CN109472778A - A kind of tall and slender structure appearance detecting method based on unmanned plane - Google Patents

A kind of tall and slender structure appearance detecting method based on unmanned plane Download PDF

Info

Publication number
CN109472778A
CN109472778A CN201811227998.3A CN201811227998A CN109472778A CN 109472778 A CN109472778 A CN 109472778A CN 201811227998 A CN201811227998 A CN 201811227998A CN 109472778 A CN109472778 A CN 109472778A
Authority
CN
China
Prior art keywords
image
unmanned plane
registration
works
matrix
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201811227998.3A
Other languages
Chinese (zh)
Other versions
CN109472778B (en
Inventor
谢海波
王培玉
刘晏玲
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.)
Hunan Tuoda Composition Monitoring Technology Co Ltd
Original Assignee
Hunan Tuoda Composition Monitoring 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 Hunan Tuoda Composition Monitoring Technology Co Ltd filed Critical Hunan Tuoda Composition Monitoring Technology Co Ltd
Priority to CN201811227998.3A priority Critical patent/CN109472778B/en
Publication of CN109472778A publication Critical patent/CN109472778A/en
Application granted granted Critical
Publication of CN109472778B publication Critical patent/CN109472778B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformations in the plane of the image
    • G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
    • G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
    • 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
    • G06T2207/10004—Still image; Photographic image
    • 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30181—Earth observation
    • G06T2207/30184—Infrastructure

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a kind of tall and slender structure appearance detecting method based on unmanned plane, the detection of high-rise structure appearance is carried out using unmanned plane, unmanned plane can be with Quick Acquisition image, and it is not influenced to carry out field process by geographical environment, propose the multi-site image mosaic technology for aiming at unmanned plane shooting image, works panoramic high-definition figure and image geometry information can be obtained by image processing techniques, obtain all visual defects information on high-rise structure surface, the defects of observing crackle, the erosion, leakage, peeling of submillimeter level information, meets the needs of Practical Project.

Description

A kind of tall and slender structure appearance detecting method based on unmanned plane
Technical field
The present invention relates to a kind of tall and slender structure appearance detecting method based on unmanned plane.
Background technique
It is towering that a kind of tall and slender structure appearance delection device and method (CN201510875509) provide two-axis table acquisition The method and apparatus of works surface disease.But this method, which has, to be had some limitations: the elevation angle of acquisition platform is greater than 45 ° It is serious to will lead to pattern distortion, image information missing；With platform and detectable substance distance increase, the limitations such as precision gradually decreases Property.
For unmanned plane in the gradually application of all trades and professions, it has changed into the tool of production that can fly, by unmanned aerial vehicle platform application It is detected to high-rise structure appearance, cooperates corresponding flight path planning method and image processing techniques, will greatly improve inspection Efficiency is surveyed, increases the scope of application, preferably solves tall and slender structure analyte detection problem.
Summary of the invention
The technical problem to be solved by the present invention is in view of the shortcomings of the prior art, provide a kind of based on the towering of unmanned plane Constructional appearance detection method obtains works panoramic high-definition figure and image geometry information by image processing techniques, obtains towering All visual defects information on works surface.
In order to solve the above technical problems, the technical scheme adopted by the invention is that: a kind of tall and slender structure based on unmanned plane Appearance detecting method, comprising the following steps:
1) according to monitoring accuracy γ requirement, camera transverse direction and short transverse resolution ratio αLAnd αH, image Duplication ol, determine Single image shoots length SLWith height SH；
2) by lens focus, size sensor, shooting length SLWith height SH, unmanned plane is calculated apart from works to be checked Minimum range d；
3) by unmanned plane during flying to the works lower right corner to be checked, and unmanned plane is adjusted to structure to be checked by distance measuring sensor Object distance, the two distance is d, and stops the longitude and latitude (Lon for obtaining this position1,Lat1) and elevation information (H1)；Then it flies To the works upper left corner to be checked, adjustment unmanned plane to works distance to be checked is d, obtains the longitude and latitude (Lon of this position2,Lat2) With elevation information (H2), thus to obtain the detection range (L, H) of works to be checked；
4) works to be checked is subjected to matrix pattern subregion, forms the location matrix M (n of unmanned plane1,n2)；Wherein, laterally It is respectively as follows: with short transverse flight range quantity
5) according to unmanned plane location matrix M (n1,n2) plan that flight path, opsition dependent matrix sequence successively give nobody automatically Machine sends aerial mission；First from the works upper left corner to be checked, that is, arrive first at matrix M (n1,n2) position, then arrive at M (n1,n2- 1), until arriving M (n1, 1), then successively reach M (n1- 1,1) and M (n2- 1,2), until last reach M (1,1), nothing It is man-machine to make a return voyage；Unmanned plane guides the position of each element mark in successively in-position matrix in RTK location information, and in element Designated position hovering is taken pictures, and image and its location information are stored；
6) the location matrix sequence high-definition image data and location information of UAV system shooting are obtained；
7) unmanned plane shooting image is mobile site shooting, i.e. multistation dot image data, using more optimal seams of weight iteration Image is carried out splicing by joint close method；
8) registration process is carried out to spliced image；
9) vector cad file determines images after registration pixel dimension, and length, the width of defect are determined according to pixel number And area.
In step 1), 20%≤ol≤50%.
In step 1), SL=γ × αL×(1-ol)；SH=γ × αH×(1-ol)。
In step 2), dL=f × SL/SSL；dW=f × SW/SSW；D=min (dL,dH)；Wherein, f is camera focus, SSL And SSHFor the length and height of camera sensor sensitive film.
In step 3),
H=H2-H1；Wherein ER is earth radius.
In step 7), include: by the specific implementation process that image carries out splicing
1) according to unmanned plane path dot sequency, stitching image I is treatedcBetween splicing sequence sorted in advance, wherein c table Show the quantity of image to be spliced；
2) SURF feature point detection algorithm is used, image characteristic point is searched；
3) adjacent image I to be registered is choseni, Ij, carried out using the geological information between RANSAC algorithm combination image special Sign point purification, obtains initial matching pair；
4) k Feature Points Matching pair, k >=4 are randomly selected from initial matching centering by successive ignition；Image is carried out thick Registration, and utilize formulaEstimate projective transformation matrix H, wherein v1= (x1,y1,z1), v2=(x2,y2,z2) it is Ii, IjIn corresponding characteristic point, enable z=1, pass through v1, v2Between homogeneous lineare transformation Relationship acquires element H in Hi,jValue, H is one 3 × 3 homogeneous matrix；
5) deformation algorithm is kept using structure, the image after rough registration is deformed, I is obtainedi', Ij'；
6) Graphcut algorithm is used, I is searchedi', Ij' overlapping region best seam；
7) formula is utilizedCalculate pixel in the characteristic point to seam in image overlapping region The distance of point, as registration error, wherein vfIndicate the characteristic point randomly selected, m indicates vfQuantity, vsIt indicates in seam Pixel, n indicate vsQuantity；
8) step 3)~step 7) is repeated, all possible image registration relationship is calculated, by the spy of registration error E (v) < δ Sign point usually takes δ=100, to all characteristic points remained to merging, calculates final image with this and match to remaining Quasi- relationship, as image to Ii, IjOptimal registration relationship；
9) step 8) is repeated, the optimal registration relationship between all adjacent images pair is calculated；
10) bundle adjustment algorithm is used, all image registration relationships are optimized；
11) Graphcut algorithm, finding step 10 are used) in final images after registration IcBetween overlapping region best seam Joint close；
12) image I is realized using multi-spectrum fusion technologycSeamless anastomosing and splicing.
Compared with prior art, the advantageous effect of present invention is that: the present invention utilize unmanned plane carry out tall and slender structure Beyond the region of objective existence sees detection, and unmanned plane can be with Quick Acquisition image, and is not influenced to carry out field process by geographical environment, proposes and aims at nothing The multi-site image mosaic technology of man-machine shooting image can obtain works panoramic high-definition figure and figure by image processing techniques As geological information, all visual defects information on high-rise structure surface are obtained, the crackle of submillimeter level is observed, corrodes, lets out The defects of leaking, peeling off information, meets the needs of Practical Project；The surface defect both macro and micro of tall and slender structure can also be obtained Evolving trend, and establish works health electronic archive system.
Detailed description of the invention
Fig. 1 is the flow chart of the high-rise structure appearance detection system provided by the invention based on unmanned plane；
Fig. 2 is unmanned plane acquisition platform provided by the invention；
100 be unmanned plane, 101 high-definition image acquisition modules, 102 rangefinders, 103GPS and RTK
Fig. 3 is unmanned plane shooting area matrix pattern block plan of the present invention；
Fig. 4 is multi-site merging algorithm for images block diagram；
Fig. 5 (1) is CAD diagram；Fig. 5 (2) is image before being registrated.
Specific embodiment
The present invention will be described in detail with reference to the accompanying drawing.
(1) unmanned plane high-definition image acquisition platform
Unmanned plane high-definition image acquisition platform includes the conventional unmanned plane module 100 such as unmanned plane during flying wing battery, centimetre Grade navigator fix RTK module 103, the image capture module 101 of high score rate (being greater than 20,000,000 pixels) imaging sensor, ranging pass Sensor 102, wireless transport module and control module；Wherein wireless transport module positions captured image information and RTK etc. Information is sent to ground control centre, and instruction is transferred to unmanned plane by control centre.Implementation process is as follows:
First according to the requirement of works detection accuracy, the camera focus of unmanned plane and its size sensor, image Duplication Determine unmanned plane course line transverse direction step-length SLWith vertical step-length SH, and unmanned plane is calculated apart from works minimum range d to be checked.
Then position is 1. put in the lower right corner, that is, figure that unmanned plane reaches Fig. 3 under the control operation of ground, adjustment unmanned plane is extremely Works distance reads the longitude and latitude (Lon of this point by unmanned plane positioning system until distance measuring sensor ranging is d1,Lat1) With elevation information (H1)；Then ground control operates and 2. puts position in the lower upper left corner, that is, figure for reaching Fig. 3, adjustment unmanned plane to knot Structure object distance reads the longitude and latitude (Lon of this point until distance measuring sensor ranging is d2,Lat2) and elevation information (H2), so far Detection zone, i.e., whole region as shown in Figure 3 are automatically determined.
Then according to detection zone and step-length, the grid dividing i.e. location matrix M (n of works is calculated1,n2), unmanned plane According to location matrix grid, flight path is planned.2 point i.e. M (n of the unmanned plane from Fig. 3 first1,n2) start to hover, pass through ground Central control system is automatically performed image focusing and shooting；Unmanned plane laterally moves to next mesh point i.e. Fig. 3 to the left automatically In black dot, hovering focusing shooting；It reaches last grid one and arranges M (n1, 1), then unmanned plane flies downwards along arrow To black dot, hovering focusing shooting, shuttle flight traversing to the right then along arrow direction, until 1. point in arrival figure It sets, hovering focusing shooting is finally maked a return voyage, and the work of this detection zone Image Acquisition terminates.
(2) image mosaic is handled
Shown in Fig. 3, unmanned plane has taken n according to the grid matrix of works1×n2High-definition image is opened, these images are nothings The a series of images of the man-machine shooting at close range on different location, different from the camera of fixed position, there is movement in these images Parallax, therefore multi-site image mosaic is the core work of technique.
It is arranged as matrix form image sequence, as shown in Figure 3 according to by these image files first.Specific stitching algorithm mistake Journey as shown in figure 4, algorithm the specific implementation process is as follows:
1) according to unmanned plane path dot sequency, stitching image I is treatedcBetween splicing sequence sorted in advance, wherein c table Show the quantity of image to be spliced；
2) SURF feature point detection algorithm is used, image characteristic point is searched；
3) adjacent image I to be registered is choseni, Ij, carried out using the geological information between RANSAC algorithm combination image special Sign point purification, obtains initial matching pair；
4) k Feature Points Matching pair, k >=4 are randomly selected from initial matching centering by successive ignition；Image is carried out thick Registration, and utilize formulaEstimate projective transformation matrix H, wherein v1= (x1,y1,z1), v2=(x2,y2,z2) it is Ii, IjIn corresponding characteristic point, enable z=1, pass through v1, v2Between homogeneous lineare transformation Relationship acquires element H in Hi,jValue, H is one 3 × 3 homogeneous matrix；
5) deformation algorithm is kept using structure, the image after rough registration is deformed, I is obtainedi', Ij'；
6) Graphcut algorithm is used, I is searchedi', Ij' overlapping region best seam；
7) formula is utilizedCalculate pixel in the characteristic point to seam in image overlapping region The distance of point, as registration error, wherein vfIndicate the characteristic point randomly selected, m indicates vfQuantity, vsIt indicates in seam Pixel, n indicate vsQuantity；
8) step 3)~step 7) is repeated, all possible image registration relationship is calculated, by the spy of registration error E (v) < δ Sign point usually takes δ=100, to all characteristic points remained to merging, calculates final image with this and match to remaining Quasi- relationship, as image to Ii, IjOptimal registration relationship；
9) step 8) is repeated, the optimal registration relationship between all adjacent images pair is calculated；
10) bundle adjustment algorithm is used, all image registration relationships are optimized；
11) Graphcut algorithm, finding step 10 are used) in final images after registration IcBetween overlapping region best seam Joint close；
12) image I is realized using multi-spectrum fusion technologycSeamless anastomosing and splicing.
(3) figure registration process
Previous step completes the splicing operation of high definition picture, and works panoramic picture has built up.Next The registration work of high-definition image, the purpose is to: high definition panorama image is cut and is calibrated so that in image works with Corresponding CAD polar plot coordinate, size are consistent, i.e., will be among works image registration to CAD diagram shape.
The method of registration be by changes in coordinates by each pixel transform in image the corresponding coordinate into CAD coordinate system Position, so that image becomes the image file for having geometric coordinate and dimension information similar with map.The specific steps of which are as follows:
A) above-mentioned high definition figure figure is subjected to gray processing, equalization and binary conversion treatment, obtains the side of works in image Boundary's location information.
B) works in panoramic picture and its practical shape and size have deviation, as shown in Fig. 5 (1) and Fig. 5 (2), figure Image data is exactly carried out geometric transformation by the purpose of shape registration, is realized and is corresponded with CAD diagram shape.By works side in image Works boundary coordinate corresponds in boundary's coordinate and CAD diagram, is changed by bilinearity, determines just for pixel each in image The two-wire mapping relations of true coordinate position (x, y), Fig. 5 (1) and Fig. 5 (2) are as follows:
C) by A and A ', corresponding 4 points such as B and B ' bring formula (9) into respectively, 8 coefficients of above-mentioned a-h can be found out.
E) it after by above-mentioned transformation, realizes that the coordinate of each pixel and practical CAD diagram paper coordinate correspond, completes image It is registrated with CAD diagram paper.
(4) Database Systems
After figure is registrated, the relationship of image pixel Yu structure actual size has been determined by vector cad file.And according to Number of pixels determines length, width and the area of defect.Image after completing registration can be by the way of atlas to structure The defect that beyond the region of objective existence is seen is marked and measures.And such as by these defective datas: data are recorded in position, length, width and area Among library, defect information database is formed.
Among image file after registration, to defect classification annotation and number, its positions and dimensions is measured, obtaining can be anti- Reflect the defect information database of works health status.History testing result is compared and analyzed, in conjunction with big data technology, is ground Study carefully and judge works health status and trend.Pipe for high-rise structure is supported and maintenance provides science, comprehensive data.

Claims (6)

1. a kind of tall and slender structure appearance detecting method based on unmanned plane, which comprises the following steps:
1) according to monitoring accuracy γ requirement, camera transverse direction and short transverse resolution ratio αLAnd αH, image Duplication ol, determine single width Image taking length SLWith height SH；
2) by lens focus, size sensor, shooting length SLWith height SH, it is minimum apart from works to be checked to calculate unmanned plane Distance d；
3) by unmanned plane during flying to the works lower right corner to be checked, and by distance measuring sensor adjust unmanned plane to works to be checked away from From the two distance is d, and stops the longitude and latitude (Lon for obtaining this position1, Lat1) and elevation information (H1)；Then flight to The works upper left corner is examined, adjustment unmanned plane to works distance to be checked is d, obtains the longitude and latitude (Lon of this position2, Lat2) and it is high Spend information (H2), thus to obtain the detection range (L, H) of works to be checked；
4) works to be checked is subjected to matrix pattern subregion, forms the location matrix M (n of unmanned plane1, n2)；Wherein, laterally and it is high Degree direction flight range quantity is respectively as follows:
5) according to unmanned plane location matrix M (n1, n2) plan that flight path, opsition dependent matrix sequence are successively sent out to unmanned plane automatically Send aerial mission；First from the works upper left corner to be checked, that is, arrive first at matrix M (n1, n2) position, then arrive at M (n1, n2- 1), until arriving M (n1, 1), then successively reach M (n1- 1,1) and M (n2- 1,2), until last reach M (1,1), unmanned plane It makes a return voyage；Unmanned plane guides the position of each element mark in successively in-position matrix in RTK location information, and in element assignment Position hovering is taken pictures, and image and its location information are stored；
6) the location matrix sequence high-definition image data and location information of UAV system shooting are obtained；
7) unmanned plane shooting image is mobile site shooting, i.e. multistation dot image data, using " more optimal sutures of weight iteration Image is carried out splicing by seam " method；
8) registration process is carried out to spliced image；
9) vector cad file determines images after registration pixel dimension, and length, width and the face of defect are determined according to pixel number Product.
2. the method according to claim 1, wherein in step 1), 20%≤ol≤50%.
3. the method according to claim 1, wherein in step 1), SL=γ × αL×(1-ol)；SH=γ × αH ×(1-ol)。
4. the method according to claim 1, wherein in step 2), dL=f × SL/SSL；dW=f × SW/SSW；d =min (dL, dH)；Wherein, f is camera focus, SSLAnd SSHFor the length and height of camera sensor sensitive film.
5. the method according to claim 1, wherein in step 3),
；
H=H2-H1；Wherein ER is earth radius.
6. the method according to claim 1, wherein image to be carried out to the specific reality of splicing in step 7) Now process includes:
1) according to unmanned plane path dot sequency, stitching image I is treatedcBetween splicing sequence sorted in advance, wherein c expression to The quantity of stitching image；
2) SURF feature point detection algorithm is used, image characteristic point is searched；
3) adjacent image I to be registered is choseni, Ij, characteristic point is carried out using the geological information between RANSAC algorithm combination image Purification, obtains initial matching pair；
4) k Feature Points Matching pair, k >=4 are randomly selected from initial matching centering by successive ignition；Image is slightly matched Standard, and utilize formulaEstimate projective transformation matrix H, wherein v1=(x1, y1, z1), v2=(x2, y2, z2) it is Ii, IjIn corresponding characteristic point, enable z=1, pass through v1, v2Between homogeneous lineare transformation close System, acquires element H in HI, jValue, H is one 3 × 3 homogeneous matrix；
5) deformation algorithm is kept using structure, the image after rough registration is deformed, I is obtainedi', Ij′；
6) Graphcut algorithm is used, I is searchedi', IjThe best seam of ' overlapping region；
7) formula is utilizedCalculate pixel in the characteristic point to seam in image overlapping region Distance, as registration error, wherein vfIndicate the characteristic point randomly selected, m indicates vfQuantity, vsIndicate pixel in seam Point, n indicate vsQuantity；
8) step 3)~step 7) is repeated, all possible image registration relationship is calculated, by the characteristic point of registration error E (v) < δ To remaining, δ=100 are usually taken, to all characteristic points remained to merging, final image registration is calculated with this and is closed System, as image to Ii, IjOptimal registration relationship；
9) step 8) is repeated, the optimal registration relationship between all adjacent images pair is calculated；
10) bundle adjustment algorithm is used, all image registration relationships are optimized；
11) Graphcut algorithm, finding step 10 are used) in final images after registration IcBetween overlapping region best seam；
12) image I is realized using multi-spectrum fusion technologycSeamless anastomosing and splicing.
CN201811227998.3A 2018-10-22 2018-10-22 Appearance detection method for towering structure based on unmanned aerial vehicle Active CN109472778B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811227998.3A CN109472778B (en) 2018-10-22 2018-10-22 Appearance detection method for towering structure based on unmanned aerial vehicle

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811227998.3A CN109472778B (en) 2018-10-22 2018-10-22 Appearance detection method for towering structure based on unmanned aerial vehicle

Publications (2)

Publication Number Publication Date
CN109472778A true CN109472778A (en) 2019-03-15
CN109472778B CN109472778B (en) 2021-11-09

Family

ID=65665742

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811227998.3A Active CN109472778B (en) 2018-10-22 2018-10-22 Appearance detection method for towering structure based on unmanned aerial vehicle

Country Status (1)

Country Link
CN (1) CN109472778B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110220918A (en) * 2019-06-17 2019-09-10 中广核核电运营有限公司 A kind of nuclear power plant containment shell appearance images acquisition methods and system based on unmanned plane
CN111008956A (en) * 2019-11-13 2020-04-14 武汉工程大学 Beam bottom crack detection method, system, device and medium based on image processing
CN111783539A (en) * 2020-05-30 2020-10-16 上海晏河建设勘测设计有限公司 Terrain measurement method, measurement device, measurement system and computer readable storage medium
EP4261776A1 (en) * 2022-03-29 2023-10-18 Rakuten Group, Inc. Information processing apparatus, method and program

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105354855A (en) * 2015-12-02 2016-02-24 湖南拓达结构监测技术有限公司 High-rise structure appearance detection device and method
CN107202793A (en) * 2017-05-16 2017-09-26 镇江市建科工程质量检测中心有限公司 A kind of detecting system and method for detecting external wall mass defect
CN107966709A (en) * 2017-11-15 2018-04-27 成都天麒科技有限公司 A kind of plant protection operation method based on laser radar mapping
WO2018076723A1 (en) * 2016-10-27 2018-05-03 上海华测导航技术股份有限公司 Unmanned aerial vehicle flight control system based on cors network differential positioning

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105354855A (en) * 2015-12-02 2016-02-24 湖南拓达结构监测技术有限公司 High-rise structure appearance detection device and method
WO2018076723A1 (en) * 2016-10-27 2018-05-03 上海华测导航技术股份有限公司 Unmanned aerial vehicle flight control system based on cors network differential positioning
CN107202793A (en) * 2017-05-16 2017-09-26 镇江市建科工程质量检测中心有限公司 A kind of detecting system and method for detecting external wall mass defect
CN107966709A (en) * 2017-11-15 2018-04-27 成都天麒科技有限公司 A kind of plant protection operation method based on laser radar mapping

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
薛爱新 等: "《近景摄影测量技术在斜拉桥桥塔健康监测中的应用研究》", 《中外公路》 *
马晔 等: "《无人机加载光学摄像及红外成像系统对海上特大桥塔索质量检测的运用技术》", 《公路交通科技》 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110220918A (en) * 2019-06-17 2019-09-10 中广核核电运营有限公司 A kind of nuclear power plant containment shell appearance images acquisition methods and system based on unmanned plane
CN111008956A (en) * 2019-11-13 2020-04-14 武汉工程大学 Beam bottom crack detection method, system, device and medium based on image processing
CN111008956B (en) * 2019-11-13 2024-06-28 武汉工程大学 Beam bottom crack detection method, system, device and medium based on image processing
CN111783539A (en) * 2020-05-30 2020-10-16 上海晏河建设勘测设计有限公司 Terrain measurement method, measurement device, measurement system and computer readable storage medium
EP4261776A1 (en) * 2022-03-29 2023-10-18 Rakuten Group, Inc. Information processing apparatus, method and program

Also Published As

Publication number Publication date
CN109472778B (en) 2021-11-09

Similar Documents

Publication Publication Date Title
US10777004B2 (en) System and method for generating three-dimensional robotic inspection plan
CN107514993B (en) The collecting method and system towards single building modeling based on unmanned plane
Loupos et al. Autonomous robotic system for tunnel structural inspection and assessment
CN109472778A (en) A kind of tall and slender structure appearance detecting method based on unmanned plane
CN106054929B (en) A kind of unmanned plane based on light stream lands bootstrap technique automatically
CN101419055B (en) Space target position and pose measuring device and method based on vision
CN104484648B (en) Robot variable visual angle obstacle detection method based on outline identification
CN106645205A (en) Unmanned aerial vehicle bridge bottom surface crack detection method and system
CN109358065B (en) Subway tunnel appearance detection method
RU2613735C2 (en) Method for detecting placement and location by virtual reference images
CN110334701B (en) Data acquisition method based on deep learning and multi-vision in digital twin environment
CN106444837A (en) Obstacle avoiding method and obstacle avoiding system for unmanned aerial vehicle
CN106767706A (en) A kind of unmanned plane reconnoitres the Aerial Images acquisition method and system of the scene of a traffic accident
CN108415453A (en) Unmanned plane tunnel method for inspecting based on BIM technology
JP2008014940A (en) Camera calibration method for camera measurement of planar subject and measuring device applying same
CN101033972A (en) Method for obtaining three-dimensional information of space non-cooperative object
CN114004977B (en) Method and system for positioning aerial data target based on deep learning
CN113340277B (en) High-precision positioning method based on unmanned aerial vehicle oblique photography
JP6877293B2 (en) Location information recording method and equipment
CN112288848A (en) Method for calculating engineering quantity through three-dimensional modeling of unmanned aerial vehicle aerial photography
Kim et al. As-is geometric data collection and 3D visualization through the collaboration between UAV and UGV
CN109342439A (en) Cable Structure appearance detecting method based on unmanned plane
CN113418448B (en) Fragment distribution detection system and method
CN116824079A (en) Three-dimensional entity model construction method and device based on full-information photogrammetry
CN113432529B (en) Seismic damage structure interlayer residual deformation detection method based on unmanned aerial vehicle camera shooting

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
GR01 Patent grant
GR01 Patent grant