Google Patents
A kind of highway patrol robot Download PDFInfo
- Publication number
- CN108777071A CN108777071A CN201810726321.8A CN201810726321A CN108777071A CN 108777071 A CN108777071 A CN 108777071A CN 201810726321 A CN201810726321 A CN 201810726321A CN 108777071 A CN108777071 A CN 108777071A
- Authority
- CN
- China
- Prior art keywords
- image
- pixel
- highway
- result
- patrol robot
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/04—Detecting movement of traffic to be counted or controlled using optical or ultrasonic detectors
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
The present invention provides a kind of highway patrol robots, including lane detection device, travel driving unit, photographic device, identification device and alarm device, the lane detection device is for being detected highway lane line, and detection information is sent to travel driving unit, the travel driving unit is for driving patrol robot to be moved, the photographic device is for obtaining real-time surrounding enviroment image, surrounding enviroment image is identified to find risk object in the identification device, the alarm device is for finding signal an alert when risk object, and the alarm signal is sent to travel driving unit.Beneficial effects of the present invention are：The colleges and universities' patrol for realizing highway, largely saves manpower and materials.
Description
Technical field
The present invention relates to robot fields, and in particular to a kind of highway patrol robot.
Background technology
Early in 20th century 20 to the thirties, highway begins to occur in western developed countries such as Italy, Germany.Meaning
Milan has been built to Simon Rex highway conducive to nineteen twenty-four greatly.Germany has built up Bonn to Cologne highway in 1932.
What is then developed is the U.S., Britain, France, Japan and other countries.
Highway is planned, Large scale construction is the main flourishing state in west at this time after last century the mid-50
Family starts to enter sustained and rapid development period from war-time economy state, and trip demand total amount constantly increases, industrial society's life
Multi items, small batch product and the high, precision and frontier product of production increase significantly, and require obviously to carry to convenience, the promptness of transport
It is high.At the same time, auto industry rapidly develops, and Automobile Transportation is increasingly becoming the basic means of transportation in the comprehensive system of transport, this
Direct impetus is played for the development of highway.
With the development of the social economy, the construction of China's highway achieves the achievement to attract people's attention.But high speed is public
Road needs a large amount of manpowers to be gone on patrol.
Invention content
In view of the above-mentioned problems, the present invention is intended to provide a kind of highway patrol robot.
The purpose of the present invention is realized using following technical scheme：
Provide a kind of highway patrol robot, including lane detection device, travel driving unit, camera shooting dress
It sets, identification device and alarm device, the lane detection device will be detected for being detected to highway lane line
Information is sent to travel driving unit, and the travel driving unit is for driving patrol robot to be moved, the camera shooting dress
It sets for obtaining real-time surrounding enviroment image, the identification device is identified surrounding enviroment image to find dangerous mesh
The alarm signal is sent to hoofing part by mark, the alarm device for finding signal an alert when risk object
Device；The travel driving unit includes drive module and locating module, and the locating module is for obtaining patrol robot
Real-time position information, the drive module are used for according to the real-time position information and lane line information-driven patrol robot edge
Highway lane line is moved, and when there is alarm signal, the drive module drives patrol robot to risk object
It is mobile.
Beneficial effects of the present invention are：The colleges and universities' patrol for realizing highway, largely saves manpower and materials.
Description of the drawings
Using attached drawing, the invention will be further described, but the embodiment in attached drawing does not constitute any limit to the present invention
System, for those of ordinary skill in the art, without creative efforts, can also obtain according to the following drawings
Other attached drawings.
Fig. 1 is the structural schematic diagram of the present invention；
Reference numeral：
Lane detection device 1, travel driving unit 2, photographic device 3, identification device 4, alarm device 5.
Specific implementation mode
The invention will be further described with the following Examples.
Referring to Fig. 1, a kind of highway patrol robot of the present embodiment, including lane detection device 1, hoofing part
Device 2, photographic device 3, identification device 4 and alarm device 5, the lane detection device 1 are used for highway lane line
It is detected, and detection information is sent to travel driving unit 2, the travel driving unit 2 is for driving patrol robot
It is moved, the photographic device 3 is for obtaining real-time surrounding enviroment image, and the identification device 4 is to surrounding enviroment image
It is identified to find risk object, the alarm device 5 is for finding signal an alert when risk object, and by the police
The number of notifying is sent to travel driving unit 2；The travel driving unit 2 includes drive module and locating module, the positioning mould
Block is used to obtain the real-time position information of patrol robot, and the drive module is used for according to the real-time position information and track
Line information-driven patrol robot is moved along highway lane line, and when there is alarm signal, the drive module is driven
Dynamic patrol robot is moved to risk object.
The present embodiment realizes colleges and universities' patrol of highway, largely saves manpower and materials.
Preferably, the lane detection device 1 includes first processing module, Second processing module, third processing module
And fourth processing module, the first processing module are used to obtain road image using camera, the Second processing module is used for
Road image is split, the third processing module is used to the road image of segmentation transforming to vertical view from image coordinate system
Map space coordinate system, the fourth processing module is for being detected lane line in vertical view space coordinates.
This preferred embodiment lane detection device by road image is split and image convert, vertical view sky
Between coordinate system lane line is detected, improve the accuracy rate and speed of lane detection.
Preferably, the Second processing module includes single treatment submodule, after-treatment submodule and handles three times sub
Module, the single treatment submodule obtain a segmentation result, after-treatment for once being divided to image
Module is used to carry out secondary splitting to image, obtains secondary splitting as a result, the submodule of processing three times is used for once dividing
As a result it is merged with secondary splitting result, obtains final image segmentation result；
The single treatment submodule obtains a segmentation result, specially for once being divided to image：To road
Road image carries out gray processing processing, obtains gray level image I (x, y)；For gray level image, it is filtered using following formula：In formula, CA (x, y) indicates that filtered gray level image, σ indicate ash
Spend the gray standard deviation of image I (x, y)；For the pixel (x, y) in gray level image, binary conversion treatment is carried out using following formula：In formula, CA (x, y) indicates the gray value of pixel (x, y), q (x, y)
The binaryzation of pixel (x, y) is indicated as a result, PL (x, y) indicates the binary-state threshold of pixel (x, y)；
The binary-state threshold PL (x, y) of pixel (x, y) is determined by following formula：PL (x, y)=lg (G+3)+E, in formula,
E indicates that the average gray of 3 × 3 neighborhood territory pixels of pixel (x, y), G indicate 3 × 3 neighborhood territory pixel gray values of pixel (x, y)
Root mean square；It regard binaryzation result q (x, y) as segmentation result of image；
This preferred embodiment obtains the abundant primary segmentation of image detail by gray processing, filtering and binary conversion treatment
As a result, specifically, determine the binary-state threshold on the location of pixels according to the gray value size cases of the neighborhood of pixel, due to
Binary-state threshold is continually changing, and the high image-region threshold value of brightness can be larger, and the threshold value of the low image-region of brightness compared with
It is small.
Preferably, the after-treatment submodule is used to carry out secondary splitting to image, obtains secondary splitting as a result, specific
For：Gray processing processing is carried out to road image, obtains gray level image I (x, y)；Image border is examined using canny algorithms
It surveys, obtains secondary splitting result TZ (x, y)；The submodule of processing three times is used for a segmentation result and secondary splitting result
It is merged, obtains final image segmentation result, specially：Melted using segmentation result of following formula pair and secondary splitting result
It closes：In formula, KW (x, y) indicates the final segmentation result of image：
This preferred embodiment after-treatment submodule is detected image border by canny algorithms, has obtained image
Secondary splitting result TZ (x, y), overcome illumination, the cloudy color and dirty interference generated to lane line such as burst, handle submodule three times
Block is by merging a segmentation result and secondary splitting result so that final image segmentation result has been provided simultaneously with once
The advantages of segmentation result and secondary splitting result, specifically, a segmentation result obtains good contours extract effect, it is secondary
Segmentation result obtains good edge extracting effect, has filtered out profile noise and edge noise well.
Preferably, the third processing module is used to the road image of segmentation transforming to vertical view sky from image coordinate system
Between coordinate system, specially：Image coordinate system is coordinate system of the image as unit of pixel, and the coordinate (x, y) of pixel represents pixel
Columns in the picture and line number, it is assumed that road is horizontal, then transforms to the pixel of vertical view space coordinates all same
One plane is obtained in the position (u, v) of vertical view space coordinates by following formula：
In formula, H indicates that height of the camera with respect to ground, m indicate that road image line number, n indicate road image columns,
β0Indicate camera tilt angles, θxIndicate vertical camera half-angle, θyIndicate level camera half-angle；
When camera shoots track, camera optical axis and road there are angle, this preferred embodiment third processing module pass through by
Road image transforms to vertical view space coordinates, can more intuitively express lane line information, convenient for subsequently to lane line
It is detected.
Highway patrol robot of the present invention chooses 5 highways and is tested, respectively in highway patrol
Highway 1, highway 2, highway 3, highway 4, highway 5 unite to patrol efficiency and patrol cost
Meter, compared with personnel go on patrol, generation has the beneficial effect that shown in table：
| Efficiency is gone on patrol to improve | Go on patrol cost reduction | |
| Highway 1 | 29% | 27% |
| Highway 2 | 27% | 26% |
| Highway 3 | 26% | 26% |
| Highway 4 | 25% | 24% |
| Highway 5 | 24% | 22% |
Finally it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than the present invention is protected
The limitation of range is protected, although being explained in detail to the present invention with reference to preferred embodiment, those skilled in the art answer
Work as understanding, technical scheme of the present invention can be modified or replaced equivalently, without departing from the reality of technical solution of the present invention
Matter and range.
Claims (6)
1. a kind of highway patrol robot, which is characterized in that including lane detection device, travel driving unit, camera shooting
Device, identification device and alarm device, the lane detection device will be examined for being detected to highway lane line
Measurement information is sent to travel driving unit, and the travel driving unit is for driving patrol robot to be moved, the camera shooting
For obtaining real-time surrounding enviroment image, the identification device is identified surrounding enviroment image to find dangerous mesh device
The alarm signal is sent to hoofing part by mark, the alarm device for finding signal an alert when risk object
Device；The travel driving unit includes drive module and locating module, and the locating module is for obtaining patrol robot
Real-time position information, the drive module are used for according to the real-time position information and lane line information-driven patrol robot edge
Highway lane line is moved, and when there is alarm signal, the drive module drives patrol robot to risk object
It is mobile.
2. highway patrol robot according to claim 1, which is characterized in that the lane detection device includes
First processing module, Second processing module, third processing module and fourth processing module, the first processing module is for using
Camera obtains road image, and for being split to road image, the third processing module is used for the Second processing module
The road image of segmentation is transformed into vertical view space coordinates from image coordinate system, the fourth processing module is for overlooking
Map space coordinate system is detected lane line.
3. highway patrol robot according to claim 2, which is characterized in that the Second processing module includes one
Secondary processing submodule, after-treatment submodule and submodule is handled three times, the single treatment submodule is used to carry out image
Primary segmentation obtains a segmentation result, and the after-treatment submodule is used to carry out secondary splitting to image, obtains secondary point
It cuts as a result, the processing submodule three times obtains final figure for being merged to a segmentation result and secondary splitting result
As segmentation result.
4. highway patrol robot according to claim 3, which is characterized in that the single treatment submodule is used for
Image is once divided, obtains a segmentation result, specially：Gray processing processing is carried out to road image, obtains gray scale
Image I (x, y)；For gray level image, it is filtered using following formula：
In formula, CA (x, y) indicates that filtered gray level image, σ indicate the gray standard deviation of gray level image I (x, y)；For gray scale
Pixel (x, y) in image carries out binary conversion treatment using following formula：In formula
In son, CA (x, y) indicates the gray value of pixel (x, y), and q (x, y) indicates the binaryzation of pixel (x, y) as a result, PL (x, y) is indicated
The binary-state threshold of pixel (x, y)；
The binary-state threshold PL (x, y) of pixel (x, y) is determined by following formula：PL (x, y)=lg (G+3)+E, in formula, E tables
Show that the average gray of 3 × 3 neighborhood territory pixels of pixel (x, y), G indicate the equal of 3 × 3 neighborhood territory pixel gray values of pixel (x, y)
Root；It regard binaryzation result q (x, y) as segmentation result of image.
5. highway patrol robot according to claim 4, which is characterized in that the after-treatment submodule is used for
Secondary splitting is carried out to image, obtains secondary splitting as a result, being specially：Gray processing processing is carried out to road image, obtains gray scale
Image I (x, y)；Image border is detected using canny algorithms, obtains secondary splitting result TZ (x, y)；It is described to locate three times
Reason submodule obtains final image segmentation result, specially for being merged to a segmentation result and secondary splitting result：
It is merged using segmentation result of following formula pair and secondary splitting result：In formula
In, KW (x, y) indicates the final segmentation result of image.
6. highway patrol robot according to claim 5, which is characterized in that the third processing module is used for will
The road image of segmentation transforms to vertical view space coordinates from image coordinate system, specially：Image coordinate system is image with picture
Element is the coordinate system of unit, and the coordinate (x, y) of pixel represents pixel columns in the picture and line number, it is assumed that road is horizontal
, then the pixel of vertical view space coordinates is transformed to all in same plane, in the position (u, v) of vertical view space coordinates
It is obtained by following formula：
In formula, H indicates that height of the camera with respect to ground, m indicate that road image line number, n indicate road image columns, β0Table
Show camera tilt angles, θxIndicate vertical camera half-angle, θyIndicate level camera half-angle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810726321.8A CN108777071A (en) | 2018-07-04 | 2018-07-04 | A kind of highway patrol robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810726321.8A CN108777071A (en) | 2018-07-04 | 2018-07-04 | A kind of highway patrol robot |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108777071A true CN108777071A (en) | 2018-11-09 |
Family
ID=64031001
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810726321.8A Withdrawn CN108777071A (en) | 2018-07-04 | 2018-07-04 | A kind of highway patrol robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108777071A (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2850608B2 (en) * | 1991-11-28 | 1999-01-27 | 日産自動車株式会社 | Roadway detection device for vehicles |
| KR20120016461A (en) * | 2010-08-16 | 2012-02-24 | 주식회사 이미지넥스트 | Road Marker Recognition System and Method |
| CN102682292A (en) * | 2012-05-10 | 2012-09-19 | 清华大学 | Method based on monocular vision for detecting and roughly positioning edge of road |
| CN103177246A (en) * | 2013-03-26 | 2013-06-26 | 北京理工大学 | Dual-model lane line identification method based on dynamic area division |
| CN103954275A (en) * | 2014-04-01 | 2014-07-30 | 西安交通大学 | Lane line detection and GIS map information development-based vision navigation method |
| CN203870474U (en) * | 2014-04-08 | 2014-10-08 | 上海好创机电工程有限公司 | Automatic navigation patrol robot for visual monitoring |
| CN105678285A (en) * | 2016-02-18 | 2016-06-15 | 北京大学深圳研究生院 | Adaptive road aerial view transformation method and road lane detection method |
| CN106341661A (en) * | 2016-09-13 | 2017-01-18 | 深圳市大道智创科技有限公司 | Patrol robot |
| CN107562054A (en) * | 2017-08-31 | 2018-01-09 | 深圳波比机器人科技有限公司 | The independent navigation robot of view-based access control model, RFID, IMU and odometer |
-
2018
- 2018-07-04 CN CN201810726321.8A patent/CN108777071A/en not_active Withdrawn
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2850608B2 (en) * | 1991-11-28 | 1999-01-27 | 日産自動車株式会社 | Roadway detection device for vehicles |
| KR20120016461A (en) * | 2010-08-16 | 2012-02-24 | 주식회사 이미지넥스트 | Road Marker Recognition System and Method |
| CN102682292A (en) * | 2012-05-10 | 2012-09-19 | 清华大学 | Method based on monocular vision for detecting and roughly positioning edge of road |
| CN103177246A (en) * | 2013-03-26 | 2013-06-26 | 北京理工大学 | Dual-model lane line identification method based on dynamic area division |
| CN103954275A (en) * | 2014-04-01 | 2014-07-30 | 西安交通大学 | Lane line detection and GIS map information development-based vision navigation method |
| CN203870474U (en) * | 2014-04-08 | 2014-10-08 | 上海好创机电工程有限公司 | Automatic navigation patrol robot for visual monitoring |
| CN105678285A (en) * | 2016-02-18 | 2016-06-15 | 北京大学深圳研究生院 | Adaptive road aerial view transformation method and road lane detection method |
| CN106341661A (en) * | 2016-09-13 | 2017-01-18 | 深圳市大道智创科技有限公司 | Patrol robot |
| CN107562054A (en) * | 2017-08-31 | 2018-01-09 | 深圳波比机器人科技有限公司 | The independent navigation robot of view-based access control model, RFID, IMU and odometer |
Non-Patent Citations (1)
| Title |
|---|
| 王俊: "无人驾驶车辆环境感知系统关键技术研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106651953B (en) | A Vehicle Pose Estimation Method Based on Traffic Signs | |
| CN102682292B (en) | Road edge detection and rough positioning method based on monocular vision | |
| CN109872329A (en) | A fast segmentation method of ground point cloud based on 3D lidar | |
| CN104236498A (en) | Method for measuring end size of switch rail | |
| CN104156731A (en) | License plate recognition system based on artificial neural network and method | |
| CN106560835A (en) | Guideboard identification method and device | |
| CN1202239A (en) | Computer stereo image system and method | |
| CN100370226C (en) | Visual Navigation Method for Artificial Landmarks | |
| Busch et al. | Lumpi: The leibniz university multi-perspective intersection dataset | |
| CN101510311A (en) | Method for rapidly sorting a large amount of building side elevation images based on GPS information | |
| CN110850394A (en) | Method for calibrating intensity of automatic driving laser radar | |
| CN103150550B (en) | A kind of road pedestrian event detection method based on gripper path analysis | |
| CN115424263A (en) | Traffic light labeling method, vehicle-mounted device and storage medium | |
| CN107918775B (en) | Zebra crossing detection method and system for assisting safe driving of vehicle | |
| Liu et al. | STCN-Net: A novel multi-feature stream fusion visibility estimation approach | |
| CN110415299B (en) | Vehicle position estimation method based on set guideboard under motion constraint | |
| Wang et al. | A new method for obstacle detection based on Kinect depth image | |
| CN107861510A (en) | A kind of intelligent vehicle control loop | |
| CN108777071A (en) | A kind of highway patrol robot | |
| CN101359403B (en) | Method for extracting contour outline of buildings from satellite imagery | |
| Philipsen et al. | Ongoing work on traffic lights: Detection and evaluation | |
| CN102364560A (en) | A traffic sign convenient for electronic identification and its identification method | |
| WO2022100960A1 (en) | Method for providing navigation data in order to control a robot, method for controlling a robot, method for producing at least one predefined point-symmetric region, and device | |
| CN108628320A (en) | A kind of intelligent automobile Unmanned Systems | |
| CN117726880A (en) | A monocular camera-based 3D real-time detection method, system, equipment and medium for traffic cones |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20181109 |