CN104637051A - High-variable precision photographing method based on mobile carrier - Google Patents

High-variable precision photographing method based on mobile carrier Download PDF

Info

Publication number
CN104637051A
CN104637051A CN201410856512.8A CN201410856512A CN104637051A CN 104637051 A CN104637051 A CN 104637051A CN 201410856512 A CN201410856512 A CN 201410856512A CN 104637051 A CN104637051 A CN 104637051A
Authority
CN
China
Prior art keywords
image
color lump
camera
predetermined requirement
size
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
Application number
CN201410856512.8A
Other languages
Chinese (zh)
Inventor
方丽华
林韶文
刘汉君
李福鹏
张新明
陶熠昆
韦磊
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.)
Zhongshan Power Supply Bureau of Guangdong Power Grid Co Ltd
Zhejiang Guozi Robot Technology Co Ltd
Original Assignee
Zhongshan Power Supply Bureau of Guangdong Power Grid Co Ltd
Zhejiang Guozi Robot Technology 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 Zhongshan Power Supply Bureau of Guangdong Power Grid Co Ltd, Zhejiang Guozi Robot Technology Co Ltd filed Critical Zhongshan Power Supply Bureau of Guangdong Power Grid Co Ltd
Priority to CN201410856512.8A priority Critical patent/CN104637051A/en
Publication of CN104637051A publication Critical patent/CN104637051A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60—Control of cameras or camera modules
    • H04N23/67—Focus control based on electronic image sensor signals
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10024—Color image

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Geometry (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a high-variable precision photographing method based on a mobile carrier. The method comprises the steps of realizing an initially positioning of a camera in a preset way; utilizing a porous model as a camera optical model, adjusting a to-be-photographed device to the central position of an image, and carrying out the precision photographing by virtue of the variable and zooming adjusting way, so that the precision image collection can be realized by virtue of the technology such as the automatic light adaptation technology.

Description

The zoomable accurate image pickup method of a kind of height based on mobile vehicle
Technical field
The present invention relates to image taking technical field, particularly the zoomable accurate image pickup method of a kind of height based on mobile vehicle.
Background technology
Come by manually transferring crusing robot to gradually in the work of patrolling and examining of some large area, rugged surroundings or remote districts.For large-scale transformer station, except Daily Round Check, regular visit, more need expatriate personnel to carry out spy to transformer station at thunderstorm weather and patrol, and under exceedingly odious weather, be manually usually cannot patrol and examine transformer station, this is for the consideration to personal safety.
But, based on the style of shooting of this mobile vehicle, existing way is the mode of omnidistance remote control, namely staff controls the video camera on the The Cloud Terrace of crusing robot in the wings by the remote control equipment moment, thus regulating the focal length of video camera in real time, the sharpness of guarantee image and subject are positioned at the center of image.Although this style of shooting achieve in the presence of a harsh environment patrol and examine work, still need manually to manipulate shooting.
Summary of the invention
The present invention is directed to prior art above shortcomings, provide the zoomable accurate image pickup method of a kind of height based on mobile vehicle.The present invention is achieved through the following technical solutions:
The zoomable accurate image pickup method of height based on mobile vehicle, comprises step:
S1, the camera position be arranged on by the mode adjustment pre-seted on The Cloud Terrace, make object appear in the visual field of video camera;
S2, using the optical model of pin-point model as video camera, by identifying the center with location object being adjusted to camera coverage;
Whether the image that S3, judgement collect meets the requirements, if the image size of object does not meet predetermined requirement, then performs S4; If the size of image meets predetermined requirement, and the bright-dark degree of image does not meet predetermined requirement, then perform S5; If the size of image and bright-dark degree all meet, then perform S6;
S4, aperture multiple according to the adjusted size video camera of the image gabarit of object, whether the size of the image gabarit of the object after checking adjustment meets predetermined requirement, repeatedly perform S4, until the size of image gabarit of object after adjustment performs S5 after meeting predetermined requirement;
S5, zoom magnification according to the image averaging pixel value of object adjustment video camera, whether the image averaging pixel value of the object after checking adjustment meets predetermined requirement, repeatedly perform S5, until the image averaging pixel value of object after adjustment performs S6 after whether meeting predetermined requirement;
S6, video camera perform shooting work, and preserve the image of shooting.
Preferably, by identifying that the center with location, object being adjusted to camera coverage comprises:
A, to camera acquisition to image carry out color lump segmentation and feature extraction;
B, the environmental information utilizing the feature of color lump also to combine correspondence verify color lump, exclusive PCR color lump;
The color lump of c, calculating object is to the required The Cloud Terrace angle deflected of picture centre, by rotary head, object is positioned to picture centre, utilize the geometry location model of mobile vehicle to object to obtain the azimuth information of object relative to mobile vehicle, complete the location to target.
Preferably, to camera acquisition to image carry out color lump segmentation and comprise: utilize color threshold to the classification of image slices vegetarian refreshments, and adopt eight together with the marginal point of algorithm keeps track color lump simultaneously, the feature obtaining color lump comprises profile, girth, area and barycenter.
The present invention, without the need to human intervention, automatically snaps each image, and according to the difference of conditions and environment, is optimized process, ensures that shooting obtains optimum image.
Accompanying drawing explanation
Shown in Fig. 1 is the schematic diagram of pin-point model;
Shown in Fig. 2 to Fig. 7 is the process flow diagram of identification of the present invention and location;
Shown in Fig. 8 is radiation symmetric schematic diagram of the present invention;
Shown in Fig. 9 is that robot coordinate system of the present invention locates geometric model;
Shown in Figure 10 is camera coordinate system target imaging geometric model of the present invention.
Embodiment
Below with reference to accompanying drawing of the present invention; clear, complete description and discussion are carried out to the technical scheme in the embodiment of the present invention; obviously; as described herein is only a part of example of the present invention; it is not whole examples; based on the embodiment in the present invention, the every other embodiment that those of ordinary skill in the art obtain under the prerequisite not making creative work, all belongs to protection scope of the present invention.
For the ease of the understanding to the embodiment of the present invention, be further explained for specific embodiment below in conjunction with accompanying drawing, and each embodiment does not form the restriction to the embodiment of the present invention.
The zoomable accurate image pickup method of a kind of height based on mobile vehicle provided by the invention, comprises step:
S1, the camera position be arranged on by the mode adjustment pre-seted on The Cloud Terrace, make object appear in the visual field of video camera;
S2, using the optical model of pin-point model as video camera, by identifying the center with location object being adjusted to camera coverage;
As shown in Figure 1, pin-point model also claims perspective projection model, and figure midplane S is two-dimensional imaging plane (i.e. view plane), and C is the position (optical centre) of aperture.Point p (x, y) on S is three dimensions point P (X, Y, Z) is being projection in plane.F=d (S, C) is called the focal length of this optical system.C is the initial point of 3 d space coordinate, and optical axis direction is Z-direction, and view plane and its x vertical with optical axis, y-axis is parallel to three-dimensional X and Y-axis respectively.
Specific as follows by identifying the center with location, object being adjusted to camera coverage:
A, to camera acquisition to image carry out color lump segmentation and feature extraction, the present invention adopts Threshold Segmentation Algorithm, the color threshold utilizing off-line learning to obtain is classified to image slices vegetarian refreshments, and adopt the marginal point of eight connectivity algorithm keeps track color lump simultaneously, obtain profile information and the girth of color lump, area, the features such as barycenter, as shown in Figure 2, as shown in Fig. 3 to Fig. 7, by first color lump of the tracking shown in Fig. 3, then color lump internal point is got rid of by the chained list shown in Fig. 4, again by second the color lump edge of the tracking shown in Fig. 5, same gets rid of color lump internal point by Fig. 6 chained list, finally complete segmentation (as shown in Figure 7),
B, the environmental information utilizing the feature of color lump also to combine correspondence verify color lump, exclusive PCR color lump;
The color lump of c, calculating object is to the required The Cloud Terrace angle deflected of picture centre, by rotary head, object is positioned to picture centre, utilize the geometry location model of mobile vehicle to object to obtain the azimuth information of object relative to mobile vehicle, complete the location to target.
Radiation symmetric algorithm is widely used a kind of algorithm in recognition of face.The characteristic that it utilizes the straight line vertical with circumferentially each tangential direction put all to intersect at a point is to detect circle.To a series of n ∈ N, by checking colors, block edge point P positive and negative gradient direction development length n obtains p +and p -2 points, wherein N is the radius transformation range determined according to color lump information.Same calculating is done to all marginal points of color lump, statistics p ±set, finally can determine the shape of figure and the position in the center of circle.As shown in Figure 8, suppose p (i, j) on representative image color lump edge a bit, g (p) is the some p (i obtained based on image segmentation algorithm, j) edge gradient vector, the circular radius calculated according to color lump area is through the influence power scope [r of suitable expanded definition p (i, j) min, r max].Utilize formula (1) (2) to set up a simple voting mechanism to the point in picture element p (i, j) coverage and can obtain the most possible center of circle.
p = p ( i , j ) + round ( g ( p ( i , j ) ) | | g ( p ( i , j ) ) | | × r ) , r ∈ [ r min , r max ] , - - - ( 1 )
o(p)＝o(p)+1, (2)
In formula, round () is bracket function, and o (p) is the result that block edge point p (i, j) undertakies by the some p that formula (1) obtains voting of checking colors.The shape that all long messages in conjunction with color lump can obtain color lump judges and home position.When satisfying condition
max ( O ( p ) ) > 1 α × perimeter , - - - ( 3 )
Time, color lump is circular, and the center of circle is
o＝p(i,j),O(p(i,j))＝max(O(p)), (4)
In formula, perimeter is the girth of color lump, and α is setting parameter.
Shown in composition graphs 9, Figure 10, in camera coordinate system Figure 10, C point is video camera photocentre position, and H is the height of video camera photocentre distance robot reference plane, and P point is the volume coordinate point of target, and coordinate is P c=[x cy cz c] t, L is the distance of target to robot; Plane O' is video camera imaging plane, wherein O'(u 0, v 0) be picture centre, p (u, v) is the projection of target P on imaging plane, φ and ψ is respectively target relative to the position angle of camera optical axis and top rake.
Robot coordinate system's coordinate P wwith camera coordinates P crelation be expressed as x w y w z w 1 = T · x c y c z c 1 , (5)
In formula
T＝T 1·T 2·T 3,
T 1 = cos θ 0 sin θ 0 0 1 0 0 - sin θ 0 cos θ h 0 0 0 1 , T 2 = cos β - sin β 0 0 sin β cos β 0 0 0 0 1 0 0 0 0 1 , T 3 = cos α 0 sin α 0 0 1 0 0 - sin α 0 cos α 0 0 0 0 1 ,
T in formula 1for robot coordinate is tied to the transition matrix of The Cloud Terrace coordinate system, θ is robot top rake, and h is the vertical range AH of robot The Cloud Terrace A point to ground, T 2for The Cloud Terrace coordinate system rotates the transition matrix behind β angle around Z axis; T 3for the transition matrix of The Cloud Terrace coordinate system behind Y-axis rotation alpha angle.
According to pin-point model, single camera vision system can only collect the two-dimensional image information of target object, namely by image a bit can not to obtain in space more only, can only the ray in corresponding space.For obtaining the volume coordinate of target object, need by the movement limit of target object in a plane, the ray that in image, impact point is corresponding and the intersection point of plane are exactly the locus of target object.
Can be obtained by video camera pin-point model
ψ = arctan ( ( v - v 0 ) × dy f ) = arctan ( v - v 0 a y ) , - - - ( 7 )
In formula, dx and dy is the dimension of object of unit picture element horizontal direction and vertical direction in image, and f is focal length of camera, θ tiltfor video camera top rake, a xand a yfor intrinsic parameters of the camera, can be demarcated by Zhang Zhengyou method [16] or Tsai [17] method and obtain.
In camera coordinate system, with x cthe camera coordinates that can obtain target object for parameter is
In formula, l is the length of connecting rod AC between steering wheel axle and video camera.
Therefore, the coordinate conversion between robot coordinate system and camera coordinate system can be expressed as
Consider that target object rest on the ground, then z w=0, can x be solved c.P can be obtained according to formula (12) w=[x wy wz w] t.
Whether the image that S3, judgement collect meets the requirements, if the image size of object does not meet predetermined requirement, then performs S4; If the size of image meets predetermined requirement, and the bright-dark degree of image does not meet predetermined requirement, then perform S5; If the size of image and bright-dark degree all meet, then perform S6;
S4, aperture multiple according to the adjusted size video camera of the image gabarit of object, whether the size of the image gabarit of the object after checking adjustment meets predetermined requirement, repeatedly perform S4, until the size of image gabarit of object after adjustment performs S5 after meeting predetermined requirement;
S5, zoom magnification according to the image averaging pixel value of object adjustment video camera, whether the image averaging pixel value of the object after checking adjustment meets predetermined requirement, repeatedly perform S5, until the image averaging pixel value of object after adjustment performs S6 after whether meeting predetermined requirement;
S6, video camera perform shooting work, and preserve the image of shooting.
The above; be only the present invention's preferably embodiment, but protection scope of the present invention is not limited thereto, is anyly familiar with those skilled in the art in the technical scope that the present invention discloses; the change that can expect easily or replacement, all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection domain of claim.

Claims (3)

1., based on the zoomable accurate image pickup method of height of mobile vehicle, it is characterized in that, comprise step:
S1, the camera position be arranged on by the mode adjustment pre-seted on The Cloud Terrace, make object appear in the visual field of video camera;
S2, using the optical model of pin-point model as video camera, by identifying the center with location object being adjusted to camera coverage;
Whether the image that S3, judgement collect meets the requirements, if the image size of object does not meet predetermined requirement, then performs S4; If the size of image meets predetermined requirement, and the bright-dark degree of image does not meet predetermined requirement, then perform S5; If the size of image and bright-dark degree all meet, then perform S6;
S4, aperture multiple according to the adjusted size video camera of the image gabarit of object, whether the size of the image gabarit of the object after checking adjustment meets predetermined requirement, repeatedly perform S4, until the size of image gabarit of object after adjustment performs S5 after meeting predetermined requirement;
S5, zoom magnification according to the image averaging pixel value of object adjustment video camera, whether the image averaging pixel value of the object after checking adjustment meets predetermined requirement, repeatedly perform S5, until the image averaging pixel value of object after adjustment performs S6 after whether meeting predetermined requirement;
S6, video camera perform shooting work, and preserve the image of shooting.
2. the zoomable accurate image pickup method of the height based on mobile vehicle according to claim 1, is characterized in that, described by identifying that the center with location, object being adjusted to camera coverage comprises:
A, to camera acquisition to image carry out color lump segmentation and feature extraction;
B, the environmental information utilizing the feature of color lump also to combine correspondence verify color lump, exclusive PCR color lump;
The color lump of c, calculating object is to the required The Cloud Terrace angle deflected of picture centre, by rotary head, object is positioned to picture centre, utilize the geometry location model of mobile vehicle to object to obtain the azimuth information of object relative to mobile vehicle, complete the location to target.
3. the zoomable accurate image pickup method of the height based on mobile vehicle according to claim 2, it is characterized in that, described to camera acquisition to image carry out color lump segmentation comprise: utilize color threshold to classify to image slices vegetarian refreshments, and employing eight is together with the marginal point of algorithm keeps track color lump simultaneously, the feature obtaining color lump comprises profile, girth, area and barycenter.
CN201410856512.8A 2014-12-31 2014-12-31 High-variable precision photographing method based on mobile carrier Pending CN104637051A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410856512.8A CN104637051A (en) 2014-12-31 2014-12-31 High-variable precision photographing method based on mobile carrier

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410856512.8A CN104637051A (en) 2014-12-31 2014-12-31 High-variable precision photographing method based on mobile carrier

Publications (1)

Publication Number Publication Date
CN104637051A true CN104637051A (en) 2015-05-20

Family

ID=53215757

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410856512.8A Pending CN104637051A (en) 2014-12-31 2014-12-31 High-variable precision photographing method based on mobile carrier

Country Status (1)

Country Link
CN (1) CN104637051A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115118844A (en) * 2021-03-23 2022-09-27 北京小米移动软件有限公司 Mobile device
CN115150559A (en) * 2022-09-06 2022-10-04 国网天津市电力公司高压分公司 Remote vision system with acquisition self-adjustment calculation compensation and calculation compensation method

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103024276A (en) * 2012-12-17 2013-04-03 沈阳聚德视频技术有限公司 Positioning and focusing method of pan-tilt camera
CN103927878A (en) * 2014-04-10 2014-07-16 中海网络科技股份有限公司 Automatic snapshot device and method for illegal parking
CN104007762A (en) * 2014-05-28 2014-08-27 国家电网公司 Navigation method of electric power inspection robot

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103024276A (en) * 2012-12-17 2013-04-03 沈阳聚德视频技术有限公司 Positioning and focusing method of pan-tilt camera
CN103927878A (en) * 2014-04-10 2014-07-16 中海网络科技股份有限公司 Automatic snapshot device and method for illegal parking
CN104007762A (en) * 2014-05-28 2014-08-27 国家电网公司 Navigation method of electric power inspection robot

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
RUI GUO 等: "A Mobile Robot for Inspection of Substation Equipments", 《2010 1ST INTERNATIONAL CONFERENCE ON APPLIED ROBOTICS FOR THE POWER INDUSTRY》 *
杜玉霞 等: "《现代教育技术》", 31 August 2013, 清华大学出版社 *
杜鑫峰 等: "仿人足球机器人视觉系统快速识别与精确定位", 《浙江大学学报(工学版)》 *
许湘明 等: "变电站机器人视觉伺服系统研究", 《西南科技大学学报》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115118844A (en) * 2021-03-23 2022-09-27 北京小米移动软件有限公司 Mobile device
CN115150559A (en) * 2022-09-06 2022-10-04 国网天津市电力公司高压分公司 Remote vision system with acquisition self-adjustment calculation compensation and calculation compensation method

Similar Documents

Publication Publication Date Title
CN110297498B (en) Track inspection method and system based on wireless charging unmanned aerial vehicle
CN101004538B (en) Omnibearing vision sensor with no dead angle
CN103558910B (en) A kind of intelligent display system of automatic tracking head pose
CN105979147A (en) Intelligent shooting method of unmanned aerial vehicle
CN113658441B (en) High-flexibility variable-view-angle roadside sensing device and beyond-the-horizon sensing method for automatic driving
CN105205785B (en) A kind of orientable oversize vehicle operation management system and its operation method
CN102902884A (en) PTZ (pan/tilt/zoom) camera automatic positioning and angle calculating method
CN103513295A (en) Weather monitoring system and method based on multi-camera real-time shoot and image processing
CN106092054A (en) A kind of power circuit identification precise positioning air navigation aid
CN112215860A (en) Unmanned aerial vehicle positioning method based on image processing
Zhang et al. Robust inverse perspective mapping based on vanishing point
CN106504363A (en) A kind of airborne pair of light cruising inspection system stabilized platform automatic tracking method of intelligence
CN110132226A (en) The distance and azimuth angle measurement system and method for a kind of unmanned plane line walking
CN107644416A (en) A kind of real-time dynamic cloud amount inversion method based on ground cloud atlas
CN112947526B (en) Unmanned aerial vehicle autonomous landing method and system
CN113111715A (en) Unmanned aerial vehicle target tracking and information acquisition system and method
CN114905512A (en) Panoramic tracking and obstacle avoidance method and system for intelligent inspection robot
CN106851229A (en) A kind of method and system of the security protection intelligent decision based on image recognition
US11703820B2 (en) Monitoring management and control system based on panoramic big data
CN108132677B (en) Sunshade unmanned aerial vehicle control system and control method
CN110825098B (en) Unmanned aerial vehicle distribution network intelligent inspection system
CN111812659A (en) Iron tower posture early warning device and method based on image recognition and laser ranging
CN105516661A (en) Master-slave target monitoring system and method in combination of fisheye camera and PTZ camera
CN104637051A (en) High-variable precision photographing method based on mobile carrier
CN207281308U (en) Unmanned vehicle reconnaissance system and there is its unmanned vehicle

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into 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: 20150520