WO2008056427A1 - Procédé de mesure de forme tridimensionnelle et dispositif correspondant - Google Patents

Procédé de mesure de forme tridimensionnelle et dispositif correspondant Download PDF

Info

Publication number
WO2008056427A1
WO2008056427A1 PCT/JP2006/322687 JP2006322687W WO2008056427A1 WO 2008056427 A1 WO2008056427 A1 WO 2008056427A1 JP 2006322687 W JP2006322687 W JP 2006322687W WO 2008056427 A1 WO2008056427 A1 WO 2008056427A1
Authority
WO
WIPO (PCT)
Prior art keywords
laser
plane
image
laser plane
projector
Prior art date
Application number
PCT/JP2006/322687
Other languages
English (en)
Japanese (ja)
Inventor
Hiroshi Kawasaki
Ryo Furukawa
Original Assignee
Techno Dream 21 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 Techno Dream 21 Co., Ltd. filed Critical Techno Dream 21 Co., Ltd.
Priority to JP2008542981A priority Critical patent/JP5307549B2/ja
Priority to PCT/JP2006/322687 priority patent/WO2008056427A1/fr
Publication of WO2008056427A1 publication Critical patent/WO2008056427A1/fr

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
    • G01B11/25—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures by projecting a pattern, e.g. one or more lines, moiré fringes on the object

Definitions

  • the present invention relates to a three-dimensional shape measuring apparatus that obtains distance information to a target object using a laser beam and an imaging apparatus, and in particular, there is a relationship between a single imaging apparatus and the imaging apparatus. It is not related to a measuring device that uses a laser oscillator that is not known (no external calibration is performed). Also, it relates to a method for measuring the three-dimensional shape of a target object using the device.
  • Patent Document 1 a method that easily acquires the positional relationship between the imaging device and the light source by attaching a marker to the light source and simultaneously photographing the marker with the imaging device.
  • Non-Patent Documents 2 and 3 a method that prepares a separate device such as a frame or a pentagrammer. Unnecessary and, moreover or taking a marker at the same time, separately If the positional relationship between the imaging device and the light source can be acquired without preparing a device, the convenience will increase and the mechanism will be simple, easy to carry, low cost, and so on. The disadvantages of the active measurement method can be solved. Therefore, a technique for three-dimensional reconstruction without using "ma" (Patent Document 4, Non-Patent Document 5) has been proposed.
  • the method proposed here is a method that directly implements Euclidean reconstruction by the above methods (Patent Document 4, Non-Patent Document 5,).
  • the coplanar information power is a method that realizes projective restoration, so the opportunity for finding a solution is greatly increased, and there are fewer restrictions on how to irradiate the line laser.
  • Non-Patent Document 2 CW Chu, S.
  • Patent Document 4 PCT JP2004 / 011406 ( Non-Patent Document 5 Hiroshi Kawasaki, Ryo Furukawa, Yasuaki Nakamura: “Proposal for 3D Restoration Method by Self-Calibration of Line Lasers—Minimum Configuration System in Active 3D Measurement-”, Image Recognition 'Understanding Symposium pp.
  • the observed coplanarity is the scene composed of planes.
  • the number of conditions obtained is not so large, and it is difficult to obtain a small-dimensional solution space.
  • a line laser was applied to the scene and the laser reflection was observed with a camera, laser reflection was observed. All the points are coplanar.
  • a very large amount of coplanarity information can be easily obtained.
  • By collecting coplanarity information while irradiating the line laser there is a possibility that 3D reconstruction can be performed in general scenes without any restrictions.
  • position information that has no remaining degrees of freedom or position information that leaves only one degree of freedom of isotropic scaling.
  • the laser projector is configured by attaching two laser planes so that they are orthogonal to each other, three-dimensional reconstruction without scaling freedom is possible. Also, the laser projectors are parallel. If the distance is accurately investigated, Euclidean reconstruction without scaling indefiniteness is possible.
  • FIG. 1 shows the intersection of a laser beam detected in the past and a laser beam detected this time from an image obtained by irradiating a target object with a planar laser beam from a laser device and shooting it with an imaging device.
  • Fig. 2 is an explanatory diagram for explaining how to obtain a linear equation from the intersection.
  • the perspective 3 ⁇ 4 shadow is represented by the composition of the homodarafi transformation shown in the figure and the orthographic projection.
  • Figure 3 shows the laser plane through which the line laser irradiated from the laser device passes, the straight line representing the line-of-sight direction, the image plane taken by the imaging device, and their relationship.
  • Figure 4 shows the two line lasers radiated from the laser device, the lines drawn on the target object and their intersections, and those detected on the image plane taken by the imager.
  • Fig. 5 is a diagram showing the intersections and the straight lines that pass through the intersections. The figure shows the multiple intersections where multiple line lasers emitted from the laser device intersect on the target object. Is shown on the image plane taken by the imaging device.
  • Fig. 6 shows the case where a marker such as an LED is attached near the exit of the line laser light source.
  • Fig. 7 shows the irradiation of multiple line lasers by combining multiple line laser light sources in an orthogonal state.
  • Fig. 8 shows the irradiation of multiple line lasers by combining multiple line laser light sources in parallel.
  • Fig. 9 shows the irradiation from the laser device. From the reflection position of each laser curve on the target object, find the intersection of the curves, and use the intersection group to determine the laser plane parameters and the three-dimensional position of the intersection.
  • Fig. 10 shows the result of the projection restoration obtained by the proposed method when the relative positional relationship is known, such as the laser plane being orthogonal. Is a flowchart showing an algorithm that upgrades to Euclidean reconstruction using the constraints.
  • Fig. 11 shows the estimation of unknown laser plane parameters using known laser plane parameters.
  • Figure 2 is a flowchart showing an algorithm that restores dense three-dimensional points by iterative processing.
  • Fig. 1 shows a laser track with white and black lattice patterns combined with a sine curve.
  • FIG. 13 shows an example of restoring the 3D shape using the proposed method.
  • Fig. 13 shows the restored laser curve for reference.
  • Fig. 14 shows an example of 3D shape restoration using the proposed method.
  • Fig. 14 shows the correct shape for reference in the restored laser curve.
  • Fig. 13 is a diagram drawn from a different perspective:
  • Fig. 15 shows a black and white grid pattern in a cubic shape with a maggot shape.
  • Fig. 16 is an example of 3D shape reconstruction using the proposed method.
  • Fig. 16 shows an example of the restored laser curve.
  • Fig. 17 shows an example of 3D shape restoration using the proposed method.
  • Fig. 17 shows the restored laser curve for reference.
  • the correct shape is displayed in an overlapping manner.
  • Fig. 16 is a diagram drawn by changing the viewpoint.
  • Fig. 16 is a diagram drawn by changing the viewpoint.
  • FIG. 18 is a square box shape.
  • Fig. 19 shows an example of three-dimensional shape reconstruction using only the laser curve shown in Fig. 18 by the proposed method.
  • Figure 20 shows the shape seen from the front.
  • Fig. 20 shows an example of the three-dimensional shape restoration of the laser curve only using the proposed method.
  • Fig. 20 shows the shape seen from directly above.
  • Fig. 21 shows the result of restoring the three-dimensional shape of Fig. 18 using the proposed method, and the dense restoration method proposed in this patent for the laser curve group that was not used for restoration.
  • Using the densely 3 Figure 2-2 shows the result of dimension restoration as seen from directly above.
  • Figure 2 shows the result of restoring the three-dimensional shape of Figure 18 using the proposed method, and a laser curve that was not used for restoration.
  • This figure shows the result of dense three-dimensional reconstruction using the dense restoration method proposed in this patent for the group, viewed from diagonally above.
  • Figure 23 shows a photograph of the cross-laser being irradiated onto paper shaped like a sine curve.
  • Figure 24 shows the shape of the sine curve shown in Figure 23. This figure shows the result of dense restoration using the dense restoration method proposed in this patent.
  • the metric constraint is used to upgrade to Euclidean restoration (Fig. 10).
  • the upgrade is the projection restored 3 This refers to estimating position information that has no remaining degrees of freedom or position information that leaves only one degree of freedom of isometric expansion / contraction from position information in the dimension space and another constraint equation.
  • a 3D shape can be obtained by projective restoration or Euclidean restoration, a more precise shape restoration can be realized (Fig. 11).
  • ⁇ ⁇ 1, 2, 3 ⁇ , ⁇ 1, 4, 5, 6 ⁇ ,.
  • ⁇ ⁇ , ⁇ , ⁇ 2 are on the same plane
  • ⁇ , ⁇ 3 , ⁇ 5 are different Index each element of the set ((which is also a set) and index (1 —
  • Coplanarity assumption of 4 points or more 'For any, J' In other words, coplanarity of 3 points or less is not included in the condition of coplanarity.
  • “Multiple coplanarity assumptions” The reason for this is that points that are included in only one plane do not become 3D reconstruction information. In the system shown in Fig. 1, this is a point set, which is an image plane of multiple laser reflection curves. Think only of the intersection (corresponding. ⁇ 'non-collinearity assumption'
  • ⁇ is a visual cone in perspective projection as shown in Fig. 2. Is a 3D Homography that transforms the region of into a rectangular parallelepiped region with a plane parallel to the image plane. This transformation moves the original point to a point at infinity in two axes. the combination of the projection becomes perspective projection shadow equivalent. Moreover, ⁇ a ⁇ ⁇ is a three-dimensional Homography that changes the coordinates only. I, projection by perspective projection camera """is three-dimensional due to O
  • the solution of the SRC, SRCO problem has more than 4 degrees of freedom. This is the solution of ⁇ 0, ⁇ , force S, SRC0, for any real number 1 , 2 , 3 , 4 Is also a solution
  • ⁇ is a set of indices of the points that exist on the plane 7 ⁇ '.
  • the plane 71, equation ' The plane parallel to the two axes cannot be expressed in this notation, but the assumption of noncollinear 3 ⁇ 4 indicates that the plane is not parallel to the two axes for any J in the SRC0 problem. Does not lose generality.
  • the coplanarity condition can be expressed as follows.
  • the solution of the SRC problem has 4 or more degrees of freedom. There are enough points for the number of planes in the scene (ie M > 3 l n l) and the planes ⁇ j ⁇ 1 ⁇ ) in the general position If there is, the rank of "is Rank ()
  • SRC or SRC0 has an angle of 4 degrees of freedom.
  • singular value decomposition can be used.
  • ⁇ is: a column matrix
  • the solution method can be left as it is.
  • five or more singular value forces S 0 When the value is close to, the above solution becomes unstable. In such a case, it may be possible to adopt a linear space of 5 or more dimensions as the solution space.
  • the ratio of the 4th smallest singular value to the 5th smallest singular value '' 1 ⁇ 1 1 5 / ⁇ ⁇ ⁇ 1 4 can be an indicator of the instability of the solution. If this value is close to 1, the solution is unstable ⁇ (Metric constrained -Cridic restoration)
  • the solution of the SRC problem obtained in the previous section is a projective restoration with four degrees of freedom by ',...,. can not be solved by only the conditions of coplanarity:.
  • additional constraints there: is, part of the freedom of Oconnection solutions Besides, 'or if all can be eliminated
  • 'or if all can be eliminated as an example, let us consider the case of obtaining a coplanarity condition by projecting a line laser.
  • the free variable was determined by.
  • the projection was performed so that the Euclidean distance between Xi and 1J was minimized from the space i " 1 " dimension of the plane parameter X to the solution space (4 dimensions) of the linear equation.
  • the laser curve should be in a lattice pattern.
  • Lower left force Decide to move to the upper right, and give an initial value close to that movement
  • a method of determining parameters is conceivable.
  • one of the four free variables of 1 is fixed and nonlinear optimization is performed for the remaining three variables.
  • two line lasers are mounted in parallel on the projector.
  • the parallelism of the planes can be added as a condition.
  • intersection of the laser plane restored in this way and the line connecting the center of the camera and the point on the laser curve on the image is It is derived as a 3D point of the laser reflection position.
  • Figure 1 2 shows the data when CG data is generated.
  • the shape of this data is' The curved surface that intersects the shape of the laser plane and the shape when the laser projector is projected from various angles on the curved surface was calculated.
  • the lattice pattern in the figure the lattice The boundary of represents a curve like this. The position of the intersection of these boundaries on the image plane is
  • the reflection curve image is shown in Fig. 5.
  • the laser was projected 20 times.
  • the direction of the projector was 5 times, starting from the upper right, upper left, lower right, and lower left.
  • the number of intersections is 40 and the number of intersections is 613.
  • the restoration results and the correct shape are shown in Fig. 16 and Fig. 17. Even in this data, almost correct shapes were obtained as restoration results.
  • a reflector can be used as a marker attached to the laser device.
  • a strobe light or a halogen lamp may be used as long as it is a pattern that passes through a plane in space.
  • a video projector may be used.
  • it may be a shadow created by a powerful light source (eg laser, strobe light, halogen lamp, sunlight, etc.).
  • Imaging device As an imaging device, a camera, digital camera, CCD, video camera, etc. are convenient. If a filter that matches the wavelength of the laser is attached, it will be easier to detect only the laser. If a filter that can control the transmission band is used, multiple lasers that are not so far apart can be efficiently identified by measuring while changing the transmission band, which is highly effective.
  • the above processing can be performed at high speed by using a computer and is convenient. Also, if a computer capable of parallel processing is used, it can be processed at high speed and is convenient. INDUSTRIAL APPLICABILITY According to the present invention, a laser device composed of a line laser is held and moved freely while the reflected light of the line laser is photographed with a camera and the image is processed. The external calibration of the laser device position is performed online, and as a result, the three-dimensional coordinates of the laser reflection position are calculated one after another.
  • This method solves in two stages: reconstruction with 4 degrees of freedom (typically 4 degrees of freedom) and satisfaction of metric constraints, so there is no need to limit the method of line laser irradiation (that is, You can create a constraint from any combination of irradiation patterns, either with a grid pattern, or 'without having to restrict all intersection points to be detected. Therefore, the same calculation algorithm can be used. For this reason, the practical degree of freedom of laser combinations in laser projectors is greatly increased, resulting in various types of laser projection. It becomes possible to create a device, and the degree of freedom in producing a hand-held 3D measuring device is greatly increased.
  • the measuring device It is also possible to accurately measure a narrow space where it is difficult to enter, and it is difficult to move it by itself. '-In addition, as in a nuclear reactor, people cannot enter and complex machines are inserted. It is also difficult to use in situations where fine control is difficult.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Length Measuring Devices By Optical Means (AREA)

Abstract

L'invention concerne un procédé de restauration par projection ou de restauration euclidienne d'une scène à partir d'une série d'images au moyen d'une équation linéaire, telles que celles photographiées en utilisant un projecteur constitué par un ou plusieurs lasers à balayage linéaire et un appareil photographique immobile, pendant que la scène est déplacée au moyen d'un projecteur. Les points irradiés avec le laser à balayage linéaire sont dans un plan, c'est-à-dire un plan de laser. Une formule de restriction obtenue à partir des informations peut être transformée en une équation linéaire. La résolution des équations linéaires simultanées obtenues à partir de nombreux points permet d'obtenir des espaces de solution dotés de quatre dimensions ou plus. Ledit procédé permet de projeter et de restaurer une forme tridimensionnelle des locus reflétés par le laser. Si la condition de restriction métrique, telle que l'orthogonalité ou le parallélisme d'un plan, est utilisée dans cet espace de solution, la restauration euclidienne de la scène peut être effectuée. La façon de constituer un projecteur de laser à balayage linéaire permet de satisfaire facilement ladite condition, telle que l'orthogonalité ou le parallélisme.
PCT/JP2006/322687 2006-11-08 2006-11-08 Procédé de mesure de forme tridimensionnelle et dispositif correspondant WO2008056427A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2008542981A JP5307549B2 (ja) 2006-11-08 2006-11-08 ３次元形状計測方法およびその装置
PCT/JP2006/322687 WO2008056427A1 (fr) 2006-11-08 2006-11-08 Procédé de mesure de forme tridimensionnelle et dispositif correspondant

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2006/322687 WO2008056427A1 (fr) 2006-11-08 2006-11-08 Procédé de mesure de forme tridimensionnelle et dispositif correspondant

Publications (1)

Publication Number Publication Date
WO2008056427A1 true WO2008056427A1 (fr) 2008-05-15

Family

ID=39364256

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2006/322687 WO2008056427A1 (fr) 2006-11-08 2006-11-08 Procédé de mesure de forme tridimensionnelle et dispositif correspondant

Country Status (2)

Country Link
JP (1) JP5307549B2 (fr)
WO (1) WO2008056427A1 (fr)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2014098602A (ja) * 2012-11-14 2014-05-29 Hitachi Power Solutions Co Ltd 水中観測装置及び水中観測方法とそれを用いた燃料集合体の放射能測定方法
JP2017023562A (ja) * 2015-07-24 2017-02-02 公立大学法人広島市立大学 ３次元形状計測装置、診断システム及び３次元形状計測方法
CN110793458A (zh) * 2019-10-30 2020-02-14 成都安科泰丰科技有限公司 一种二维激光位移传感器共面调整方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0694428A (ja) * 1992-07-29 1994-04-05 Matsushita Electric Ind Co Ltd ３次元形状入力装置
JP2003130621A (ja) * 2002-11-14 2003-05-08 Techno Dream 21:Kk ３次元形状計測方法およびその装置
JP2005098978A (ja) * 2003-08-20 2005-04-14 Japan Science & Technology Agency 三次元計測装置、三次元計測方法、三次元計測プログラムおよび記録媒体
WO2005100910A1 (fr) * 2004-04-09 2005-10-27 Techno Dream 21 Co., Ltd. Méthode de mesure d'une forme en trois dimensions et son équipement
WO2006013635A1 (fr) * 2004-08-03 2006-02-09 Techno Dream 21 Co., Ltd. Procédé de mesure de forme en trois dimensions et appareil correspondant

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2581863B2 (ja) * 1991-12-26 1997-02-12 三菱電機株式会社 立体形状計測装置及び立体形状計測用センサ
JPH05248838A (ja) * 1992-03-06 1993-09-28 Mitsubishi Electric Corp イメージセンサ
JP3254475B2 (ja) * 1994-01-11 2002-02-04 日本電信電話株式会社 レンジ・センサにおけるキャリブレーション方法
CN1300551C (zh) * 2002-07-25 2007-02-14 索卢申力士公司 使用光学标记自动排列三维扫描数据的设备和方法
JP3963219B2 (ja) * 2002-08-27 2007-08-22 学校法人東京電機大学 非接触式形状計測装置

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH0694428A (ja) * 1992-07-29 1994-04-05 Matsushita Electric Ind Co Ltd ３次元形状入力装置
JP2003130621A (ja) * 2002-11-14 2003-05-08 Techno Dream 21:Kk ３次元形状計測方法およびその装置
JP2005098978A (ja) * 2003-08-20 2005-04-14 Japan Science & Technology Agency 三次元計測装置、三次元計測方法、三次元計測プログラムおよび記録媒体
WO2005100910A1 (fr) * 2004-04-09 2005-10-27 Techno Dream 21 Co., Ltd. Méthode de mesure d'une forme en trois dimensions et son équipement
WO2006013635A1 (fr) * 2004-08-03 2006-02-09 Techno Dream 21 Co., Ltd. Procédé de mesure de forme en trois dimensions et appareil correspondant

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2014098602A (ja) * 2012-11-14 2014-05-29 Hitachi Power Solutions Co Ltd 水中観測装置及び水中観測方法とそれを用いた燃料集合体の放射能測定方法
JP2017023562A (ja) * 2015-07-24 2017-02-02 公立大学法人広島市立大学 ３次元形状計測装置、診断システム及び３次元形状計測方法
CN110793458A (zh) * 2019-10-30 2020-02-14 成都安科泰丰科技有限公司 一种二维激光位移传感器共面调整方法

Also Published As

Publication number Publication date
JP5307549B2 (ja) 2013-10-02
JPWO2008056427A1 (ja) 2010-02-25

Similar Documents

Publication Publication Date Title
JP6007178B2 (ja) ３次元撮像システム
JP6426968B2 (ja) 情報処理装置およびその方法
US8218858B2 (en) Enhanced object reconstruction
JP4290733B2 (ja) ３次元形状計測方法及びその装置
US8208029B2 (en) Method and system for calibrating camera with rectification homography of imaged parallelogram
US20120176478A1 (en) Forming range maps using periodic illumination patterns
US20120177284A1 (en) Forming 3d models using multiple images
US20120176380A1 (en) Forming 3d models using periodic illumination patterns
JP2012504771A (ja) 三次元及び距離の面間推定を与えるための方法及びシステム
Reichinger et al. Evaluation of methods for optical 3-D scanning of human pinnas
Ahmadabadian et al. An automatic 3D reconstruction system for texture-less objects
JP2016537612A (ja) モデリング構成及び三次元表面のトポグラフィーをモデリングする方法及びシステム
US20220180560A1 (en) Camera calibration apparatus, camera calibration method, and nontransitory computer readable medium storing program
US20190220952A1 (en) Method of acquiring optimized spherical image using multiple cameras
WO2020075252A1 (fr) Dispositif de traitement d'informations, programme, et procédé de traitement d'informations
Hafeez et al. Image based 3D reconstruction of texture-less objects for VR contents
JP2019511851A (ja) ピクセルビームを表すデータを生成する方法及び装置
WO2008056427A1 (fr) Procédé de mesure de forme tridimensionnelle et dispositif correspondant
CN108428244A (zh) 图像匹配方法及深度数据测量方法和***
JP7414090B2 (ja) 撮像装置および撮像装置の制御方法
Mengoni et al. An exploratory study on the application of reverse engineering in the field of small archaeological artefacts
Ettl Introductory review on ‘Flying Triangulation’: a motion-robust optical 3D measurement principle
Skabek et al. Comparison of photgrammetric techniques for surface reconstruction from images to reconstruction from laser scanning
Chikurtev et al. 3D modelling for object recognition with depth sensors
WO2020047692A1 (fr) Scanner intrabuccal 3d utilisant l'imagerie par champ lumineux

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 06823389

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2008542981

Country of ref document: JP

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 06823389

Country of ref document: EP

Kind code of ref document: A1