CN108759773A - A kind of monocular vision distance measuring method applied to robot crawl - Google Patents
A kind of monocular vision distance measuring method applied to robot crawl Download PDFInfo
- Publication number
- CN108759773A CN108759773A CN201810427821.1A CN201810427821A CN108759773A CN 108759773 A CN108759773 A CN 108759773A CN 201810427821 A CN201810427821 A CN 201810427821A CN 108759773 A CN108759773 A CN 108759773A
- Authority
- CN
- China
- Prior art keywords
- monocular vision
- crawl
- robot
- image
- center
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C11/00—Photogrammetry or videogrammetry, e.g. stereogrammetry; Photographic surveying
- G01C11/04—Interpretation of pictures
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C3/00—Measuring distances in line of sight; Optical rangefinders
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Electromagnetism (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Manipulator (AREA)
Abstract
The invention discloses a kind of monocular vision distance measuring method applied to robot crawl, this method goes out the marker that will be captured on object first with the image recognition that visual sensor collects and positions its center；Then adjustment monocular vision posture is controlled by robot, makes to indicate that object image coordinate is in the center of whole sub-picture in the image of monocular vision acquisition；Finally, the distance for capturing object to monocular vision coordinate origin can be calculated according to the size of mark object image on crawl object.It realizes monocular vision and replaces measurement of the binocular vision to object space coordinate, solve the problems such as binocular vision calibration, complicated calculations and structure design, the actual application ability of monocular vision technique is promoted, development cost is saved, and algorithm is realized simple, operation efficiency is high, real-time, using simplicity.
Description
Technical field
The present invention relates to technical field of robot vision, specifically a kind of monocular vision ranging applied to robot crawl
Method.
Background technology
Extensive use with robot technology in fields such as industry, agricultural, service industries, more and more robots open
The replacement that begins is accomplished manually related work, the sorting such as product on production line and feeding, discharge, the fruit picking in farm, in dining room
Serve a dish service etc..The diversification of demand makes requirement of the every field to robot higher and higher, how to allow robot in shape
In the different object of shape, size, accurately identifying positioning and capturing object becomes one of the problem of automatic field.In face of work
Industry manufacture and assembly production line, a variety of vision guide industrial robots position crawl technology and are applied in real time.Pass through vectoring aircraft
Device people captures target automatically, can efficiently solve identification and this problem of positioning of object, realize needed for every field
Full-automatic functional requirement improves production work efficiency.Wherein, the vision positioning of target object, which predicts its three dimensional space coordinate, is
The top priority of crawl is the basis that robot automation completes work in every below.And existing vision technique is double in utilization
Visually feel that camera cures with target relative position during obtaining target 3 d pose information, algorithm operation quantity is big and robustness
The shortcomings of poor, process realization is cumbersome, and which limits the uses of the Popularization And Development of vision robot, while also improving robot and answering
With technical difficulty, it is seen that visual token technical method is extremely important.Therefore for these disadvantage problems of solution, robot is improved
The reliability and stability of work reduce the restriction degree of its use condition, it is necessary to which redesigning one kind can effectively solve now
Deposit the ranging technology method of various disadvantages.
Invention content
It is above-mentioned to solve the purpose of the present invention is to provide a kind of monocular vision distance measuring method applied to robot crawl
The problem of being proposed in background technology.
To achieve the above object, the present invention provides the following technical solutions：
A kind of monocular vision distance measuring method applied to robot crawl, including following steps：
Step (1)：The image recognition that visual sensor collects goes out the marker that will be captured on object；
Step (2)：Identify the center of witness marker object image；
Step (3)：Robot control adjustment monocular vision posture, makes mark object image in the image of monocular vision acquisition sit
Center of the mark in whole sub-picture；
Step (4)：Calculate the size for indicating object image on crawl object；
Step (5)：The distance for capturing object to monocular vision coordinate origin is calculated using camera national forest park in Xiaokeng.
As a further solution of the present invention：The concrete methods of realizing of the step (1) is：First with solid in robot
Image in fixed monocular vision camera shooting, collecting operative scenario, by template matches identification technology, searching, which identifies, to be wanted
The object of crawl obtains the substantially place orientation of crawl object.
As further scheme of the invention：The concrete methods of realizing of the step (2) is：Step (1) is identified
Crawl object on the Marker Identity that pastes come out, and the center of witness marker object image.
As further scheme of the invention：The concrete methods of realizing of the step (3) is：Robot carries monocular and regards
Feel that camera motion, control adjustment monocular vision posture change marker picture centre and be in the whole secondary center for generating image,
I.e. the center of marker is on the optical axis of vision camera.
As further scheme of the invention：The concrete methods of realizing of the step (4) is：Calculation flag object is square vertically
To top edge to the number of pixels of the center, i.e., the length between them has reacted monocular vision camera and has captured object
Apart from size degree.
As further scheme of the invention：The step (5) is specifically：Using camera pinhole imaging system principle model,
According to the geometrical correspondence between them, the distance for calculating crawl object to monocular vision coordinate origin is derived.
Compared with prior art, the beneficial effects of the invention are as follows：1, the object dimensional spatial information of robot crawl is realized
Monocular vision measures, and improves the accuracy and reliability of vision measurement technology, effectively solves to ask existing for binocular vision technology
Topic can simplify the process of crawl object dimensional space measurement, and the actual application value for improving monocular vision measuring technique is horizontal,
System development design cost is reduced, the development in vision robotics field is pushed.2, procedure it is easy, it is versatile, meter
Calculation amount is small, computational efficiency is high, accuracy of measurement is high.
Description of the drawings
Fig. 1 is monocular geometrical relationship figure of the present invention；
Fig. 2 is monocular ranging technology flow chart of the present invention.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Please refer to Fig.1~2, in the embodiment of the present invention, a kind of robot monocular vision measuring technique method, including it is following several
A step：
Step (1)：The image recognition that visual sensor collects goes out the marker that will be captured on object；
Step (2)：Identify the center of witness marker object image；
Step (3)：Robot control adjustment monocular vision posture, makes mark object image in the image of monocular vision acquisition sit
Center of the mark in whole sub-picture；
Step (4)：Calculate the size for indicating object image on crawl object；
Step (5)：The distance for capturing object to monocular vision coordinate origin is calculated using camera national forest park in Xiaokeng.
The step (1) is first with the figure in fixed monocular vision camera shooting, collecting operative scenario in robot
Picture, by template matches identification technology, searching identifies object to be crawled, obtains the substantially place orientation of crawl object,
Identify that its marker reduces the range that identification is found for later step.
The step (2) makes a kind of special sign thing, and marker is tightly attached to crawl body surface, then identification
The Marker Identity pasted on the crawl object gone out comes out, and positions the center of extraction mark object image；
Step (3) robot carries monocular vision camera motion, and the posture of control adjustment monocular vision changes mark
Image space of the object in monocular camera, it is final so that marker picture centre is in the whole secondary center for generating image, i.e.,
The center of marker is on the optical axis of vision camera；
On step (4) the calculation flag object vertical direction each edge to the center number of pixels, i.e., they it
Between respective distance length, reacted monocular vision camera and crawl object at a distance from size degree；
The step (5) utilizes camera pinhole imaging system principle model, is constituted according to each length distance image-forming principle
Geometrical correspondence, derives the distance for calculating crawl object to monocular vision coordinate origin, and last each length computation obtains
The information arrived is averaged value, finally obtain monocular vision measurement as a result, this completes monocular vision measuring technique method mistakes
Journey.
This method realizes that monocular vision ranging process is as follows：
For this method as a kind of monocular vision ranging technology, being gone out using the image recognition that visual sensor collects will
It captures the marker on object and positions its central point q, adjustment monocular vision posture is controlled by robot, monocular vision is made to adopt
Indicate that the center that object image coordinate is in whole sub-picture is point o in the image of collection so that marker picture centre is in camera
On optical axis, it is op to calculate mark size of the object image edge far from center on crawl object in the vertical direction, sets camera focus
For f, the pitch angle of camera is α, is h apart from ground level, capture the mark center point q of object to camera lens distance be d, most
The geometrical relationship figure between each parameter is obtained by camera pinhole imaging system principle model afterwards, according to relationship model figure can set β=α+
γ obtains tg β=h/d, tg γ=op/f, final in this way to obtain d=h/tg (γ+α)=h/g [α+arctg (op/f)], because should
In formula other than depth distance d is unknown quantity, remaining is all known quantity, you can calculates crawl object to monocular by formula derivation
The three dimensional space coordinate information of visual coordinate system origin, as crawl target object motion control below provide space bit confidence
Breath.
As described above, the present invention can be better realized.
It is obvious to a person skilled in the art that invention is not limited to the details of the above exemplary embodiments, Er Qie
In the case of without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Profit requires rather than above description limits, it is intended that all by what is fallen within the meaning and scope of the equivalent requirements of the claims
Variation is included within the present invention.Any reference signs in the claims should not be construed as limiting the involved claims.
In addition, it should be understood that although this specification is described in terms of embodiments, but not each embodiment is only wrapped
Containing an independent technical solution, this description of the specification is merely for the sake of clarity, and those skilled in the art should
It considers the specification as a whole, the technical solutions in the various embodiments may also be suitably combined, forms those skilled in the art
The other embodiment being appreciated that.
Claims (6)
1. a kind of monocular vision distance measuring method applied to robot crawl, which is characterized in that including following steps：
Step (1)：The image recognition that visual sensor collects goes out the marker that will be captured on object；
Step (2)：Identify the center of witness marker object image；
Step (3)：Robot control adjustment monocular vision posture, makes in the image of monocular vision acquisition at mark object image coordinate
In the center of whole sub-picture；
Step (4)：Calculate the size for indicating object image on crawl object；
Step (5)：The distance for capturing object to monocular vision coordinate origin is calculated using camera national forest park in Xiaokeng.
2. the monocular vision distance measuring method according to claim 1 applied to robot crawl, which is characterized in that the step
Suddenly the concrete methods of realizing of (1) is：First in fixed monocular vision camera shooting, collecting operative scenario in robot
Image, by template matches identification technology, searching identifies object to be crawled, obtains the substantially place side of crawl object
Position.
3. the monocular vision distance measuring method according to claim 1 applied to robot crawl, which is characterized in that the step
Suddenly the concrete methods of realizing of (2) is：The Marker Identity pasted on the crawl object that step (1) identifies is come out, and positions mark
The center of will object image.
4. the monocular vision distance measuring method according to claim 1 applied to robot crawl, which is characterized in that the step
Suddenly the concrete methods of realizing of (3) is：Robot carries monocular vision camera motion, and control adjustment monocular vision posture changes mark
Will object image center is in the whole secondary center for generating image, i.e. the center of marker is on the optical axis of vision camera.
5. the monocular vision distance measuring method according to claim 1 applied to robot crawl, which is characterized in that the step
Suddenly the concrete methods of realizing of (4) is：Calculation flag object vertical direction top edge to the center number of pixels, i.e., between them
Length, reacted monocular vision camera and crawl object at a distance from size degree.
6. the monocular vision distance measuring method according to claim 2 applied to robot crawl, which is characterized in that the step
Suddenly (5) are specifically：Using camera pinhole imaging system principle model, according to the geometrical correspondence between them, derivation, which calculates, grabs
Take object to the distance of monocular vision coordinate origin.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810427821.1A CN108759773A (en) | 2018-05-07 | 2018-05-07 | A kind of monocular vision distance measuring method applied to robot crawl |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810427821.1A CN108759773A (en) | 2018-05-07 | 2018-05-07 | A kind of monocular vision distance measuring method applied to robot crawl |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108759773A true CN108759773A (en) | 2018-11-06 |
Family
ID=64009861
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810427821.1A Pending CN108759773A (en) | 2018-05-07 | 2018-05-07 | A kind of monocular vision distance measuring method applied to robot crawl |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108759773A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109724560A (en) * | 2018-12-17 | 2019-05-07 | 南京理工大学 | A kind of monocular range-measurement system based on Inertial Measurement Unit |
| CN109764807A (en) * | 2019-01-14 | 2019-05-17 | 广东原点智能技术有限公司 | The 2D visible detection method and detection system of engine cylinder body location position |
| CN109855602A (en) * | 2019-01-14 | 2019-06-07 | 南通大学 | Move the monocular visual positioning method under visual field |
| CN110363905A (en) * | 2019-07-31 | 2019-10-22 | 浪潮金融信息技术有限公司 | The method of the positioning of vending machine detection zone and detection shipment based on sliding rail |
-
2018
- 2018-05-07 CN CN201810427821.1A patent/CN108759773A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109724560A (en) * | 2018-12-17 | 2019-05-07 | 南京理工大学 | A kind of monocular range-measurement system based on Inertial Measurement Unit |
| CN109764807A (en) * | 2019-01-14 | 2019-05-17 | 广东原点智能技术有限公司 | The 2D visible detection method and detection system of engine cylinder body location position |
| CN109855602A (en) * | 2019-01-14 | 2019-06-07 | 南通大学 | Move the monocular visual positioning method under visual field |
| CN109764807B (en) * | 2019-01-14 | 2021-03-05 | 广东原点智能技术有限公司 | 2D visual detection method and detection system for engine cylinder position calibration |
| CN110363905A (en) * | 2019-07-31 | 2019-10-22 | 浪潮金融信息技术有限公司 | The method of the positioning of vending machine detection zone and detection shipment based on sliding rail |
| CN110363905B (en) * | 2019-07-31 | 2021-09-17 | 浪潮金融信息技术有限公司 | Vending machine detection area positioning and goods discharge detection method based on sliding rail |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108759773A (en) | A kind of monocular vision distance measuring method applied to robot crawl | |
| CN110426051B (en) | Lane line drawing method and device and storage medium | |
| CN108109174A (en) | A kind of robot monocular bootstrap technique sorted at random for part at random and system | |
| CN109029257A (en) | Based on stereoscopic vision and the large-scale workpiece pose measurement system of structure light vision, method | |
| CN110599541A (en) | Method and device for calibrating multiple sensors and storage medium | |
| CN107917700B (en) | Small-amplitude target three-dimensional attitude angle measurement method based on deep learning | |
| CN110675453B (en) | Self-positioning method for moving target in known scene | |
| CN209174850U (en) | The device of big packet collector nozzle is positioned using machine vision | |
| CN108827250A (en) | A kind of robot monocular vision ranging technology method | |
| CN104457758B (en) | Video-acquisition-based Visual Map database establishing method and indoor visual positioning method using database | |
| CN112949478A (en) | Target detection method based on holder camera | |
| CN107798720A (en) | A kind of method for drawing map and its device, mobile terminal | |
| CN206863817U (en) | Camera review automatic acquisition device and camera calibration systems | |
| Momeni-k et al. | Height estimation from a single camera view | |
| CN103852060A (en) | Visible light image distance measuring method based on monocular vision | |
| CN111307046B (en) | Tree height measuring method based on hemispherical image | |
| CN110889829A (en) | Monocular distance measurement method based on fisheye lens | |
| CN111028271A (en) | Multi-camera personnel three-dimensional positioning and tracking system based on human skeleton detection | |
| CN104048649B (en) | A kind of multi-view images and the rapid registering method of threedimensional model | |
| CN106504287A (en) | Monocular vision object space alignment system based on template | |
| CN108180825B (en) | A kind of identification of cuboid object dimensional and localization method based on line-structured light | |
| CN114492070A (en) | High-precision mapping geographic information virtual simulation technology and device | |
| CN110532865A (en) | Spacecraft structure recognition methods based on visible light and laser fusion | |
| CN110424754B (en) | Butt joint method and system for wall building robot | |
| CN104200469B (en) | Data fusion method for vision intelligent numerical-control system |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181106 |