CN101825445A - Three-dimension measuring system for dynamic object - Google Patents
Three-dimension measuring system for dynamic object Download PDFInfo
- Publication number
- CN101825445A CN101825445A CN 201010166876 CN201010166876A CN101825445A CN 101825445 A CN101825445 A CN 101825445A CN 201010166876 CN201010166876 CN 201010166876 CN 201010166876 A CN201010166876 A CN 201010166876A CN 101825445 A CN101825445 A CN 101825445A
- Authority
- CN
- China
- Prior art keywords
- image
- dlp projector
- ccd camera
- computing machine
- dimensional
- 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
Links
Images
Abstract
The invention discloses a three-dimension measuring system for a dynamic object, which comprises a clock synchronous controller, a digital light processing (DLP) projector, two charge coupled device (CCD) cameras, an image acquisition card and a computer, wherein a color wheel for generating colorful images is removed from the DLP projector, included angles between optical mandrels of the CCD cameras and the optical mandrel of the DLP projector are both between 20 and 60 degrees, and the relative positions of the DLP projector and the CCD cameras are unchanged during measurement; the computer is a graphic card with a calculation-based uniform equipment framework; and the clock synchronous controller is connected with the DLP projector and the CCD cameras respectively, the DLP projector is connected with the computer, and the CCD cameras are connected with the computer through the graphic card respectively. The system has a most obvious characteristic of high measuring speed and can calculate and display the three-dimensional information of the dynamic object in real time.
Description
Technical field
The invention belongs to the precision measurement field, be specifically related to a kind of three-dimension measuring system of dynamic object, can carry out real-time three-dimensional measurement the object in the motion.
Background technology
Along with the high speed development of Computer Science and Technology, dynamic object three-dimensional measurement technology has application prospect more and more widely in many industries.In the industrial automation field, can utilize this technology that production line is monitored in real time, the quality problems in time finding to produce; In fields such as sports analysis, medical treatment, 3d gaming and film tricks, can utilize this commercial measurement human body or animal form, the expressive features in motion process, and realize the three-dimensional digital processing; In the vehicle safety testing experiment, also can adopt this technology that vehicle crash test is carried out complete three-dimensional measurement, by the stress-strain data in the computer-aided analysis collision process, help the designer to improve relevant design, improve the security of automobile.This development of technology will be brought help to the development of many industries.
In dynamic object three-dimensional measurement technology, realize measuring in real time meaning with particular importance.As in production line on-line monitoring process, if three-dimensional measurement speed can not be caught up with the speed of production line operation, measurement will lose meaning.Though do not need so strong real-time requirement in fields such as sports analysis, medical treatment, but because measurement data quantity is huge, if can not be being controlled at computing time in the suitable scope, Computerized three-dimensional is measured just can not in time feed back to user of service's metrical information, follow-up work can't be carried out, so these fields also need three-dimensional measurement to possess weak real-time.
The structural light measurement technology is because some advantages itself that have, is adapted at fully being developed in the research that dynamic process measures.Last century the eighties, when America and Asian countries were devoted to study the 3-d laser measurement technology, Germany had begun opposite structural light three-dimensional measuring technique and has studied.1985, the M.A.N. optical measuring technique center that is positioned at German Munich-Karlsfeld takes the lead in utilizing the phase shift interference method, and (Phase Shift Interferometry PSI) realized deformation measurement and vibration analysis.1986, the researchist Dr.Breuckmann at this center introduces measuring three-dimensional morphology with the PSI technology, formed a kind of new measuring three-dimensional morphology technology: phase measuring profilometer (PMP), and set up oneself laboratory, specialize in the research of this technical elements.The measuring system of the different model that released one after another in nearly 20 years, and obtained in a plurality of fields such as industrial detection, cultural relic digitalization, anthropological measurings using widely.Except that Dr.Breuckmann, the Reinhold Ritter professor of Dr.Steinbichler, Dr.Wolf and German TechnicalUniversity of Braunschweig, also be the pioneer of structural light measurement technical field, they last century the nineties set up SteinbichlerGmbH, Dr.WolfGmbH and GOM GmbH respectively, and many moneys structured light measurement system that released one after another, as: the COMET5 type structural light three-dimensional measuring system of Steinbichler GmbH, the Atos-II type structural light three-dimensional measuring system of GOMGmbH etc.But above-mentioned commercial structured light measurement system all needs to take many images in measuring process, and the processing time needs the several seconds, can not satisfy the demand that dynamic object is measured in real time.
The real-time three-dimensional measurement technology of dynamic object worldwide also is not very ripe at present, but development in recent years is rapid.From present development trend, mainly can be divided into two big classes based on the dynamic object three-dimensional measurement technology of area-structure light: (1) uses single image to carry out three-dimensional reconstruction; (2) use multiple image to carry out three-dimensional reconstruction.
At present existing a lot of research units are studied the method for using single image to carry out the dynamic object three-dimensional measurement.2006, research and development centre of Siemens developed the real-time three-dimensional structured light measurement system of a cover based on the coloud coding principle.This system is made up of one camera-single projector, throw a coloud coding image to testee during measurement, then by the coloured image of taking being decoded and the three-dimensional three-dimensional appearance that reconstructs testee that mates, the resolution of this system is 640 * 400, measuring speed is a per second 17-25 frame, but computation process is processed offline, need expend the long time.2008, the Hiroshi Kawasaki of the Ryusuke Sagawa of Osaka, Japan university, the beautiful university of Saitama and the people such as Ryo Furukawa of Hiroshima Institute of Technology have developed a kind of structured light system that uses the color screen table images to realize slow changing matter bulk measurement jointly, this system can successfully record the human face expression of slow variation, and measuring error is 0.52mm.The professor of the He An of Institutes Of Technology Of Nanjing has proposed a cover and has moved the dynamic 3 D topography measurement system of principle based on colour pase in 2007, and successfully realizes the three-dimensionalreconstruction of dynamic object.The Chen Shengyong of Zhejiang Polytechnical University professor has also developed a kind of area-structure light measuring system based on the color fringe coding principle in 2008, this system has realized under common hardware condition the topography measurement of object at a slow speed.Above-mentioned several single width measuring technique is all used the coloud coding pattern, and measuring accuracy generally can be subjected to the influence of testee surface color, and the Processing Algorithm of above-mentioned several method is all comparatively complicated, also can't realize real-time processing at present.
Another method of using single image to carry out three-dimensional measurement be based on the Fourier transform technology of profiling that Takeda and Moutoh propose in nineteen eighty-three (Fourier Transform Profilometry, FTP).The Su Xianyu of photoelectric technology research institute of domestic Sichuan University professor and seminar thereof apply to the FTP method during the dynamic object 3 d shape measures, and successful measurement be in the respiratory, the situation of torso model fluctuations that pace of change is slower; In the fluid mechanics fields of measurement, perfect measurement and reproduced the generation of liquid whirlpool and the process of intensification.And, for the three-dimensional measurement and the research in fields such as material deformation, blasting process, collision deformation provides a kind of new method.This seminar level that is in a leading position in high speed kinetic measurement field, but because the inherent characteristic of FTP, this method are having certain limitation aspect the complex object measuring, and because the algorithm complexity, the real-time of measurement is still waiting further raising.
Above-mentioned two class single image measuring methods all improve processing speed by sacrifice in measurement accuracy.In order to improve the measuring accuracy of dynamic object, the method that also has a lot of R﹠D institutions to propose to take fast at short notice several coded images realizes the three-dimensional measurement of dynamic object.The Szymon Rusinkiewicz professor and the team thereof of U.S. Princeton university, a kind of structured light face scanning system based on 4 frame optical strip images was proposed in 2002, allow measuring system or testee to do motion slowly in the measuring process, this system's per second can throw 60 width of cloth stripe patterns, and promptly per second can be gathered 15 frame three-dimensional datas.The Li Zhang professor of University of Washington has proposed a kind of dynamic human face measuring method based on Time-Space in 2004, this method calculates the dynamic object three-dimensional data by one group of black and white encoded light of projection according to three-dimensional matching method of time.These two kinds of methods use binary-coded measuring method the same with other, and its Measurement Resolution is subjected to the influence of encoding precision, can't realize high-resolution three-dimensional measurement, and because its decoding algorithm is comparatively complicated, are difficult to realize real-time three-dimensional reconstruction.
From the development of the real-time measuring technique of dynamic three-dimensional body, GPU calculates and more and more shows its importance, and many calculating can be finished on GPU in the dynamic 3 D measuring technique, therefore can improve computing velocity greatly.Initial GPU plays up custom-designedly for three-dimensional picture, and the three-dimensional picture that is used for computer display card shows to be quickened.Because in the design of GPU, multiple transistor more is used for the parallel processing of data, but not metadata cache (caching) that traditional C PU is emphasized and flow process control (flowcontrol), therefore GPU can handle the geometrical calculation work that reaches hundreds of three-dimensional vertices and pixel simultaneously, possesses and powerful parallel processing capability.Numerous scholars begin one's study by parallel numerical being calculated flexible mode for the three-dimensional picture demonstration, utilize the method for GPU high-performance calculation ability, GPU calculates gradually and has obtained Preliminary Applications in research fields such as physical simulation, calculating finance, calculation biologies.Simultaneously, the production firm of GPU makes it no longer only be confined to handle at three-dimensional picture also at the hardware structure of constantly improveing GPU, but towards general-purpose computations.In the GPU of a new generation framework, GPU is made up of the computing unit that is called as stream handle (Streaming Processors) in a large number, and they all follow the architecture mode of SIMT (single-instruction, multiple-thread, single instrction, multithreading).Multiprocessor can be mapped to each thread a scalar processor core, and each scalar thread uses oneself instruction address and buffer status independently to carry out.At this moment GPU just begins really possess the ability of handling complicated algorithm.
2007, the Thibaut Weise professor and the seminar thereof of Swiss Federal Institute of Technology were used for real time dynamic measurement with GPU, successfully develop a cover Real-time and Dynamic three-dimension measuring system, and speed reached per second 15 frames.
U.S. Song Zhang also with GPU since Real-time and Dynamic three-dimensional measurement technology in, and realized the measurement that human face expression changes, but its algorithm also can't well utilize the function of GPU parallel computation, therefore can only be applied to measure the slower object of movement velocity at present.
In sum: at present, though the structural light measurement technology can be carried out high-acruracy survey to the object dimensional surface configuration, but all need to take many images in the measuring process, processing time is long, can't satisfy the demand that the Real-time and Dynamic object is measured, in most cases can only carry out the data processing of off-line.Therefore existing structural light measurement technology also can't realize high-resolution real time dynamic measurement.Simultaneously, along with the development of GPU, the computation capability that GPU is powerful also is applied to a plurality of fields gradually, and can well be applied to real-time, high-resolution dynamic object three-dimensional measurement.
Summary of the invention
The object of the present invention is to provide a kind of dynamic object three-dimension measuring system, this system can carry out three-dimensional real time to dynamic object.
The three-dimension measuring system of dynamic object provided by the invention is characterized in that: this system comprises clock synchronizer controller, DLP projector, first, second CCD camera, image pick-up card and computing machine; Wherein, the DLP projector has removed the colour wheel that is used to produce coloured image, the photocentre axle clamp angle of the photocentre axle of first, second CCD camera and DLP projector and keeps the relative position of DLP projector and first, second CCD camera constant when measuring all between 20 to 60 degree; Computing machine is the figure video card that has based on calculating unified equipment frame;
Clock synchronizer controller links to each other with first, second CCD camera with the DLP projector respectively, and the DLP projector links to each other with computing machine, and first, second CCD camera all links to each other with computing machine by image pick-up card;
In the process of measuring, at first by clock synchronizer controller trigger the DLP projector to testee with one group of black and white sinusoidal grating of the frame per second projection image more than per second 90 frames, the sinusoidal grating image is all on the vertical direction of grid line 1/3 of translation pitch during projection, equidistant translation secondary, the projection of every frame three-dimensional data goes out three width of cloth sinusoidal grating images; Simultaneously, clock synchronizer controller sends signal for first, second CCD camera, and the projection frame per second of two image of camera acquisition frame rate and DLP projector is consistent; The image of first, second CCD camera collection sends computing machine to through image pick-up card, and the figure video card GPU in the computing machine handles the data of gathering, and calculates the three-dimensional data of dynamic object in real time.
Trigger DLP (Digital Light processing by clock controller in the native system, digital optical processor) projector goes out one group of black and white sinusoidal grating image to testee with the frame per second projection more than per second 90 frames, simultaneously, two CCD cameras of clock controller control are taken synchronously with the frame per second identical with projector, send shot image data to computing machine by image pick-up card then, computing machine is handled image and real-time calculating three-dimensional data fast, because each frame three-dimensional data needs 3 width of cloth images to calculate, so the frame per second of dynamic 3 D data can reach more than per second 30 frames.The data processing of part is finished by the GPU of computer display card is auxiliary in this process.The feature the most significantly of this system: measuring speed is fast, the three-dimensional information of calculating that can be real-time and demonstration dynamic object.
Description of drawings
Fig. 1 is the structural drawing of dynamic measurement system;
Fig. 2 is the structural drawing of DLP projector
Fig. 3 is the synchronizing signal synoptic diagram;
Fig. 4 is flow chart of data processing figure;
Fig. 5 is the schematic diagram of polar curve constraint;
Fig. 6 is the matching algorithm schematic diagram of polar curve constraint;
Fig. 7 is binocular stereo vision figure;
Fig. 8 is the system-computed framework map.
Embodiment
The present invention will be further described in detail below in conjunction with accompanying drawing and example.
As Fig. 1, system of the present invention comprises clock synchronizer controller 101, DLP projector 102, first, second CCD camera 103,104, image pick-up card 105 and computing machine 106.Wherein the photocentre axle clamp angle of the photocentre axle of first, second CCD camera and DLP projector is all between 20 to 60 degree, and the strict relative position of DLP projector and first, second CCD camera that keeps is constant when measuring.
In the process of measuring, at first by the clock synchronizer controller in the system 101 trigger DLP projector 102 to testee with one group of black and white sinusoidal grating of the frame per second projection image more than per second 90 frames, the sinusoidal grating image is all on the vertical direction of grid line 1/3 of translation pitch during projection, equidistant translation 2 times is so every frame three-dimensional data needs projection to go out 3 width of cloth sinusoidal grating images.Simultaneously, clock synchronizer controller 101 sends signal for first, second CCD camera 103,104, and the projection frame per second of two image of camera acquisition frame rate and DLP projector 102 is consistent.The image that first, second CCD camera 103,104 is gathered sends computing machine 106 to through image pick-up card 105.Computing machine 106 machines are handled the data of gathering, and calculate the three-dimensional data of dynamic object in real time.
The projection speed that requires DLP projector 102 in the native system is at least more than per second 90 frames, could be with the speed smoothness of per second 24 frames obtain the dynamic 3 D data.Yet the speed of DLP projector projection on the market is generally lower at present.Therefore the DLP projector 102 in the native system is reequiped on the basis of commercial projector.As shown in Figure 2, common commercial DLP projector mainly comprises 201, two lens 202 of light source and 203, DMD (Digital Micromirror Device) chip 205, colour wheel 204 and camera lens 206.For obtaining higher projection speed, DLP projector 102 has removed the colour wheel 204 that is used to produce coloured image in the native system.Because therefore the sinusoidal grating image that is black and white launched of DLP projector 102 removes the not influence of 204 pairs of native systems of colour wheel, can also utilize color channel in the DLP projector 102 to generate black and white sinusoidal grating image on the contrary and improve projection speed.Yet after DLP projector 102 removed colour wheel, the light emitting diode behind the colour wheel 204 can't produce trigger pip and trigger dmd chip 205 generation pictures.Therefore, the clock synchronizer controller 101 in the native system sends trigger pip can for DLP projector 102, makes dmd chip 205 generate image.
Native system realize one of gordian technique that the dynamic object real-time three-dimensional is measured be the DLP projector 102 that makes the high speed projection with first, second CCD camera 103 and 104 between synchronous working.Clock synchronizer controller 101 is by sending the synchronous working that trigger pip keeps total system.DLP projector with red (R), green (G), blue (B) and white (W) four passages is an example, as shown in Figure 3, and l
pBe the timing curve of DLP projector projected channel, l
dFor synchronous clock controller 101 sends to the trigger pip of dmd chip 205, l
cFor synchronous clock controller 101 sends to the signal of CCD camera 103 and 104, be used for controlling two camera photographic images.Synchronous clock controller 101 sends trigger pip to dmd chip 205 among the figure, when dmd chip 205 is experienced the negative edge of trigger pip, just generates the sinusoidal grating image of black and white successively by red (R), green (G), blue (B) and white (W) four passages.Simultaneously the signal sent by synchronous clock controller 101 of CCD camera 103 and 104 is gathered four images that passage generated successively, to reach the purpose of synchronous working.After the removal of the colour wheel in the DLP projector, the projection speed of system is four times of former projection speed.
The colour wheel technology that present commercial projector adopted has nothing in common with each other, but all can reequip in a manner described, and its principle of work is identical, all can system increase substantially projection speed.
After finishing, data acquisition just begins to calculate three-dimensional data, computing machine that system of the present invention adopts 106 is for having based on universal computing device framework (Compute Unified Device Architecture, CUDA) graphic presentation chip (GPU), this chip has powerful parallel computation function, can add the computation process of three-dimensional data.As shown in Figure 8,801 is CCD camera 103 and 104 captured Any Digit images, and all pixels are divided into several pixel blocks in 802.803 is the central processor CPU (Central Processing Unit) of computing machine 106, and 804 is the figure video card GPU (Graphic Processing Unit) of computing machine 106, and 805 is the stream handle of GPU.At first all pixels that will launch according to the thread block number of GPU804 are divided into several pixel blocks 802, to guarantee to allow the maximized concurrent operation of GPU804, use CPU803 that each piece is identified simultaneously, after first block of pixels disposes, CPU803 can import new pixel information among the GPU804 at once, handle successively, finally obtain complete three-dimensional model information, and show.The piecemeal of other computation process such as image, data transmission etc. are finished on the CPU803 again.Like this, by the acceleration of GPU804 concurrent operation model, the system that makes can finish the processing of mass data at the utmost point in the short time, high-resolution real-time measurement can be realized.
The computation process of pixel in 805 pairs of block of pixels of each stream handle below is described, as shown in Figure 4, its detailed process is:
(1) computing machine at first carried out for three steps to distortion black and white sinusoidal grating and transfers from one place to another under escort phase mutually.
The black and white sinusoidal grating that DLP projector 102 projects continuously fast, when the black and white sinusoidal grating projects to the testee surface, form deformed grating on body surface, suppose that the projection light intensity is that standard sine distributes, then the intensity distribution function of deformed grating image is:
Wherein be I (x, y) be pixel on the camera photographic images (x, light intensity y), a (x, y) and b (x is the light wave amplitude relevant with background y), and k is called the wave system number,
Be pixel (x, phase value y).Three the step transfer from one place to another under escort mutually the phase method with grating on the vertical direction of grid line translation pitch 1/3, then phase place will move 2 π/3 in the light intensity expression of formula (1), for the equidistant translation of the sinusoidal grating of same frequency 2 times, obtain 3 width of cloth images.After carrying out three step phase shifts, (x is y) at the light intensity function I in each step for pixel
1, I
2, I
3Be respectively:
Can get by formula (3) to (5):
I
1, I
2, I
3(x, grating y) be light intensity, 1/3 o'clock light intensity of grating translation pitch and 2/3 o'clock the light intensity of grating translation pitch during translation not to be respectively pixel on the camera photographic images.
So just can calculate the relative phase value of each pixel in the image
Therefore (claiming the phase place main value again), it is a monotonically increasing in a phase cycling, but should value in whole measurement space not unique, can't be directly seeks match point on the camera image by the phase place main value, so just can not carry out stereo reconstruction.
(2) utilize the match point that obtains the captured image of two CCD cameras based on the Stereo Matching Algorithm of limit constraint;
Adopt a kind of Stereo Matching Algorithm herein based on the polar curve constraint.The polar curve constraint principles as shown in Figure 5, establishing the picture point of measured point P on the captured image 301 of CCD camera 103 is P
1, the picture point on the captured image 302 of CCD camera 104 is P
2, and O
1And O
2Respectively with the photocentre point of CCD camera 103 and 104, straight line O
1O
2Intersection point e with the plane at the plane at image 301 places and image 302 places
1And e
2Be limit, plane P O
1O
2With the intersection of image 301 and image 302 be polar curve l
1And l
2, then easily know polar curve l
1Certain for limit e
1, polar curve l
2Certain for limit e
2, picture point P
1One fixes on polar curve l
1On, picture point P2 is at polar curve l
2On, that is to say can be by the some P on the image 301
1Coordinate is obtained and P on image 302
1Corresponding polar curve equation.This is the polar curve constraint principles.
This system's neutral body matching algorithm the steps include: as shown in Figure 6
Get certain 1 P in the image 301 when (1) calculating
1
(2) according to the polar curve constraint principles, according to P
1Coordinate figure just can on the projected image 303 of DLP projector 102, obtain a polar curve l
3, here projector 303 to be treated as camera, the image that projector 303 projections go out also can be regarded the image that projector 303 is captured as;
(3) in the image 303 of DLP projection, find at least two and P
1Point has the straight line l of same phase main value
31And l
32, establish polar curve l
3With straight line l
31And l
32Meet at a P
31And P
32
(4) on image 302, obtain and P according to limit constraint
1, P
31And P
32Corresponding polar curve l
2, l
21And l
22Equation;
(5) polar curve l
2, l
21And l
22In two lines P that intersects at a point on image 302 is arranged
2, P then
2Be P
1Match point, P simultaneously
2Also be the picture point of P point on image 302;
(6) by match point P
1, P
2Utilize the binocular stereo vision principle to carry out a cloud reconstruct, obtain the three-dimensional data of object.
Just can use the binocular stereo vision principle to carry out a cloud reconstruct after coupling is finished, calculate the three-dimensional point coordinate on testee surface.
As shown in Figure 7, spatial point P is at world coordinate system O
wX
wY
wZ
wUnder coordinate figure be (X
w, Y
w, Z
w), the picture point P that P is ordered on the image 301
1Image coordinate be (u
1, v
1), by the pinhole imaging system model, can list the transformational relation equation (8) of P point from the image coordinate to the world coordinates, wherein k
1Be the Z axial coordinate of P point under camera 103 coordinate systems, a
1x=f
1S
X1And a
1y=f
1S
Y1, f
1Be the focal length of camera 103, S
X1And S
Y1Be the pixel count (pixels/mm) on the camera 103 captured plane of delineation unit distances, (u
10, v
10) be the coordinate of image center, R
1And T
1For be tied to rotation matrix and the translation matrix between camera 103 coordinate systems from world coordinates.Picture point P on the image 302
2Image coordinate be (u
2, v
2).Equally also can list equation (9), wherein k
2Be the Z axial coordinate of P point under camera 104 coordinate systems, a
2x=f
2S
X2And a
2y=f
2S
Y2, f
2Be the focal length of camera 104, S
2xAnd S
2yBe the pixel count (pixels/mm) on the camera 104 captured plane of delineation unit distances, (u
20, v
20) be the coordinate of image center, R
2And T
2For be tied to rotation matrix and the translation matrix between camera 104 coordinate systems from world coordinates.Calculate the three-dimensional coordinate (X of measured object according to equation (8) and (9)
w, Y
w, Z
w), k wherein
1And k
2For scale factor is obtained in the solution procedure of equation.
Native system utilizes the GPU+CPU mixed architecture to quicken the computation process of Real-time and Dynamic three-dimensional measurement.It all is to be that unit calculates with the pixel that in the above computation process three goes on foot coupling, stereo reconstruction and the data presentation transfering from one place to another under escort phase mutually, retrain based on polar curve, and independent uncorrelated between the pixel on the algorithm.Therefore, this four step calculates and all can utilize the parallel computational model extensive, graininess of GPU to be optimized acceleration, makes the calculating of a plurality of pixels carry out simultaneously.
Above DLP projector 102 projection speed and CCD camera 103 and 104 picking rates all are per second 90 frames, calculate each frame three-dimensional appearance data and need three secondary pictures, calculating and demonstration in every frame three-dimensional data on the basis based on the GPU+CPU hybrid structure can be finished in the time at 30ms, therefore the present invention can realize the real-time three-dimensional measurement of dynamic object, and the measurement frame per second can reach for 30 frame/seconds.
The above is preferred embodiment of the present invention, but the present invention should not be confined to the disclosed content of this embodiment and accompanying drawing.So everyly do not break away from the equivalence of finishing under the spirit disclosed in this invention or revise, all fall into the scope of protection of the invention.
Claims (2)
1. the three-dimension measuring system of a dynamic object, it is characterized in that: this system comprises clock synchronizer controller (101), DLP projector (102), first, second CCD camera (103,104), image pick-up card (105) and computing machine (106); Wherein, DLP projector (102) has removed the colour wheel (204) that is used to produce coloured image, the photocentre axle clamp angle of the photocentre axle of first, second CCD camera and DLP projector and keeps the relative position of DLP projector and first, second CCD camera constant when measuring all between 20 to 60 degree; Computing machine (106) is for having based on the figure video card that calculates unified equipment frame;
Clock synchronizer controller (101) links to each other with first, second CCD camera (103,104) with DLP projector (102) respectively, DLP projector (102) links to each other with computing machine (106), and first, second CCD camera (103,104) all links to each other with computing machine (106) by image pick-up card (105);
In the process of measuring, at first by clock synchronizer controller (101) trigger DLP projector (102) to testee with one group of black and white sinusoidal grating of the frame per second projection image more than per second 90 frames, the sinusoidal grating image is all on the vertical direction of grid line 1/3 of translation pitch during projection, equidistant translation secondary, the projection of every frame three-dimensional data goes out three width of cloth sinusoidal grating images; Simultaneously, clock synchronizer controller (101) sends signal for first, second CCD camera (103,104), and the projection frame per second of two image of camera acquisition frame rate and DLP projector (102) is consistent; The image that first, second CCD camera (103,104) is gathered sends computing machine (106) to through image pick-up card (105), and the figure video card GPU in the computing machine (106) handles the data of gathering, and calculates the three-dimensional data of dynamic object in real time.
2. the three-dimension measuring system of dynamic object according to claim 1, it is characterized in that: computing machine 106 carries out data processing according to following process:
The 1st step computing machine at first carried out for three steps to distortion black and white sinusoidal grating and transfers from one place to another under escort phase mutually;
The 2nd step was obtained the match point of the captured image of first, second CCD camera according to following process:
Get certain 1 P in the image 301 of institute's the one CCD camera shooting when (2.1) calculating
1
(2.2) according to the polar curve constraint principles, according to P
1Coordinate figure on the projected image 303 of DLP projector 102, obtain a polar curve l
3
(2.3) in the projected image 303 of DLP projection, find at least two and P
1Point has the straight line l of same phase main value
31And l
32, establish polar curve l
3With straight line l
31And l
32Meet at a P
31And P
32
(2.4) on the image 302 that the 2nd CCD camera is taken according to the limit constraint obtain and P
1, P
31And P
32Corresponding polar curve l
2, l
21And l
22Equation;
(2.5) polar curve l
2, l
21And l
22In two lines P that intersects at a point on image 302 is arranged
2, P then
2Be P
1Match point, P simultaneously
2Also be the picture point of P point on image 302;
(2.6) by match point P
1, P
2Carry out a cloud reconstruct, obtain the three-dimensional data of object;
Use the binocular stereo vision principle to carry out a cloud reconstruct after the 3rd step, coupling was finished, calculate the three-dimensional point coordinate on testee surface.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101668765A CN101825445B (en) | 2010-05-10 | 2010-05-10 | Three-dimension measuring system for dynamic object |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101668765A CN101825445B (en) | 2010-05-10 | 2010-05-10 | Three-dimension measuring system for dynamic object |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101825445A true CN101825445A (en) | 2010-09-08 |
| CN101825445B CN101825445B (en) | 2011-11-30 |
Family
ID=42689500
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010101668765A Active CN101825445B (en) | 2010-05-10 | 2010-05-10 | Three-dimension measuring system for dynamic object |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101825445B (en) |
Cited By (35)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102175182A (en) * | 2011-01-27 | 2011-09-07 | 浙江大学宁波理工学院 | Structured light three-dimensional measurement device and complete point cloud data acquisition method thereof |
| CN102207371A (en) * | 2011-03-16 | 2011-10-05 | 西南石油大学 | Three-dimensional point coordinate measuring method and measuring apparatus thereof |
| CN102261896A (en) * | 2011-04-19 | 2011-11-30 | 长春东瑞科技发展有限公司 | Method and system for measuring three-dimensional shape of object based on phase measurement |
| CN102316355A (en) * | 2011-09-15 | 2012-01-11 | 丁少华 | Generation method of 3D machine vision signal and 3D machine vision sensor |
| CN102506758A (en) * | 2011-10-12 | 2012-06-20 | 北京航空航天大学 | Object surface three-dimensional morphology multi-sensor flexible dynamic vision measurement system and method |
| CN102679937A (en) * | 2011-03-17 | 2012-09-19 | 镇江亿海软件有限公司 | Ship steel plate dynamic three-dimension measurement method based on multi-camera vision |
| CN103020080A (en) * | 2011-09-23 | 2013-04-03 | 鸿富锦精密工业（深圳）有限公司 | Method and system for rapidly reading point cloud document |
| CN103148806A (en) * | 2011-12-07 | 2013-06-12 | 史金龙 | Ship steel plate dynamic three-dimensional measuring system based on projection and multi-view vision |
| CN103148792A (en) * | 2011-12-07 | 2013-06-12 | 史金龙 | Ship steel plate dynamic three-dimensional measuring method |
| CN103292733A (en) * | 2013-05-27 | 2013-09-11 | 华中科技大学 | Corresponding point searching method based on phase shift and trifocal tensor |
| CN103335611A (en) * | 2013-06-13 | 2013-10-02 | 华中科技大学 | Method for GPU-based object three-dimensional shape measurement |
| CN103411564A (en) * | 2013-08-29 | 2013-11-27 | 衢州亿龙信息技术有限公司 | Optical scanning device of integrated structure |
| CN103900504A (en) * | 2014-04-10 | 2014-07-02 | 华中科技大学 | Nano-scale real-time three-dimensional visual information feedback method |
| CN103968782A (en) * | 2014-05-23 | 2014-08-06 | 四川大学 | Real-time three-dimensional measurement method based on color sine structured light coding |
| CN104457562A (en) * | 2013-09-13 | 2015-03-25 | 武汉惟景三维科技有限公司 | Adapter on-line detection system based on surface structured light |
| CN104964656A (en) * | 2015-06-26 | 2015-10-07 | 天津大学 | Self-positioning flowing-type rapid scanning measuring device and method based on inertial navigation |
| CN105091748A (en) * | 2015-05-29 | 2015-11-25 | 南京南车浦镇城轨车辆有限责任公司 | Measuring system for tolerance dimensions of railway vehicle |
| CN105136030A (en) * | 2015-08-24 | 2015-12-09 | 苏州市博得立电源科技有限公司 | Automatic detection system |
| CN105313126A (en) * | 2014-07-29 | 2016-02-10 | 精工爱普生株式会社 | Control system, robot system, and control method |
| CN105403143A (en) * | 2015-10-28 | 2016-03-16 | 苏州西博三维科技有限公司 | Measurement method and system thereof for acquiring displacement changes of earthquake simulator |
| CN105627942A (en) * | 2015-12-25 | 2016-06-01 | 华南理工大学 | Imaging apparatus and method for detecting object surface micro deformation through machine vision |
| CN105953746A (en) * | 2016-05-16 | 2016-09-21 | 西安工业大学 | Three-dimensional contour real-time measurement method of dynamic objects |
| CN106210702A (en) * | 2016-08-30 | 2016-12-07 | 四川川大智胜软件股份有限公司 | A kind of synchronization three dimensional data collection instrument, acquisition system and acquisition method |
| CN106840251A (en) * | 2015-12-07 | 2017-06-13 | 中国电力科学研究院 | A kind of 3 D scanning system for the detection of low-voltage current mutual inductor outward appearance |
| CN106969723A (en) * | 2017-04-21 | 2017-07-21 | 华中科技大学 | High speed dynamic object key point method for three-dimensional measurement based on low speed camera array |
| CN109087350A (en) * | 2018-08-07 | 2018-12-25 | 西安电子科技大学 | Fluid light intensity three-dimensional rebuilding method based on projective geometry |
| CN109283538A (en) * | 2018-07-13 | 2019-01-29 | 上海大学 | A kind of naval target size detection method of view-based access control model and laser sensor data fusion |
| WO2019237782A1 (en) * | 2018-06-15 | 2019-12-19 | 杭州海康机器人技术有限公司 | 3d information detection device |
| CN110631507A (en) * | 2019-11-07 | 2019-12-31 | 成都铁安科技有限责任公司 | Three-dimensional measurement method and system based on structured light |
| CN110686599A (en) * | 2019-10-31 | 2020-01-14 | 中国科学院自动化研究所 | Three-dimensional measurement method, system and device based on colored Gray code structured light |
| CN110720897A (en) * | 2018-07-16 | 2020-01-24 | 大立光电股份有限公司 | Optical pulse wave image measuring instrument and pulse wave measuring method |
| CN110738206A (en) * | 2019-10-15 | 2020-01-31 | 四川川大智胜软件股份有限公司 | three-dimensional face polarized illumination and information acquisition system and method thereof |
| CN112362088A (en) * | 2020-12-11 | 2021-02-12 | 中国石油大学(华东) | Synchronous acquisition method and system for multi-grating data |
| CN112525109A (en) * | 2020-12-18 | 2021-03-19 | 成都立鑫新技术科技有限公司 | Method for measuring object motion attitude angle based on GPU |
| CN113251952A (en) * | 2021-05-17 | 2021-08-13 | 四川川大智胜软件股份有限公司 | Three-dimensional measurement system and three-dimensional measurement method for grating translation structured light |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103475887A (en) * | 2013-07-12 | 2013-12-25 | 黑龙江科技大学 | Image synchronization method and device in camera visual system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7146036B2 (en) * | 2003-02-03 | 2006-12-05 | Hewlett-Packard Development Company, L.P. | Multiframe correspondence estimation |
| CN101105393A (en) * | 2006-07-13 | 2008-01-16 | 周波 | Vision measuring method for projecting multiple frequency grating object surface tri-dimensional profile |
| CN101245998A (en) * | 2008-02-01 | 2008-08-20 | 黑龙江科技学院 | Imaging method of three-dimensional measuring system |
| US20090169095A1 (en) * | 2008-01-02 | 2009-07-02 | Spatial Integrated Systems, Inc. | System and method for generating structured light for 3-dimensional image rendering |
-
2010
- 2010-05-10 CN CN2010101668765A patent/CN101825445B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7146036B2 (en) * | 2003-02-03 | 2006-12-05 | Hewlett-Packard Development Company, L.P. | Multiframe correspondence estimation |
| CN101105393A (en) * | 2006-07-13 | 2008-01-16 | 周波 | Vision measuring method for projecting multiple frequency grating object surface tri-dimensional profile |
| US20090169095A1 (en) * | 2008-01-02 | 2009-07-02 | Spatial Integrated Systems, Inc. | System and method for generating structured light for 3-dimensional image rendering |
| CN101245998A (en) * | 2008-02-01 | 2008-08-20 | 黑龙江科技学院 | Imaging method of three-dimensional measuring system |
Non-Patent Citations (2)
| Title |
|---|
| 《中国优秀博硕士学位论文全文数据库（硕士）工程科技II辑》 20060515 秦大辉 结构光三维测量系统关键技术的研究 正文第3.1,3.2.1节 1,2 , 第05期 2 * |
| 《北京信息科技大学学报》 20100331 吕乃光 等 结构光三维视觉测量关键技术的研究 正文第3,4节、图6,7 1,2 第25卷, 第1期 2 * |
Cited By (50)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102175182A (en) * | 2011-01-27 | 2011-09-07 | 浙江大学宁波理工学院 | Structured light three-dimensional measurement device and complete point cloud data acquisition method thereof |
| CN102175182B (en) * | 2011-01-27 | 2012-10-10 | 浙江大学宁波理工学院 | Structured light three-dimensional measurement device and complete point cloud data acquisition method thereof |
| CN102207371A (en) * | 2011-03-16 | 2011-10-05 | 西南石油大学 | Three-dimensional point coordinate measuring method and measuring apparatus thereof |
| CN102679937A (en) * | 2011-03-17 | 2012-09-19 | 镇江亿海软件有限公司 | Ship steel plate dynamic three-dimension measurement method based on multi-camera vision |
| CN102261896A (en) * | 2011-04-19 | 2011-11-30 | 长春东瑞科技发展有限公司 | Method and system for measuring three-dimensional shape of object based on phase measurement |
| CN102316355A (en) * | 2011-09-15 | 2012-01-11 | 丁少华 | Generation method of 3D machine vision signal and 3D machine vision sensor |
| CN103020080A (en) * | 2011-09-23 | 2013-04-03 | 鸿富锦精密工业（深圳）有限公司 | Method and system for rapidly reading point cloud document |
| CN102506758A (en) * | 2011-10-12 | 2012-06-20 | 北京航空航天大学 | Object surface three-dimensional morphology multi-sensor flexible dynamic vision measurement system and method |
| CN103148806A (en) * | 2011-12-07 | 2013-06-12 | 史金龙 | Ship steel plate dynamic three-dimensional measuring system based on projection and multi-view vision |
| CN103148792A (en) * | 2011-12-07 | 2013-06-12 | 史金龙 | Ship steel plate dynamic three-dimensional measuring method |
| CN103292733B (en) * | 2013-05-27 | 2016-03-02 | 华中科技大学 | A kind of corresponding point lookup method based on phase shift and trifocal tensor |
| CN103292733A (en) * | 2013-05-27 | 2013-09-11 | 华中科技大学 | Corresponding point searching method based on phase shift and trifocal tensor |
| CN103335611A (en) * | 2013-06-13 | 2013-10-02 | 华中科技大学 | Method for GPU-based object three-dimensional shape measurement |
| CN103335611B (en) * | 2013-06-13 | 2016-02-24 | 华中科技大学 | Based on the object dimensional surface shape measurement method of GPU |
| CN103411564A (en) * | 2013-08-29 | 2013-11-27 | 衢州亿龙信息技术有限公司 | Optical scanning device of integrated structure |
| CN104457562A (en) * | 2013-09-13 | 2015-03-25 | 武汉惟景三维科技有限公司 | Adapter on-line detection system based on surface structured light |
| CN103900504A (en) * | 2014-04-10 | 2014-07-02 | 华中科技大学 | Nano-scale real-time three-dimensional visual information feedback method |
| CN103968782A (en) * | 2014-05-23 | 2014-08-06 | 四川大学 | Real-time three-dimensional measurement method based on color sine structured light coding |
| CN105313126A (en) * | 2014-07-29 | 2016-02-10 | 精工爱普生株式会社 | Control system, robot system, and control method |
| CN105313126B (en) * | 2014-07-29 | 2019-01-01 | 精工爱普生株式会社 | control system, robot system and control method |
| CN105091748A (en) * | 2015-05-29 | 2015-11-25 | 南京南车浦镇城轨车辆有限责任公司 | Measuring system for tolerance dimensions of railway vehicle |
| CN105091748B (en) * | 2015-05-29 | 2017-11-28 | 南京中车浦镇城轨车辆有限责任公司 | Rail vehicle tolerance dimension measuring system |
| CN104964656A (en) * | 2015-06-26 | 2015-10-07 | 天津大学 | Self-positioning flowing-type rapid scanning measuring device and method based on inertial navigation |
| CN105136030A (en) * | 2015-08-24 | 2015-12-09 | 苏州市博得立电源科技有限公司 | Automatic detection system |
| CN105403143A (en) * | 2015-10-28 | 2016-03-16 | 苏州西博三维科技有限公司 | Measurement method and system thereof for acquiring displacement changes of earthquake simulator |
| CN105403143B (en) * | 2015-10-28 | 2017-10-13 | 苏州西博三维科技有限公司 | A kind of measuring method and its system of the change in displacement for obtaining simulated earthquake vibration stand |
| CN106840251B (en) * | 2015-12-07 | 2020-04-14 | 中国电力科学研究院 | Three-dimensional scanning system for appearance detection of low-voltage current transformer |
| CN106840251A (en) * | 2015-12-07 | 2017-06-13 | 中国电力科学研究院 | A kind of 3 D scanning system for the detection of low-voltage current mutual inductor outward appearance |
| CN105627942A (en) * | 2015-12-25 | 2016-06-01 | 华南理工大学 | Imaging apparatus and method for detecting object surface micro deformation through machine vision |
| CN105953746A (en) * | 2016-05-16 | 2016-09-21 | 西安工业大学 | Three-dimensional contour real-time measurement method of dynamic objects |
| CN105953746B (en) * | 2016-05-16 | 2018-12-14 | 西安工业大学 | The 3-D profile real-time measuring method of dynamic object |
| CN106210702A (en) * | 2016-08-30 | 2016-12-07 | 四川川大智胜软件股份有限公司 | A kind of synchronization three dimensional data collection instrument, acquisition system and acquisition method |
| CN106969723B (en) * | 2017-04-21 | 2019-07-09 | 华中科技大学 | High speed dynamic object key point method for three-dimensional measurement based on low speed camera array |
| CN106969723A (en) * | 2017-04-21 | 2017-07-21 | 华中科技大学 | High speed dynamic object key point method for three-dimensional measurement based on low speed camera array |
| WO2019237782A1 (en) * | 2018-06-15 | 2019-12-19 | 杭州海康机器人技术有限公司 | 3d information detection device |
| EP3809095A4 (en) * | 2018-06-15 | 2021-06-23 | Hangzhou Hikrobot Technology Co., Ltd. | 3d information detection device |
| CN109283538A (en) * | 2018-07-13 | 2019-01-29 | 上海大学 | A kind of naval target size detection method of view-based access control model and laser sensor data fusion |
| CN109283538B (en) * | 2018-07-13 | 2023-06-13 | 上海大学 | Marine target size detection method based on vision and laser sensor data fusion |
| CN110720897A (en) * | 2018-07-16 | 2020-01-24 | 大立光电股份有限公司 | Optical pulse wave image measuring instrument and pulse wave measuring method |
| CN109087350B (en) * | 2018-08-07 | 2020-06-26 | 西安电子科技大学 | Fluid light intensity three-dimensional reconstruction method based on projective geometry |
| CN109087350A (en) * | 2018-08-07 | 2018-12-25 | 西安电子科技大学 | Fluid light intensity three-dimensional rebuilding method based on projective geometry |
| CN110738206A (en) * | 2019-10-15 | 2020-01-31 | 四川川大智胜软件股份有限公司 | three-dimensional face polarized illumination and information acquisition system and method thereof |
| CN110738206B (en) * | 2019-10-15 | 2023-04-18 | 四川川大智胜软件股份有限公司 | Three-dimensional face polarized illumination and information acquisition system and method thereof |
| CN110686599B (en) * | 2019-10-31 | 2020-07-03 | 中国科学院自动化研究所 | Three-dimensional measurement method, system and device based on colored Gray code structured light |
| CN110686599A (en) * | 2019-10-31 | 2020-01-14 | 中国科学院自动化研究所 | Three-dimensional measurement method, system and device based on colored Gray code structured light |
| CN110631507B (en) * | 2019-11-07 | 2021-04-20 | 成都盛锴科技有限公司 | Three-dimensional measurement method and system based on structured light |
| CN110631507A (en) * | 2019-11-07 | 2019-12-31 | 成都铁安科技有限责任公司 | Three-dimensional measurement method and system based on structured light |
| CN112362088A (en) * | 2020-12-11 | 2021-02-12 | 中国石油大学(华东) | Synchronous acquisition method and system for multi-grating data |
| CN112525109A (en) * | 2020-12-18 | 2021-03-19 | 成都立鑫新技术科技有限公司 | Method for measuring object motion attitude angle based on GPU |
| CN113251952A (en) * | 2021-05-17 | 2021-08-13 | 四川川大智胜软件股份有限公司 | Three-dimensional measurement system and three-dimensional measurement method for grating translation structured light |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101825445B (en) | 2011-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101825445B (en) | Three-dimension measuring system for dynamic object | |
| CN111344716A (en) | Full convolution point of interest detection and description via homography transformation adaptation | |
| JP2024028494A (en) | Method and apparatus for angle detection using neural networks and angle detectors | |
| CN106056620B (en) | Line laser camera measurement system calibrating method | |
| CN103697815B (en) | Mixing structural light three-dimensional information getting method based on phase code | |
| CN105353873A (en) | Gesture manipulation method and system based on three-dimensional display | |
| CN102945564A (en) | True 3D modeling system and method based on video perspective type augmented reality | |
| CN109556535B (en) | Three-dimensional surface type one-step reconstruction method based on color fringe projection | |
| CN108492017B (en) | Product quality information transmission method based on augmented reality | |
| CN208887565U (en) | A kind of three-dimension measuring system of non-contact type high-precision | |
| CN102261896A (en) | Method and system for measuring three-dimensional shape of object based on phase measurement | |
| CN103308028A (en) | Binocular stereovision measuring device and method for attitude angle of triaxial air floating platform | |
| CN102980513A (en) | Monocular panoramic three-dimensional vision sensor focusing on objects | |
| CN109934935A (en) | A kind of clothes are exposed the false with human body image, match the method and system of migration | |
| CN102914295A (en) | Computer vision cube calibration based three-dimensional measurement method | |
| CN103747196A (en) | Kinect sensor-based projection method | |
| CN110264527A (en) | Real-time binocular stereo vision output method based on ZYNQ | |
| CN104714646A (en) | 3D virtual touch control man-machine interaction method based on stereoscopic vision | |
| CN103196393A (en) | Geometrical shape and surface color real time imaging device | |
| CN103985121A (en) | Method for calibrating underwater projector | |
| CN110728745A (en) | Underwater binocular stereoscopic vision three-dimensional reconstruction method based on multilayer refraction image model | |
| CN109360267A (en) | A kind of thin objects quick three-dimensional reconstructing method | |
| Lin et al. | Using hybrid sensoring method for motion capture in volleyball techniques training | |
| CN105844623A (en) | Target object depth information obtaining method based on De sequence hybrid coding | |
| CN108492336A (en) | A kind of camera calibration method of structure Virtual Calibration object of reference |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |