CN106892356B - A kind of tyre crane running gear automatic correction method based on machine vision - Google Patents
A kind of tyre crane running gear automatic correction method based on machine vision Download PDFInfo
- Publication number
- CN106892356B CN106892356B CN201710070800.4A CN201710070800A CN106892356B CN 106892356 B CN106892356 B CN 106892356B CN 201710070800 A CN201710070800 A CN 201710070800A CN 106892356 B CN106892356 B CN 106892356B
- Authority
- CN
- China
- Prior art keywords
- image
- camera
- path planning
- tyre crane
- point
- 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.)
- Active
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66C—CRANES; LOAD-ENGAGING ELEMENTS OR DEVICES FOR CRANES, CAPSTANS, WINCHES, OR TACKLES
- B66C13/00—Other constructional features or details
- B66C13/18—Control systems or devices
- B66C13/48—Automatic control of crane drives for producing a single or repeated working cycle; Programme control
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66C—CRANES; LOAD-ENGAGING ELEMENTS OR DEVICES FOR CRANES, CAPSTANS, WINCHES, OR TACKLES
- B66C13/00—Other constructional features or details
- B66C13/16—Applications of indicating, registering, or weighing devices
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66C—CRANES; LOAD-ENGAGING ELEMENTS OR DEVICES FOR CRANES, CAPSTANS, WINCHES, OR TACKLES
- B66C9/00—Travelling gear incorporated in or fitted to trolleys or cranes
- B66C9/16—Travelling gear incorporated in or fitted to trolleys or cranes with means for maintaining alignment between wheels and track
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
Abstract
The invention discloses a kind of tyre crane running gear automatic correction method based on machine vision, selects tyre crane side to install a camera respectively before and after it as guiding side and calibrated；Collection image simultaneously carries out appropriate pretreatment；Establish BP neural network and selected pixels point set is repeatedly trained；The minimum network of test error is selected image is identified；By the point that the image simplification after identification is series of discrete, straight-line detection is carried out using Hough transform and least square method and institute's detection of straight lines is merged into one as path planning；Splice the image of front camera and rear camera, and new coordinate system is established as origin using the upper left corner of image after splicing and re-establishes two planning path equations；Euclidean distance of two path plannings in pole coordinate parameter space is calculated, if distance is more than setting value, recalibrates camera, otherwise merges two path plannings；The angle and position of intersecting point of calculating path planning and image longitudinal midline simultaneously formulate correction strategy.
Description
Technical field
The present invention relates to machine vision and field of image recognition, particularly a kind of automobile navigation side based on machine vision
Method.
Background technology
Tyre crane has that airdrome maneuver is flexible, is not limited at harbour by Fixed Point Operation region, it is possible to increase harbour facilities profit
With rate, the features such as portal crane configures quantity is reduced.However as the fast development of logistics and the rising of cost of labor, to wheel
The service efficiency that tire is hung proposes higher requirement.It is that driver observes by the naked eye drawing for ground during traditional tyre crane traveling
Wire is using mode of manually rectifying a deviation, to ensure that tyre crane drives safely on fixed course.Such a drive manner efficiency is not
It is high and need driver to concentrate, energy is consumed larger, the accident such as easily cause maloperation or misoperation to collide.
Therefore realize that the automatic deviation correction of tyre crane running gear is significant, can further improve operating efficiency.
Gps carrier phase difference location technology is a kind of in the skill that location navigation and automatic deviation correction can be provided for tyre crane of utilization
Art.But the technical scheme needs to build GPS reference station and movement station in stockyard and differential received antenna is installed on tyre crane
Etc. equipment, equipment investment is big, technical requirements are high, and in order to reduce screening effect of the tyre crane body to gps signal, GPS days
Line need to be arranged on tyre crane top, be had a great influence away from ground level up to 15m-20m, positioning precision by stockyard surface water Pingdu.
Therefore the present invention proposes the tyre crane automatic correction method based on machine vision, by the identification tyre crane guide line of camera simultaneously
Path planning is calculated accordingly formulates correction strategy.It is more economical compared to using gps carrier phase difference technology, the program.To the greatest extent
Pipe can limit the normal work of the visual field effects camera of camera, the present invention is to weather under the weather conditions such as heavy snow dense fog
There is preferable robustness, draw not forming a wide range of ponding, accumulated snow and the enough misty rain weather of visible range and remaining to normal identification
Wire.
The content of the invention
It is an object of the invention to provide a kind of tyre crane running gear automatic correction method based on machine vision, can pass through
The guide line of camera effective detection tyre crane running gear simultaneously calculates path planning formulation correction strategy.
1. the technical solution adopted for the present invention to solve the technical problems comprises the following steps：
(a) selection selection tyre crane running gear side is as guiding side, its front and back position install respectively one it is same
Model and the camera of configuration are simultaneously calibrated；
(b) mapping to be checked of particular size and resolution ratio is obtained by appropriate image preprocessing to the image collected
Picture；
(c) establish BP neural network and choose a large amount of pixel point sets and repeatedly trained；
(d) choose the minimum network of test error image is identified, described image recognition is by each pixel
And its rgb value input neutral net of pixel is differentiated in peripheral extent；
(e) point by the image simplification after identification for series of discrete, using mixing Hough transform and the side of least square
Method carries out straight-line detection to image, by taking the average value of detected line correspondences parameter to merge the straight line detected, and
As path planning；
(f) splice front camera and rear camera image, new coordinate system and again is established by origin of the upper left corner of image after splicing
Establish the equation of two path plannings；
(g) two planning path parameters are transformed into pole coordinate parameter space, are expressed as (ρ1,θ1), (ρ2,θ2), meter
2 points of Euclidean distance is calculated, if distance is more than setting value, front camera and rear camera position there are variation needs to re-start calibration, otherwise
It is one to merge two planning path equations；
(h) according to the center of guiding side and the folder of the position relationship and path planning and image longitudinal midline of path planning
Formulate correction strategy in angle.
Further, the Installation And Calibration of the camera described in the step a, before front camera and rear camera should be arranged on guiding side
Rear sustained height and with horizontal plane folded by angle it is consistent, specific calibration steps is as follows：
Step1：Folded by straight line where making the guiding side of tyre crane be centrally located on guide line and making guiding side and guide line
Angle is more than 15 degree；
Step2：Front camera and rear camera gathers image and pre-processed respectively, the guide line and image extended in each image
Longitudinal midline makes it intersect at outside image.
Step3：The distance of measure intersection point to image is calculated as l1, l2 respectively, if the two is inconsistent, adjustment camera makes it
It is consistent and be calculated as l；
Further, the large format high-definition picture that the pretreatment described in the step b refers to collect is divided
Resolution reduces and intercepts area-of-interest, if picture noise seriously will also carry out image filtering processing.
Further, the foundation and training of the BP neural network described in the step c, it is 75 to establish input layer, defeated
Go out the BP neural network that node layer number is 1, choosing size outside the pixel point set and guide line that size on a large amount of guide lines is 5*5 is
5*5 pixel point set is simultaneously trained it as positive example and counter-example to neutral net, and this positive counter-example, which is chosen, should be included respectively
The image shot under kind working condition, and the positive and negative number of cases amount chosen under various operating modes needs to reach certain amount level.
Further, the image simplification described in the step e and the straight-line detection side of mixing Hough transform and least square
The method of method, wherein image simplification is as follows：
Step1：The bianry image obtained to identification takes a horizontal straight line every the width of 5 pixels；
Step2：The midpoint of each connected region on the straight line is taken to simplify bianry image.
The line detection method for mixing Hough transform and least square is as follows：
Step1：Point in simplification figure picture is changed from rectangular co-ordinate image space to pole coordinate parameter space；
Step2：Maximum is carried out to parameter space and suppresses the point that simultaneously Selecting All Parameters spatial value is more than 0.3 times of maximum；
Step3：Calculate the pixel and linear equation of image space corresponding to selected point；
Step4：Reject laterally or close to the very few straight line of the straight line and corresponding pixel points of cross direction profiles；
Step5：Straight line close to linear equation parameter and that corresponding pixel points are inconsistent merges；
Step6：The point of every institute's detection of straight lines certain distance of selected distance, re-starts least squares line fitting.
Further, the image mosaic described in the step f is to keep preceding camera image constant, rear camera image
It is positioned over after rotation 180 degree immediately below preceding camera image and is spaced 2*l.
Further, rectified a deviation in the step h formulation of strategy, concrete mode is as follows：
The angle span for defining path planning and image longitudinal midline first be [- 90,90), it is right positioned at image center line
Side is just, left side is negative；
(1) intersection point of path planning and stitching image longitudinal midline is located at stitching image overcentre, and left and right tire is synchronized
Motion；
(2) intersection point of path planning and stitching image longitudinal midline is located at stitching image center or below and angle is
Just, left side tire speed is more than right side tire speed, tyre crane is deflected to the right；
(3) intersection point of path planning and stitching image longitudinal midline is located at stitching image center or below and angle is
Negative, right side tire speed is more than left side tire speed, tyre crane is deflected to the left；
(4) path planning is parallel with image longitudinal midline, if path planning is located on the right side of image longitudinal midline, left side wheels
Tire speed is more than right side tire speed, tyre crane is deflected to the right；If path planning is located on the left of image longitudinal midline, right-hand wheel
Tire speed is more than left side tire speed, tyre crane is deflected to the left；
(5) path planning is completely superposed with image longitudinal midline, and left and right tire moves at the same speed.
The beneficial effects of the invention are as follows：
Compared to the correction airmanship based on GPS, the automatic correction method based on machine vision can effectively reduce tire
Hang the cost of self-navigation correction.Simultaneously to overcome bad weather to identify the influence of guide line to camera, the present invention is using god
The image under various operating modes is trained through network, therefore there is stronger robustness to a variety of bad working environments and weather.I.e.
Make have situations such as car trace, greasy dirt, fade on guide line, can still accurately identify.In addition do not formed a wide range of ponding, accumulated snow and
The enough misty rain weather of visible range remains to normally identify guide line, does not influence tyre crane normal work.
Brief description of the drawings
Fig. 1：Preceding camera calibrates schematic diagram
Fig. 2：Pretreated image
Fig. 3：The location drawing picture of neural network recognization
Fig. 4：Bianry image simplifies design sketch
Fig. 5：Straight-line detection design sketch
Fig. 6：Stitching image schematic diagram
Fig. 7：Straight line transformation parameter schematic diagram
Embodiment
The embodiment of the present invention is described further below in conjunction with the accompanying drawings.
For the present invention using tyre crane running gear as control object, the system as core and includes steering using PLC control circuits
The mechanical structures such as device, wheel, driver.In specific embodiment, the posture of tyre crane is judged by installing machine visual unit additional
With the next step direction of motion and result is transferred to PLC control unit, so as to realize the correction of tyre crane running gear.The machine
Visual unit is handled by former and later two camera collection images and being transferred in industrial computer, finally transmits result
To PLC.
Specific implementation step is broadly divided into image preprocessing, guide line identification, image mosaic, tactful four portions of formulation correction
Point.Image preprocessing is mainly to be filtered denoising to the image of collection, and image resolution ratio is reduced to size and intercepted and feels emerging
Interesting region；The identification of guide line is trained and identified using BP neural network and entered by Hough transform and least square method
The further detection of row；Image mosaic is the image mosaic by front camera and rear camera shooting into piece image；Finally according to guiding side
Center and the angle of position relationship and path planning and image longitudinal midline of path planning formulate correction strategy.Specifically such as
Under：
1st, image preprocessing
1.1 install and calibrate camera
The CCD camera of two same models and configuration is separately mounted to the anteroposterior position above the tire of tyre crane side
The sustained height put and respectively with horizontal plane folded by angle it is consistent.To ensure that installation is correct, it is necessary to carry out necessary calibration：
Step1：Folded by straight line where making the guiding side of tyre crane be centrally located on guide line and making guiding side and guide line
Angle is more than 15 degree；
Step2：Front camera and rear camera gathers image and pre-processed respectively, the guide line and image extended in each image
It is outer (as shown in Figure 1) that longitudinal midline makes it intersect at image.
Step3：The distance of measure intersection point to image is calculated as l1, l2 respectively, if the two is inconsistent, adjustment camera makes it
It is consistent and be calculated as l；
1.2 images obtain and pretreatment
The frame per second for setting camera is 30fps, and setting image resolution ratio is 1280*960, and it is in image to set image ROI
Heart 350*650 region.A sub-picture is gathered every 3 frames, and is 175*325 by image down, then using in 3*3 templates
Value filtering carries out medium filtering, and the pending image finally obtained is as shown in Figure 2.
2nd, guide line identifies
2.1 training BP neural networks
First, the image totally 500 that should include being shot under various working conditions is chosen, includes each period of fine day, it is cloudy
My god, (there is secondary light source at night), rain, mist, snow (not causing deposite snow situation) etc..Then guiding is chosen in each sub-picture
The positive example that the pixel point set that size is 5*5 on line is trained as BP neural network, choosing non-guide line part size in image is
The counter-example that 5*5 pixel point set is trained as BP neural network.The quantity that each sub-picture gathers positive and negative example is respectively 15~25
Individual, 500 images, which amount to, gathers positive example 11342, counter-example 10975.
It is 75 to establish input layer, and output layer nodes are 1, and hidden layer node is 172 BP neural network.Nerve net
The transmission function of network hidden layer is bipolarity S function, and output layer transmission function is linear function.By each positive example or counter-example number
It is trained according to form and the neutral net that has built up of input that 75*1 is stored as by row, trains 20 times altogether and count each instruction
Practice the test error of successful network.
2.2 image recognition
Choose the minimum neutral net of test error to differentiate each pixel in image, and will differentiate that result is carried out
Binaryzation be identified after bianry image, as shown in Figure 3.For identifying width of the obtained bianry image every 5 pixels
Degree takes a horizontal straight line and takes the midpoint of each connected region on the straight line to simplify bianry image, as shown in Figure 4.In figure
" * " represents the point after simplifying.
2.3 inductive line detection
Straight-line detection and every institute's detection of straight lines one of selected distance are carried out to the bianry image after simplification using Hough transform
The point of set a distance, re-starts least squares line fitting, and testing result is as shown in Figure 5.Four straightways in figure represent inspection
The guide line measured.The line detection method further constrained the straight line of detection on the basis of Hough transform, including with
Lower step：
The line detection method for mixing Hough transform and least square is as follows：
Step1：Point in simplification figure picture is changed from rectangular co-ordinate image space to pole coordinate parameter space；
Step2：Maximum is carried out to parameter space and suppresses the point that simultaneously Selecting All Parameters spatial value is more than 0.3 times of maximum；
Step3：Calculate the pixel and linear equation of image space corresponding to selected point；
Step4：Reject horizontal or the straight line close to cross direction profiles and corresponding straight line of the points less than 10；
Step5：Straight line close to linear equation parameter and that corresponding pixel points are inconsistent merges；
Step6：The point of every institute's detection of straight lines certain distance of selected distance, re-starts least squares line fitting.
By taking the average value of detected line correspondences parameter to merge the straight line detected, and it is used as path planning.
3 image mosaics
Image mosaic is the information for the detection for merging front camera and rear camera, to carry out self-test to camera position state and to be
Determine that the formulation of tyre crane posture and correction strategy provides help.
It is constant that image split-joint method is to maintain preceding camera image, is positioned over after rear camera image rotation 180 degree proactive
As head image underface and it is spaced 2*l.New coordinate system is established as origin using the upper left corner of image after splicing and re-establishes two
The equation (as shown in Figure 6) of bar path planning.Two planning path parameters are transformed into pole coordinate parameter space (such as Fig. 7 institutes
Show), it is expressed as (ρ1,θ1), (ρ2,θ2), the Euclidean distance of 2 points of calculating, if distance is more than setting value, front camera and rear camera
Position has variation needs to re-start calibration, and it is one otherwise to merge two planning path equations.
4 formulate correction strategy
According to the center of guiding side and the angle of the position relationship and path planning and image longitudinal midline of path planning
Formulate correction strategy.The angle span for defining path planning and image longitudinal midline first be [- 90,90), positioned at image
For just, left side is negative on the right side of center line.
(1) intersection point of path planning and stitching image longitudinal midline is located at stitching image overcentre, and left and right tire is synchronized
Motion；
(2) intersection point of path planning and stitching image longitudinal midline is located at stitching image center or below and angle is
Just, left side tire speed is more than right side tire speed, tyre crane is deflected to the right；
(3) intersection point of path planning and stitching image longitudinal midline is located at stitching image center or below and angle is
Negative, right side tire speed is more than left side tire speed, tyre crane is deflected to the left；
(4) path planning is parallel with image longitudinal midline, if path planning is located on the right side of image longitudinal midline, left side wheels
Tire speed is more than right side tire speed, tyre crane is deflected to the right；If path planning is located on the left of image longitudinal midline, right-hand wheel
Tire speed is more than left side tire speed, tyre crane is deflected to the left；
(5) path planning is completely superposed with image longitudinal midline, and left and right tire moves at the same speed.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " illustrative examples ",
The description of " example ", " specific example " or " some examples " etc. means to combine specific features, the knot that the embodiment or example describe
Structure, material or feature are contained at least one embodiment or example of the present invention.In this manual, to above-mentioned term
Schematic representation is not necessarily referring to identical embodiment or example.Moreover, specific features, structure, material or the spy of description
Point can combine in an appropriate manner in any one or more embodiments or example.
Although an embodiment of the present invention has been shown and described, it will be understood by those skilled in the art that：Not
In the case of departing from the principle and objective of the present invention a variety of change, modification, replacement and modification can be carried out to these embodiments, this
The scope of invention is limited by claim and its equivalent.
Claims (6)
1. a kind of tyre crane running gear automatic correction method based on machine vision, it is characterised in that comprise the following steps：
(a) select tyre crane running gear side that one same model is installed respectively in its front and back position and matched somebody with somebody as guiding side
The camera put simultaneously is calibrated；The Installation And Calibration of middle camera, front camera and rear camera should be installed in same before and after guiding side
Height and with horizontal plane folded by angle it is consistent, specific calibration steps is as follows:Stepa1：Make the guiding side of tyre crane be centrally located to draw
On wire and guiding side place straight line is set to be more than 15 degree with angle folded by guide line；Stepa2：Front camera and rear camera gathers figure respectively
Picture is simultaneously pre-processed, and the guide line and image longitudinal midline extended in each image makes it intersect at outside image；Stepa3：Survey
The distance for determining intersection point to image is calculated as l1, l2 respectively, if the two is inconsistent, adjustment camera makes it consistent and is calculated as l；
(b) image to be detected of particular size and resolution ratio is obtained by appropriate image preprocessing to the image collected；
(c) establish BP neural network and choose a large amount of pixel point sets and repeatedly trained；
(d) the minimum network of test error is chosen image is identified, described image recognition be by each pixel and its
The rgb value input neutral net of pixel is differentiated in the certain limit of periphery；
(e) point by the image simplification after identification for series of discrete, using mixing Hough transform and the method pair of least square
Image carries out straight-line detection, by taking the average value of detected line correspondences parameter to merge the straight line detected, and conduct
Path planning；
(f) splice front camera and rear camera image, using the upper left corner of image after splicing as origin establish new coordinate system lay equal stress on it is newly-built
Found the equation of two path plannings；
(g) two planning path parameters are transformed into pole coordinate parameter space, are expressed as (ρ1,θ1), (ρ2,θ2), calculate two
The Euclidean distance of point, if distance is more than setting value, front camera and rear camera position has variation needs to re-start calibration, otherwise merged
Two planning path equation is one；
(h) according to the center of guiding side and the angle system of the position relationship and path planning and image longitudinal midline of path planning
Fixed correction strategy.
2. according to the method for claim 1, it is characterised in that pretreatment refers to collect significantly in the step b
Face high-definition picture carries out resolution ratio reduction and intercepts area-of-interest, if picture noise seriously will be also carried out at image filtering
Reason.
3. according to the method for claim 1, it is characterised in that the foundation of the BP neural network described in the step c and
Training, it is 75 to establish input layer, and output layer nodes are 1 BP neural network, and it is 5*5 to choose size on a large amount of guide lines
Pixel point set and the outer size of guide line be 5*5 pixel point set and neutral net is carried out using it as positive example and counter-example
Training, this positive counter-example choose the image that should include being shot under various working conditions, and the positive and negative number of cases amount chosen under various operating modes
Need to reach certain amount level.
4. according to the method for claim 1, it is characterised in that in the step e image simplification and mixing Hough transform and
The method of the line detection method of least square, wherein image simplification is as follows：
Stepe1：The bianry image obtained to identification takes a horizontal straight line every the width of 5 pixels；
Stepe2：The midpoint of each connected region on the straight line is taken to simplify bianry image；
The line detection method for mixing Hough transform and least square is as follows：
Step1：Point in simplification figure picture is changed from rectangular co-ordinate image space to pole coordinate parameter space；
Step2：Maximum is carried out to parameter space and suppresses the point that simultaneously Selecting All Parameters spatial value is more than 0.3 times of maximum；
Step3：Calculate the pixel and linear equation of image space corresponding to selected point；
Step4：Reject laterally or close to the very few straight line of the straight line and corresponding pixel points of cross direction profiles；
Step5：Straight line close to linear equation parameter and that corresponding pixel points are inconsistent merges；
Step6：The point of every institute's detection of straight lines certain distance of selected distance, re-starts least squares line fitting.
5. according to the method for claim 1, it is characterised in that image mosaic in the step f, keep preceding camera image
It is constant, it is positioned over after rear camera image rotation 180 degree immediately below preceding camera image and is spaced 2*l.
6. according to the method for claim 1, it is characterised in that the formulation for strategy of being rectified a deviation in the step h, concrete mode is such as
Under：
The angle span for defining path planning and image longitudinal midline first be [- 90,90), be on the right side of image center line
Just, left side is negative；
The intersection point of path planning and stitching image longitudinal midline is located at stitching image overcentre, and left and right tire moves at the same speed；
The intersection point of path planning and stitching image longitudinal midline is located at stitching image center or below and angle is just left side
Tire speed is more than right side tire speed, tyre crane is deflected to the right；
The intersection point of path planning and stitching image longitudinal midline is located at stitching image center or below and angle is negative, right side
Tire speed is more than left side tire speed, tyre crane is deflected to the left；
Path planning is parallel with image longitudinal midline, if path planning is located on the right side of image longitudinal midline, left side tire speed
More than right side tire speed, tyre crane is set to deflect to the right；If path planning is located on the left of image longitudinal midline, right side tire speed
More than left side tire speed, tyre crane is set to deflect to the left；
Path planning is completely superposed with image longitudinal midline, and left and right tire moves at the same speed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710070800.4A CN106892356B (en) | 2017-02-09 | 2017-02-09 | A kind of tyre crane running gear automatic correction method based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710070800.4A CN106892356B (en) | 2017-02-09 | 2017-02-09 | A kind of tyre crane running gear automatic correction method based on machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106892356A CN106892356A (en) | 2017-06-27 |
| CN106892356B true CN106892356B (en) | 2018-02-27 |
Family
ID=59198094
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710070800.4A Active CN106892356B (en) | 2017-02-09 | 2017-02-09 | A kind of tyre crane running gear automatic correction method based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106892356B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109993788B (en) * | 2017-12-29 | 2021-10-26 | 西门子（中国）有限公司 | Deviation rectifying method, device and system for tyre crane |
| CN107934789B (en) * | 2018-01-17 | 2020-03-31 | 潘莲英 | Intelligent container loading and unloading method based on polar coordinates |
| CN107973221B (en) * | 2018-01-17 | 2020-04-21 | 温州铭泰工业设计有限公司 | Intelligent crane based on infrared measurement |
| CN108549877B (en) * | 2018-04-23 | 2022-02-18 | 重庆大学 | Tracking robot track identification method based on neural network |
| CN110054089B (en) * | 2019-04-29 | 2020-06-09 | 北京航天自动控制研究所 | Automatic vision deviation rectifying system and method for tire crane |
| CN110844785B (en) * | 2019-11-28 | 2020-12-29 | 重庆中星微人工智能芯片技术有限公司 | Method, device, equipment and medium for generating information of tower crane boom |
| CN110963409B (en) * | 2019-11-29 | 2021-02-09 | 北京航天自动控制研究所 | Method for measuring automatic deviation rectification deviation of tire crane machine vision |
| CN111017727B (en) * | 2019-11-29 | 2021-11-16 | 北京航天自动控制研究所 | Automatic deviation-rectifying control shutdown judgment method for tire crane |
| CN113419526B (en) * | 2021-06-17 | 2022-08-23 | 江苏大学 | Automatic pond transferring system of aquaculture workboat and pose adjusting method |
| CN113860172A (en) * | 2021-09-30 | 2021-12-31 | 广州文远知行科技有限公司 | Deviation rectifying method and device, vehicle and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SU1440853A1 (en) * | 1986-01-15 | 1988-11-30 | П. П. Гил ров | Running gear of overhead traveling crane |
| CN201280386Y (en) * | 2008-09-12 | 2009-07-29 | 天津港联盟国际集装箱码头有限公司 | Automatic deviation correction system of tire crane |
| CN102060234A (en) * | 2010-10-26 | 2011-05-18 | 常州超媒体与感知技术研究所有限公司 | Tire crane traveling track video correction device and method |
| CN201882785U (en) * | 2010-02-02 | 2011-06-29 | 南通通镭软件有限公司 | Position image recognition and correction system for container gantry crane |
| CN105438998A (en) * | 2014-09-26 | 2016-03-30 | 上海海镭激光科技有限公司 | Rubber-tired crane walking positioning and deviation rectifying anticollision method |
-
2017
- 2017-02-09 CN CN201710070800.4A patent/CN106892356B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SU1440853A1 (en) * | 1986-01-15 | 1988-11-30 | П. П. Гил ров | Running gear of overhead traveling crane |
| CN201280386Y (en) * | 2008-09-12 | 2009-07-29 | 天津港联盟国际集装箱码头有限公司 | Automatic deviation correction system of tire crane |
| CN201882785U (en) * | 2010-02-02 | 2011-06-29 | 南通通镭软件有限公司 | Position image recognition and correction system for container gantry crane |
| CN102060234A (en) * | 2010-10-26 | 2011-05-18 | 常州超媒体与感知技术研究所有限公司 | Tire crane traveling track video correction device and method |
| CN105438998A (en) * | 2014-09-26 | 2016-03-30 | 上海海镭激光科技有限公司 | Rubber-tired crane walking positioning and deviation rectifying anticollision method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106892356A (en) | 2017-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106892356B (en) | A kind of tyre crane running gear automatic correction method based on machine vision | |
| CN104299240B (en) | Camera calibration method and system for lane shift early warning | |
| CN105955259A (en) | Monocular vision AGV accurate positioning method and system based on multi-window real-time range finding | |
| CN107025432B (en) | A kind of efficient lane detection tracking and system | |
| CN106441286B (en) | Unmanned plane tunnel cruising inspection system based on BIM technology | |
| CN103413313B (en) | The binocular vision navigation system of electrically-based robot and method | |
| CN105700532B (en) | The Intelligent Mobile Robot navigator fix control method of view-based access control model | |
| CN104933409B (en) | A kind of parking stall recognition methods based on panoramic picture dotted line feature | |
| CN107462223A (en) | Driving sight distance self-operated measuring unit and measuring method before a kind of highway is turned | |
| CN103714538B (en) | road edge detection method, device and vehicle | |
| CN104637073B (en) | It is a kind of based on the banding underground structure detection method for shining upon shadow compensation | |
| CN103593671B (en) | The wide-range lane line visible detection method worked in coordination with based on three video cameras | |
| CN106651953A (en) | Vehicle position and gesture estimation method based on traffic sign | |
| CN105511462B (en) | A kind of AGV air navigation aids of view-based access control model | |
| CN104848851A (en) | Transformer substation patrol robot based on multi-sensor data fusion picture composition and method thereof | |
| CN106056100A (en) | Vehicle auxiliary positioning method based on lane detection and object tracking | |
| CN106774313A (en) | A kind of outdoor automatic obstacle-avoiding AGV air navigation aids based on multisensor | |
| CN107229908A (en) | A kind of method for detecting lane lines | |
| CN105447853A (en) | Flight device, flight control system and flight control method | |
| CN103386975A (en) | Vehicle obstacle avoidance method and system based on machine vision | |
| CN102682292A (en) | Method based on monocular vision for detecting and roughly positioning edge of road | |
| CN105136153B (en) | A kind of lane line exact position harvester and acquisition method | |
| CN108364466A (en) | A kind of statistical method of traffic flow based on unmanned plane traffic video | |
| CN106584451A (en) | Visual navigation based transformer substation automatic composition robot and method | |
| CN110398979A (en) | A kind of unmanned engineer operation equipment tracking method and device that view-based access control model is merged with posture |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |