CN108174111A - Crusing robot target image grasping means - Google Patents

Crusing robot target image grasping means Download PDF

Info

Publication number
CN108174111A
CN108174111A CN201810354614.8A CN201810354614A CN108174111A CN 108174111 A CN108174111 A CN 108174111A CN 201810354614 A CN201810354614 A CN 201810354614A CN 108174111 A CN108174111 A CN 108174111A
Authority
CN
China
Prior art keywords
crusing robot
image
target
reference picture
inspection
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201810354614.8A
Other languages
Chinese (zh)
Other versions
CN108174111B (en
Inventor
庄孟文
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
CHANGZHOU YINGNENG ELECTRIC CO LTD
Original Assignee
CHANGZHOU YINGNENG ELECTRIC CO LTD
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by CHANGZHOU YINGNENG ELECTRIC CO LTD filed Critical CHANGZHOU YINGNENG ELECTRIC CO LTD
Priority to CN201810354614.8A priority Critical patent/CN108174111B/en
Publication of CN108174111A publication Critical patent/CN108174111A/en
Application granted granted Critical
Publication of CN108174111B publication Critical patent/CN108174111B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60Control of cameras or camera modules
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/20Image preprocessing
    • G06V10/255Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00Cameras or camera modules comprising electronic image sensors; Control thereof
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00Television systems
    • H04N7/18Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • H04N7/183Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a single remote source
    • H04N7/185Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a single remote source from a mobile camera, e.g. for remote control

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Processing (AREA)
  • Manipulator (AREA)

Abstract

The present invention relates to a kind of crusing robot target image grasping means, including:1)Crusing robot runs to assigned target position, the target reference picture of photographic subjects object;2)When crusing robot moves inspection, it is continuously shot inspection image;3)Crusing robot compares and analyzes inspection image with target reference picture, if inspection image is overlapped with target reference picture, crusing robot is using the inspection image as target image.The present invention can solve crusing robot be completely dependent on travel distance positioning when, due to wheel slip, wear the problem of positioning accuracy occurred is poor, it can realize picture of the extraction available for instrumental analysis from the multiple image of lasting camera shooting, meet crusing robot image identification positioning accuracy request;It can be achieved to shoot without pausing and without periodic calibration, reduce monitoring time, improve routing inspection efficiency, reduce the workload of operation maintenance personnel.

Description

Crusing robot target image grasping means
Technical field
The present invention relates to robot field, especially a kind of crusing robot target image grasping means.
Background technology
Interior distribution station is as power grid end, and large number of, inspection pressure is huge, and the appearance of rail mounted crusing robot is big The big inspection problem for alleviating interior distribution station.In order to carry out inspection to indoor electric equipment, it usually needs preset inspection Target formulates inspection route, setting patrol task.When crusing robot performs patrol task, run to and patrol according to inspection route Target location is examined, by included shooting unit photographic subjects object after stopping, is then proceeded to next inspection target location It moves, in motion process, robot carries out automatic identification to the target shot before without picture, reads and backs up meter therein Numerical value or position of the switch information.
However it is this stop the mode taken pictures in object corresponding position, routing inspection efficiency is relatively low;And since machinery is beaten Sliding, abrasion or movement cumulative errors, can cause stop position to generate offset, and then influence shooting effect or even influence to patrol Examine the normal inspection of robot.
Invention content
The technical problem to be solved by the present invention is to:A kind of crusing robot target image grasping means is proposed, to solve to patrol Inspection robot be completely dependent on travel distance positioning when, due to wheel slip, abrasion occur positioning accuracy it is poor the problem of, it can be achieved that It is shot without pausing and without periodic calibration, reduces monitoring time, improves routing inspection efficiency, reduce the workload of operation maintenance personnel.
The technical solution adopted in the present invention is:A kind of crusing robot target image grasping means, includes the following steps:
1)Crusing robot runs to assigned target position, the target reference picture of photographic subjects object;
2)When crusing robot moves inspection, it is continuously shot inspection image;
3)Crusing robot compares and analyzes inspection image with target reference picture, if inspection image and object reference figure As overlapping, then crusing robot is using the inspection image as target image.
Step 3 of the present invention)In, inspection image is overlapped with target reference picture as inspection image and object reference figure The key feature object of picture overlaps, and the key feature object includes cabinet in, cabinet angle, artificial target, instrument sideline or angle, nameplate Line or angle, centre of figure point or focus point.
Further, the acquisition methods of key feature object of the present invention include:
A, tri- single channel figures of RGB corresponding to picture are obtained;
B, according to different device types, corresponding single channel figure is chosen;
C, according to different device types, corresponding thresholding is chosen, thresholding extraction then is carried out to single channel figure, thresholding is obtained and carries Take figure;
D, maximum connected domain is obtained from thresholding extraction figure;
E, the connected domain is filled;
F, the connected domain profile of filling is obtained, and then obtains the key feature object of profile.
Step 1 of the present invention)In, crusing robot is during assigned target position is run to, according to fixed intervals Shooting walking reference picture;The fixed intervals include Fixed Time Interval or fixed range interval;Walking reference picture and mesh Mark reference picture establishes path image arrangement according to the sequencing of photo opporunity.
Step 2 of the present invention)In, when crusing robot carries out mobile inspection, inspection image is continuously shot, the inspection figure As not only with target reference picture comparative analysis, also with walk reference picture comparative analysis, if inspection image with walking image It overlaps, then the crusing robot reaches the walking image position in path image arrangement.
When crusing robot determines the position currently in path image arrangement, crusing robot is from assigned target position When farther out, the higher speed of travel is taken;When crusing robot is nearer from assigned target position, slower walking speed is taken Degree.
Crusing robot is nearer from assigned target position, and crusing robot is reached from current location according to the current speed of travel The time of assigned target position carries out the identification of inspection image, feedback and the time acted, but be no more than and patrol more than crusing robot It examines twice that robot carries out the time of the identification of inspection image, feedback and action;Crusing robot from assigned target position farther out, Crusing robot reaches the time of assigned target position from current location according to the current speed of travel, is carried out more than crusing robot The identification of inspection image, feed back and act more than twice of time.
Crusing robot carries out the identification of inspection image, feedback and the time acted, identifies, feeds back and moves including 2 images The time of work.
When crusing robot of the present invention is reached before and after assigned target position, crusing robot under current movement speed, Shooting speed can make in the inspection image being continuously shot, and at least one overlaps with target reference picture.
The beneficial effects of the invention are as follows:Inspection is carried out using the present invention, is realized by lasting camera shooting and is shot without pausing, is expert at Captured picture is compared and analyzed after walking road, homeward journey even recurrence origin, so as to reduce monitoring time, improve and patrol Examine efficiency;Equally, it positions and calibrates without the front and rear phase, reduce the workload of operation maintenance personnel, reduce system complexity;It can be real Now picture of the extraction available for instrumental analysis from the multiple image of lasting camera shooting meets crusing robot image identification positioning accurate Degree requirement.
Description of the drawings
The present invention is further described with reference to the accompanying drawings and examples.
Fig. 1 is the flow diagram of the present invention.
Fig. 2 is the key feature object acquisition methods flow diagram of the present invention.
Fig. 3 is original image(Reference picture).
Fig. 4 is original image(Inspection picture).
Fig. 5 is the passage layered figures of R(Reference picture).
Fig. 6 is thresholding extraction figure(Reference picture R channels).
Fig. 7 is largest connected domain figure(Reference picture R channels).
Fig. 8 is connected domain blank map(Reference picture R channels).
Fig. 9 is connected domain profile diagram(Reference picture R channels).
Figure 10 is profile key feature object figure(Reference picture R channels).
Figure 11 is center point coordinate calculating figure(Reference picture).
Figure 12 is center point coordinate calculating figure(Inspection picture).
Specific embodiment
Presently in connection with attached drawing and preferred embodiment, the present invention is described in further detail.These attached drawings are simplified Schematic diagram only illustrates the basic structure of the present invention, therefore it only shows composition related to the present invention in a schematic way.
A kind of crusing robot target image grasping means as shown in Figure 1, including,
1st, crusing robot runs to assigned target position, the target reference picture of photographic subjects object;Crusing robot is being walked To assigned target position, walking reference picture is shot according to fixed intervals.The fixed intervals were included between the set time Every or fixed range interval.Walking reference picture and target reference picture establish path image according to the sequencing of photo opporunity Arrangement.Crusing robot is nearer from assigned target position, refers to that crusing robot reaches predeterminated target according to the current speed of travel The time of position carries out the identification of inspection image, feedback and the time acted, but no more than crusing robot more than crusing robot Carry out twice of the time of the identification of inspection image, feedback and action.
2nd, when crusing robot carries out mobile inspection, inspection image is continuously shot, if inspection image and walking image weight It closes, then the crusing robot reaches the walking image position in path image arrangement.Crusing robot reaches predeterminated target When before and after position, under current movement speed, shooting speed can make in the inspection image being continuously shot crusing robot, until Rare one overlaps with target reference picture.
3rd, crusing robot compares and analyzes inspection image and target reference picture, calculates object in inspection image Coordinate position of the key feature object in inspection image.Compare the key feature object of object in inspection image in inspection image In coordinate position and target reference picture in same object coordinate of the same a key feature object in target reference picture Whether position overlaps.If overlapped, show that inspection image is overlapped with target reference picture, then crusing robot is by the inspection figure As being used as target image.
Shown in Fig. 3 is that crusing robot runs to assigned target position, the target reference picture of the object taken; Crusing robot can shoot walking reference picture during assigned target position is run to according to fixed intervals.Shooting row The purpose for walking reference picture is to determine optimal target location;So as to shoot best target reference picture;Meanwhile it walks Reference picture and target reference picture can establish path image arrangement according to the sequencing of photo opporunity;And establish path image The purpose of arrangement is to judge distance of the crusing robot from assigned target position;As crusing robot runs to predeterminated target During position, according to 1 meter apart from multiple walking reference pictures of interval shooting;It is assumed that optimum photographing position, i.e., predetermined mesh Cursor position is apart from the position of 10 meters of far point, then the 10th walking reference picture is target reference picture.
When crusing robot moves inspection, inspection image is continuously shot, inspection image is not only compared with target reference picture Analysis, also with walk reference picture comparative analysis.It can determine crusing robot from predeterminated target with the comparison of walking reference picture The distance of position.When crusing robot determines the position currently in path image arrangement, crusing robot is from predeterminated target Position farther out when, take the higher speed of travel;When crusing robot is nearer from assigned target position, slower walking is taken Speed.
Inspection image is as shown in Figure 4;Due to the object of shooting(That is equipment)Background color differ, therefore, according to Different device types chooses clarity and the passage layered figures of the highest R of contrast, as shown in Figure 5;Due to view be only it is black White color system, therefore applicant does the virtualization processing of image to illustrate herein.
As shown in Fig. 2, the method for obtaining the key feature object of object in inspection image is as follows:
I. tri- single channel figures of RGB corresponding to picture are obtained;
II. according to different device types, corresponding single channel figure is chosen;
III. according to different device types, corresponding thresholding is chosen, thresholding extraction then is carried out to single channel figure, obtains thresholding Extraction figure;As shown in Figure 6;
IV. maximum connected domain is obtained from thresholding extraction figure;As shown in Figure 7;
V. the connected domain is filled;As shown in Figure 8;
VI. the connected domain profile of filling is obtained, as shown in Figure 9;And then the key feature object of profile, i.e. centre of figure point are obtained, As shown in Figure 10.
Figure 11 is the center point coordinate calculating figure of reference picture;Figure 12 is inspection center picture point coordinates calculating figure.
The specific embodiment of the only present invention described in description above, various illustrations are not to the reality of the present invention Matter Composition of contents limits, and person of an ordinary skill in the technical field can be to described in the past specific after specification has been read Embodiment is made an amendment or is deformed, without departing from the spirit and scope of the invention.

Claims (7)

1. a kind of crusing robot target image grasping means, it is characterised in that include the following steps:
1)Crusing robot runs to assigned target position, the target reference picture of photographic subjects object;
2)When crusing robot moves inspection, it is continuously shot inspection image;
3)Crusing robot compares and analyzes inspection image with target reference picture, if inspection image and object reference figure As overlapping, then crusing robot is using the inspection image as target image;
The step 3)In, inspection image overlaps special for the key of inspection image and target reference picture with target reference picture It levies object to overlap, the key feature object includes cabinet side, cabinet angle, artificial target, instrument sideline or angle, nameplate sideline or angle, figure Shape central point or focus point;
The acquisition methods of the key feature object include:
A, tri- single channel figures of RGB corresponding to picture are obtained;
B, corresponding single channel figure is chosen;It is required that selection clarity and that highest channel of contrast;
C, corresponding thresholding is chosen, thresholding extraction then is carried out to single channel figure, obtains thresholding extraction figure;
D, maximum connected domain is obtained from thresholding extraction figure;
E, the connected domain is filled;
F, the connected domain profile of filling is obtained, and then obtains the key feature object of profile.
2. crusing robot target image grasping means as described in claim 1, it is characterised in that:The step 1)In, it patrols Robot is examined during assigned target position is run to, walking reference picture is shot according to fixed intervals;The fixed intervals Including Fixed Time Interval or fixed range interval;Walking reference picture and target reference picture are suitable according to the priority of photo opporunity Sequence establishes path image arrangement.
3. crusing robot target image grasping means as claimed in claim 2, it is characterised in that:The crusing robot into During row movement inspection, inspection image is continuously shot, which not only with target reference picture comparative analysis, also joins with walking Image comparison analysis is examined, if inspection image is overlapped with walking reference picture, which reaches path image arrangement In the walking reference picture position.
4. crusing robot target image grasping means as claimed in claim 3, it is characterised in that:When crusing robot determines The currently position in path image arrangement, crusing robot take higher walking speed when from assigned target position farther out Degree;When crusing robot is nearer from assigned target position, the slower speed of travel is taken.
5. crusing robot target image grasping means as claimed in claim 4, it is characterised in that:Crusing robot is from predetermined Target location is nearer, and crusing robot reaches the time of assigned target position from current location according to the current speed of travel, is more than Crusing robot carries out the identification of inspection image, feedback and the time acted, but carries out inspection image knowledge no more than crusing robot Not, twice of the time fed back and acted;Crusing robot from assigned target position farther out, from current location press by crusing robot The time of assigned target position is reached according to the current speed of travel, the identification of inspection image is carried out more than crusing robot, feeds back and move More than twice of the time of work.
6. crusing robot target image grasping means as claimed in claim 5, it is characterised in that:Crusing robot is patrolled The identification of inspection image, the time fed back and acted, including 2 image identifications, the times fed back and acted.
7. crusing robot target image grasping means as described in claim 1, it is characterised in that:Crusing robot reaches pre- When setting the goal before and after position, for crusing robot under current movement speed, shooting speed can make the inspection figure being continuously shot As in, at least one overlaps with target reference picture.
CN201810354614.8A 2018-04-19 2018-04-19 Crusing robot target image grasping means Active CN108174111B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810354614.8A CN108174111B (en) 2018-04-19 2018-04-19 Crusing robot target image grasping means

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810354614.8A CN108174111B (en) 2018-04-19 2018-04-19 Crusing robot target image grasping means

Publications (2)

Publication Number Publication Date
CN108174111A true CN108174111A (en) 2018-06-15
CN108174111B CN108174111B (en) 2018-07-20

Family

ID=62511411

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810354614.8A Active CN108174111B (en) 2018-04-19 2018-04-19 Crusing robot target image grasping means

Country Status (1)

Country Link
CN (1) CN108174111B (en)

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110248155A (en) * 2019-05-16 2019-09-17 普瑞特机械制造股份有限公司 A kind of tank deck job management method and system
CN110516522A (en) * 2019-05-30 2019-11-29 中通服创立信息科技有限责任公司 A kind of method for inspecting and system
CN111176305A (en) * 2020-04-14 2020-05-19 常州市盈能电气有限公司 Visual navigation method
CN111324131A (en) * 2020-03-31 2020-06-23 中通服创立信息科技有限责任公司 Following monitoring method of track type inspection robot based on human body radar
CN111860242A (en) * 2020-07-07 2020-10-30 北京海益同展信息科技有限公司 Robot inspection method and device and computer readable medium
CN114407005A (en) * 2021-12-02 2022-04-29 国能铁路装备有限责任公司 Robot and walking control method and device thereof

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101447029A (en) * 2008-12-11 2009-06-03 赵怀志 Method for automatically indentifying structures in front of roads
US20110196535A1 (en) * 2010-02-10 2011-08-11 Electric Power Research Institute, Inc. Line inspection robot and system
CN103927507A (en) * 2013-01-12 2014-07-16 山东鲁能智能技术有限公司 Improved multi-instrument reading identification method of transformer station inspection robot
CN104809732A (en) * 2015-05-07 2015-07-29 山东鲁能智能技术有限公司 Electrical equipment appearance abnormity detection method based on image comparison
CN106127180A (en) * 2016-06-30 2016-11-16 广东电网有限责任公司电力科学研究院 A kind of robot assisted localization method and device

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101447029A (en) * 2008-12-11 2009-06-03 赵怀志 Method for automatically indentifying structures in front of roads
US20110196535A1 (en) * 2010-02-10 2011-08-11 Electric Power Research Institute, Inc. Line inspection robot and system
CN103927507A (en) * 2013-01-12 2014-07-16 山东鲁能智能技术有限公司 Improved multi-instrument reading identification method of transformer station inspection robot
CN104809732A (en) * 2015-05-07 2015-07-29 山东鲁能智能技术有限公司 Electrical equipment appearance abnormity detection method based on image comparison
CN106127180A (en) * 2016-06-30 2016-11-16 广东电网有限责任公司电力科学研究院 A kind of robot assisted localization method and device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
杨旭东等: "变电站巡检机器人研究现状综述", 《山东电力技术》 *

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110248155A (en) * 2019-05-16 2019-09-17 普瑞特机械制造股份有限公司 A kind of tank deck job management method and system
CN110516522A (en) * 2019-05-30 2019-11-29 中通服创立信息科技有限责任公司 A kind of method for inspecting and system
CN110516522B (en) * 2019-05-30 2020-11-27 中通服创立信息科技有限责任公司 Inspection method and system
CN111324131A (en) * 2020-03-31 2020-06-23 中通服创立信息科技有限责任公司 Following monitoring method of track type inspection robot based on human body radar
CN111324131B (en) * 2020-03-31 2023-09-01 中通服创立信息科技有限责任公司 Tracking monitoring method of track type inspection robot based on human body radar
CN111176305A (en) * 2020-04-14 2020-05-19 常州市盈能电气有限公司 Visual navigation method
CN111860242A (en) * 2020-07-07 2020-10-30 北京海益同展信息科技有限公司 Robot inspection method and device and computer readable medium
CN111860242B (en) * 2020-07-07 2024-05-21 京东科技信息技术有限公司 Robot inspection method, device and computer readable medium
CN114407005A (en) * 2021-12-02 2022-04-29 国能铁路装备有限责任公司 Robot and walking control method and device thereof

Also Published As

Publication number Publication date
CN108174111B (en) 2018-07-20

Similar Documents

Publication Publication Date Title
CN108174111B (en) Crusing robot target image grasping means
CN110487816A (en) A kind of tunnel inspection image collecting device and tunnel cruising inspection system
CN105203046B (en) Multi-thread array laser 3 D scanning system and multi-thread array laser 3-D scanning method
CN106125744A (en) The Intelligent Mobile Robot cloud platform control method of view-based access control model servo
CN1238691C (en) Combined stereovision, color 3D digitizing and motion capture system
CN109066861A (en) Intelligent inspection robot charging controller method based on machine vision
CN106625676A (en) Three-dimensional visual accurate guiding and positioning method for automatic feeding in intelligent automobile manufacturing
CN110246175A (en) Intelligent Mobile Robot image detecting system and method for the panorama camera in conjunction with holder camera
CN105513072A (en) PTZ correction method
CN104793637B (en) The real-time tracking system and tracking of a kind of mobile device
CN107421466B (en) A kind of synchronous acquisition device and acquisition method of two and three dimensions image
CN110605718A (en) Transformer substation inspection robot system and inspection method
CN102253057B (en) Endoscope system and measurement method using endoscope system
CN106996748A (en) A kind of wheel footpath measuring method based on binocular vision
CN107389687A (en) A kind of electronic component appearance images harvester and its acquisition method
CN108190774B (en) A kind of row's rope fault detection method and its device based on projection
CN105357490A (en) Visual monitoring and target tracking method and system based on real-time three-dimensional model of transformer substation
CN117250200B (en) Square pipe production quality detection system based on machine vision
CN105307116A (en) Distributed vision positioning system and method based on mobile robot
CN112082770A (en) Rail vehicle body detection system and method based on artificial intelligence technology
CN104786227A (en) Drop type switch replacing control system and method based on high-voltage live working robot
CN107896326A (en) Binocular solid camera automatic leveling equipment, its levelling control system and control method
CN104858877B (en) High-tension line drop switch changes the control method of control system automatically
CN105068139B (en) A kind of characterization processes of piston cooling nozzle installment state
CN112258633B (en) SLAM technology-based scene high-precision reconstruction method and device

Legal Events

Date Code Title Description
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant