CN109887027A - A kind of method for positioning mobile robot based on image - Google Patents
A kind of method for positioning mobile robot based on image Download PDFInfo
- Publication number
- CN109887027A CN109887027A CN201910003337.0A CN201910003337A CN109887027A CN 109887027 A CN109887027 A CN 109887027A CN 201910003337 A CN201910003337 A CN 201910003337A CN 109887027 A CN109887027 A CN 109887027A
- Authority
- CN
- China
- Prior art keywords
- image
- width
- length
- unit pixel
- pixel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Image Analysis (AREA)
- Manipulator (AREA)
- Image Processing (AREA)
Abstract
The present invention proposes a kind of method for positioning mobile robot based on image;The present invention be directed to prior arts to carry out distortion correction to image, then processing image does the process positioned, a kind of method for rapidly positioning directly positioned on fault image is provided, it makes full use of the aberration rate of image to be calculated, to realize the positioning to mobile robot more accurately and fast.The present invention does not need to carry out distortion correction, directly carries out framing using fault image, while ensure that positioning accuracy, substantially reduces the time required for positioning.
Description
Technical field
The invention belongs to localization for Mobile Robot fields, and in particular to a kind of image position method of mobile robot.
Background technique
With the continuous development of robot technology, the application range of mobile robot in real life constantly expands, people
Higher demand has been located for mobile robot.In previous traditional localization method, absolute fix mainly passes through
Global positioning system (GPS) is completed, but indoors and in occluded environment, and signal transmission will receive limitation, so that positioning accurate
Degree decline, or even failure.Relative positioning is the encoder and used by the realizations positioning such as encoder, inertial sensor and laser radar
Property sensor can not directly acquire location information, and laser radar is expensive, these all give the commercialization, daily of mobile robot
Change brings inconvenience.
Recently, framing technology is more and more applied in the industrial production, main to obtain week by video camera
The image of environmental scenery is enclosed, and image information is handled, the positioning of Lai Shixian robot.But video camera obtain image with
The visual field picture that people can see is compared, and still has certain gap in coverage, image quality, the image of shooting can generate one
Fixed pattern distortion, this distortion will cause detrimental effect to the processing analysis of framing, can determine mobile robot
Error is caused in position, or even cannot achieve positioning.
Summary of the invention
In view of the deficiencies of the prior art, the present invention proposes a kind of method for positioning mobile robot based on image.
The present invention be directed to prior arts to carry out distortion correction to image, then handles image and does the process positioned, mentions
For the method for rapidly positioning that one kind is directly positioned on fault image, it makes full use of the aberration rate of image to be calculated,
To realize the positioning to mobile robot more accurately and fast.
Localization method of the present invention includes the following steps:
Step 1: video camera is suspended on certain altitude vertically downward, guarantee camera lens and ground level, it is known on the ground
Paste N number of cross mark in the region of position;It connects video camera to be shot, obtain original with N number of cross mark
Color image;
Step 2: handling image using Raw color image as input, the cross mark of known location in image is known
It does not come out;
Step 3: being directed to the cross mark of each position, length in pixels and width in image are obtained, further according to every
The physical length and width of a cross mark calculate the corresponding physical length of position unit pixel and width;
Step 4: utilizing each cross mark physical length obtained in step 3 using picture centre as coordinate origin
With the ratio of length in pixels, it is established that the functional relation of the length and width that unit pixel represents and the point to reference axis distance,
I.e. pixel deformation is distributed coordinate system;
Step 5: the mobile robot to be positioned in identification image, handles image, by the location of pixels of mobile robot
It identifies;
Step 6: the actual calculation of location of moving trolley is gone out using the pixel deformation distribution coordinate system established in step 4
Come.
In the step 3, for each cross mark, the length in pixels and width in image is obtained, this is calculated
The corresponding physical length of position unit pixel and width, are handled in the following ways:
1) length in pixels and width are obtained: image first being switched into hsv color space by RGB, all colours in image are sieved
Choosing, color of object is retained, other colors are set to black, then is done opening and closing operations to image, found profile, required for filtering out
Cross mark, and minimum circumscribed rectangle is drawn out, the parameter of boundary rectangle is exported;
2) it according to the length and width parameter of boundary rectangle and the cross actual length and width of mark, calculates
The corresponding physical length of position unit pixel and width out.
Using picture centre as coordinate origin in the step 4, it is established that the length and width and the point that unit pixel represents
To the functional relation of reference axis distance, i.e. pixel deformation distribution coordinate system refers to:
It is divided into fault image and is furnished with N number of cross mark, the physical length of each cross mark or width are removed
With its length in pixels or width in the picture, can obtain at each identification point, physical length representated by unit pixel
And width, due to the presence of distortion, length and width representated by the unit pixel of this N number of identification point is different, then root
According to the location information of N number of identification point in the picture, the actual range and pixel to seat of unit pixel representative can be set up
The functional relation of the distance of parameter;
If identification point N1Distance apart from Y-axis isThe corresponding physical length of unit pixel isIdentification point N2Distance
The distance of Y-axis isThe corresponding physical length of unit pixel is... identification point NnDistance apart from Y-axis isUnit
The corresponding physical length of pixel isI.e.
Integration obtains unit pixel length and pixel to the functional relation of y-axis distance:
PX(x)=fX(x)
It can similarly obtain the functional relation of unit pixel width with pixel to x-axis distance:
PY(y)=fY(y)
Coordinate system is distributed the actual bit of moving trolley using the pixel deformation established in step 4 in the step 6
It sets to calculate and refer to:
If the pixel coordinate of moving trolley in the picture is (xc,yc), then trolley physical location (xcr,ycr) can be asked by following formula
:
Wherein, fX(l) and fYIt (l) is function of the actual range of unit pixel representative at a distance from pixel to reference axis
Relationship.
Beneficial effects of the present invention: (the advantages of the present invention over the prior art are that :)
The present invention does not need to carry out distortion correction, directly carries out framing using fault image, ensure that positioning accurate
While spending, the time required for positioning is substantially reduced.
Detailed description of the invention
Fig. 1 is positioning system figure;
Fig. 2 is a kind of flow diagram of the image position method of mobile robot of the present invention;
Fig. 3 (A) is original color image;
Fig. 3 (B) is dithering image;
Fig. 3 (C) is cross mark identification image;
Fig. 3 (D) is the image of moving trolley identification.
Specific embodiment
The present invention is described in further detail with example with reference to the accompanying drawing.It is understood that described herein
Specific example be used only for explaining the present invention rather than limiting the invention.It is related to the present invention shown in attached drawing
Part, and not all content.
The specific embodiment of whole process of the present invention illustrated below is following (each step effect picture is referring to figure):
Positioning system is as shown in Figure 1.
The process of localization method is as shown in Figure 2.
Step 1 inevitably generates distortion due to obtain biggish angular field of view, the picture of shooting, and marginal portion is outstanding
To be obvious, center in localization region installs a video camera vertically downward, guarantees camera lens and ground basic horizontal, takes the photograph
As the other end connection of head is used to handle the computer of image information;Using localization region central point as origin, horizontal line is x-axis, is erected
Straight line is that y-axis establishes coordinate system, is needing to paste N number of cross mark in the region positioned, and shoot picture, is obtaining and have ten
The Raw color image of font mark, as shown in Fig. 3 (A).
Step 2 computer acquires image using video camera, and collected Raw color image is transferred to image procossing system
In system, system handles image information, identifies cross, and export length in pixels and width, such as Fig. 3
(C) shown in.
The length in pixels of obtained cross mark and width are divided by by step 3 with physical length and width, obtain the position
Set the corresponding length and width of unit pixel of cross mark.
Step 4 using picture centre as coordinate origin, using each cross mark physical length obtained in step 3 with
The ratio of length in pixels, it is established that the functional relation of the length and width that unit pixel represents and the point to reference axis distance, i.e.,
Pixel deformation is distributed coordinate system.
Step 5 utilizes video camera to obtain moving machine when the mobile robot with color identifier is when localization region is mobile
The image of device people obtains the location of pixels coordinate (x of mobile robot in the picture by the operation in step 2c,yc), such as Fig. 3
(D) shown in.
Step 6 comes out the actual calculation of location of moving trolley using the pixel deformation distribution coordinate system established in step 4.
Unit pixel in the step 3 corresponds to what length and width obtained in the following ways:
1) collected cross mark original color image is transformed into hsv color space by BGR, then to cross mark
Know color gamut Screening Treatment, obtains doing open and close operation such as Fig. 3 (B), remove noise, connect domain, then image is done
Binaryzation, then find the profile in image, and draw out the minimum circumscribed rectangle of profile, by the length and width of boundary rectangle with
And center point coordinate export.
2) it according to the length and width parameter of boundary rectangle and the cross actual length and width of mark, calculates
The corresponding physical length of position unit pixel and width out.
Using picture centre as coordinate origin in the step 4, it is established that the length and width and the point that unit pixel represents
To the functional relation of reference axis distance, i.e. pixel deformation distribution coordinate system refers to:
It is divided into fault image and is furnished with N number of cross mark, the physical length of each cross mark or width are removed
It with its length or width in the picture, can obtain at each identification point, physical length representated by unit pixel and width
Degree, due to the presence of distortion, length and width representated by the unit pixel of this N number of identification point be it is different, further according to N number of
The location information of identification point in the picture, the actual range and pixel that can set up unit pixel representative arrive reference axis
The functional relation of distance.
If identification point N1Distance apart from Y-axis isThe corresponding physical length of unit pixel isIdentification point N2Distance
The distance of Y-axis isThe corresponding physical length of unit pixel is... identification point NnDistance apart from Y-axis isUnit
The corresponding physical length of pixel isI.e.
Integrate the functional relation of available unit pixel length with pixel to y-axis distance:
PX(x)=fX(x)
It can similarly obtain the functional relation of unit pixel width with pixel to x-axis distance:
PY(y)=fY(y)
Coordinate system is distributed the physical location of moving trolley using the pixel deformation established in step 4 in the step 6
It calculates and refers to:
If the pixel coordinate of moving trolley in the picture is (xc,yc), then trolley physical location (xcr,ycr) can be asked by following formula
:
Wherein, fX(l) and fYIt (l) is function of the actual range of unit pixel representative at a distance from pixel to reference axis
Relationship.
Claims (2)
1. a kind of method for positioning mobile robot based on image, which is characterized in that this method specifically includes the following steps:
Step 1: video camera is suspended on certain altitude vertically downward, guarantee camera lens and ground level, on the ground known location
Region paste N number of cross mark;It connects video camera to be shot, obtains the original color for having N number of cross mark
Image;
Step 2: handling image using Raw color image as input, the cross mark of known location in image being identified
Come;
Step 3: being directed to the cross mark of each position, length in pixels and width in image are obtained, further according to each ten
The physical length and width of font mark, calculate the corresponding physical length of position unit pixel and width;
Step 4: utilizing each cross mark physical length and picture obtained in step 3 using picture centre as coordinate origin
The ratio of plain length, it is established that the length and width that unit pixel represents and the point to the functional relation of reference axis distance, i.e. picture
Plain deformation is distributed coordinate system;
Step 5: the mobile robot to be positioned in identification image, handles image, the location of pixels of mobile robot is identified
Out;
Step 6: the actual calculation of location of moving trolley is come out using the pixel deformation distribution coordinate system established in step 4.
2. a kind of method for positioning mobile robot based on image according to claim 1, it is characterised in that: the step
In three, for each cross mark, the length in pixels and width in image is obtained, it is corresponding to calculate the position unit pixel
Physical length and width, handled in the following ways:
1) length in pixels and width are obtained: image first being switched into hsv color space by RGB, all colours in image are screened, it will
Color of object retains, other colors are set to black, then does opening and closing operations to image, finds profile, filters out required cross
Type mark, and minimum circumscribed rectangle is drawn out, the parameter of boundary rectangle is exported;
2) according to the length and width parameter of boundary rectangle and the cross actual length and width of mark, this is calculated
The corresponding physical length of position unit pixel and width;
Using picture centre as coordinate origin in the step 4, it is established that the length and width that unit pixel represents and the point to seat
The functional relation of parameter distance, i.e. pixel deformation distribution coordinate system refer to:
It is divided into fault image and is furnished with N number of cross mark, by the physical length of each cross mark or width divided by it
Length in pixels or width in the picture, can obtain at each identification point, physical length representated by unit pixel and width
Degree, due to the presence of distortion, length and width representated by the unit pixel of this N number of identification point be it is different, further according to N number of
The location information of identification point in the picture, the actual range and pixel that can set up unit pixel representative arrive reference axis
The functional relation of distance;
If identification point N1Distance apart from Y-axis isThe corresponding physical length of unit pixel isIdentification point N2Apart from Y-axis
Distance isThe corresponding physical length of unit pixel isIdentification point NnDistance apart from Y-axis isUnit pixel
Corresponding physical length isI.e.
Integration obtains unit pixel length and pixel to the functional relation of y-axis distance:
PX(x)=fX(x)
It can similarly obtain the functional relation of unit pixel width with pixel to x-axis distance:
PY(y)=fY(y)
Coordinate system is distributed the physical location meter of moving trolley using the pixel deformation established in step 4 in the step 6
It calculates and refers to:
If the pixel coordinate of moving trolley in the picture is (xc, yc), then trolley physical location (xcr, ycr) can be acquired by following formula:
Wherein, fX(l) and fYIt (l) is functional relation of the actual range of unit pixel representative at a distance from pixel to reference axis.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910003337.0A CN109887027A (en) | 2019-01-03 | 2019-01-03 | A kind of method for positioning mobile robot based on image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910003337.0A CN109887027A (en) | 2019-01-03 | 2019-01-03 | A kind of method for positioning mobile robot based on image |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109887027A true CN109887027A (en) | 2019-06-14 |
Family
ID=66925548
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910003337.0A Pending CN109887027A (en) | 2019-01-03 | 2019-01-03 | A kind of method for positioning mobile robot based on image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109887027A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112215892A (en) * | 2020-10-22 | 2021-01-12 | 常州大学 | Method for monitoring position and motion path of site robot |
CN112991742A (en) * | 2021-04-21 | 2021-06-18 | 四川见山科技有限责任公司 | Visual simulation method and system for real-time traffic data |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104732577A (en) * | 2015-03-10 | 2015-06-24 | 山东科技大学 | Building texture extraction method based on UAV low-altitude aerial survey system |
CN106950399A (en) * | 2017-03-30 | 2017-07-14 | 河海大学 | Broad surface flow field figure is as test system automatic calibration device and method |
CN108873943A (en) * | 2018-07-20 | 2018-11-23 | 南京奇蛙智能科技有限公司 | A kind of image processing method that unmanned plane Centimeter Level is precisely landed |
-
2019
- 2019-01-03 CN CN201910003337.0A patent/CN109887027A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104732577A (en) * | 2015-03-10 | 2015-06-24 | 山东科技大学 | Building texture extraction method based on UAV low-altitude aerial survey system |
CN106950399A (en) * | 2017-03-30 | 2017-07-14 | 河海大学 | Broad surface flow field figure is as test system automatic calibration device and method |
CN108873943A (en) * | 2018-07-20 | 2018-11-23 | 南京奇蛙智能科技有限公司 | A kind of image processing method that unmanned plane Centimeter Level is precisely landed |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112215892A (en) * | 2020-10-22 | 2021-01-12 | 常州大学 | Method for monitoring position and motion path of site robot |
CN112215892B (en) * | 2020-10-22 | 2024-03-12 | 常州大学 | Method for monitoring position and motion path of site robot |
CN112991742A (en) * | 2021-04-21 | 2021-06-18 | 四川见山科技有限责任公司 | Visual simulation method and system for real-time traffic data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109765245B (en) | Large-size display screen defect detection and positioning method | |
WO2017080102A1 (en) | Flying device, flying control system and method | |
CN104299215B (en) | The image split-joint method that a kind of characteristic point is demarcated and matched | |
CN106600644B (en) | Parameter correction method and device for panoramic camera | |
CN111583116A (en) | Video panorama stitching and fusing method and system based on multi-camera cross photography | |
CN112669394B (en) | Automatic calibration method for vision detection system | |
CN108776980A (en) | A kind of scaling method towards lenticule light-field camera | |
CN106017458B (en) | Mobile robot combined navigation method and device | |
CN104657982A (en) | Calibration method for projector | |
WO2001047285A1 (en) | Method and apparatus for calibrating projector-camera system | |
CN103268596B (en) | A kind of method for reducing picture noise and making color be near the mark | |
CN103886107A (en) | Robot locating and map building system based on ceiling image information | |
CN103065334A (en) | Color cast detection and correction method and device based on HSV (Hue, Saturation, Value) color space | |
CN104021527B (en) | Rain and snow removal method in image | |
CN102622739A (en) | Method for correcting non-uniformity of image of Bayer filter array color camera | |
CN108257187B (en) | Camera-projector system calibration method | |
US20220148223A1 (en) | Lens Calibration System | |
CN109887027A (en) | A kind of method for positioning mobile robot based on image | |
US11948286B2 (en) | Dark image enhancement | |
CN108958256A (en) | A kind of vision navigation method of mobile robot based on SSD object detection model | |
CN104749801B (en) | High Precision Automatic optical detecting method and system | |
CN114331835A (en) | Panoramic image splicing method and device based on optimal mapping matrix | |
CN109447946A (en) | A kind of Overhead optical cable method for detecting abnormality | |
CN106546196B (en) | A kind of optical axis real-time calibration method and system | |
CN104977154B (en) | Spatial light modulator defect classification method with sub-pixel structure |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190614 |
|
RJ01 | Rejection of invention patent application after publication |