CN106875468A - Three-dimensional reconstruction apparatus and method - Google Patents
Three-dimensional reconstruction apparatus and method Download PDFInfo
- Publication number
- CN106875468A CN106875468A CN201510927095.6A CN201510927095A CN106875468A CN 106875468 A CN106875468 A CN 106875468A CN 201510927095 A CN201510927095 A CN 201510927095A CN 106875468 A CN106875468 A CN 106875468A
- Authority
- CN
- China
- Prior art keywords
- luminescence unit
- reconstruction
- light source
- dimensional
- model
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
Abstract
The invention discloses a kind of three-dimensional reconstruction apparatus and method, the device includes：Image acquisition unit, the image information for obtaining target object；Multiple luminescence units, are arranged on around described image acquiring unit；Synchronous control unit, the light on and off for controlling the multiple luminescence unit, when each luminescence unit is lighted, Synchronization Control described image acquiring unit photographic subjects object obtains the image information of the lower target object of each luminescence unit irradiation；Reconstruction unit, for the luminous model according to described image information and luminescence unit, rebuilds to target object；The luminous model is output intensity, the relation of shooting angle on output intensity and each exit direction on each luminescence unit primary optical axis.The technical scheme that the present invention is provided, reduces the operation complexity of three-dimensional reconstruction, improves the precision and efficiency of three-dimensional reconstruction.
Description
Technical field
The present invention relates to three-dimensional reconstruction field, more particularly to a kind of three-dimensional reconstruction apparatus and method.
Background technology
At present, three-dimensional output and the appearance of the end product for showing are supported with various, for the demand urgency of three-dimensional content
Increase severely plus, and existing three dimensional data collection mode is main with 3-D scanning, such as：Laser scanning, based on structure light scan technology,
Related product structure is complex, although can obtain accurate three-dimensional information, but its hardware cost and operating method are common
User is difficult to bear and be difficult what is grasped, for example：Chinese patent application the CN101383046Ath discloses a kind of based on image
Three-dimensional rebuilding method, the method extracts the characteristic point of each image first with SIFT feature extraction algorithm, and special to each
Levying a little carries out being three-dimensional reconstruction.It is combined by by feature points constraint and colour consistency, using the characteristic point around picture point
As information distance search constraints, the hunting zone of space of points position to be reconstructed is determined, and further obtain the space of reconstruction point
Position；Chinese patent application CN104331924A discloses the three-dimensional rebuilding method based on single camera SFS algorithms, specifically presses
Following steps are implemented：Calibrating camera, then extracts objective contour first, finally carries out target based on voxel using SFS algorithms
Three-dimensional reconstruction.
There is a kind of three-dimensional rebuilding method based on light measurement in existing method, light measurement is a kind of skill in 3D vision
Art means, its general principle is irradiated by the light source of different directions, is shot by camera, analyzes object under different irradiating angles
Light and shade information change recover the three-dimensional normal direction information on surface, and then obtain surface three dimension shape.Existing light measurement is three-dimensional
It is that method for reconstructing employs spot light or source of parallel light it is assumed that do not consider the characteristics of luminescence of light source itself, when object away from
From light source farther out or when light source meets directional light or spot light characteristic in itself, existing method can obtain preferable three-dimensional reconstruction
Effect, but this is higher to light source oneself requirement, and to obtain directional light, light source needs larger volume in itself, such as uses LED gusts
Array apparatus.
From the foregoing, existing three-dimensional reconstruction is primarily present following shortcoming：
1), using laser or method of structured light, although precision is higher but expensive, complex operation, domestic consumer is difficult to
Bear and skillfully grasp；
2), the three-dimensional rebuilding method simple structure based on light measurement, hardware cost is low, but existing method is assumed to be used
Light source is directional light or ideal point light source, it is necessary to uniformity professional light source product higher could obtain preferable three-dimensional reconstruction
Effect, it is difficult to work on closely；
3), the existing closely light measurement three-dimensional reconstruction based on single source, simply in image processing stage pair
The illuminance uniformity of image is corrected, for improving three-dimensional reconstruction effect, not in view of light source itself the characteristics of luminescence with
And the factor such as incident angle, irradiation distance, therefore be also difficult to obtain preferable three-dimensional reconstruction effect.
In sum, existing three-dimensional reconstruction scheme, complex operation, rebuilds efficiency and precision is low.
The content of the invention
A kind of three-dimensional reconstruction apparatus are the embodiment of the invention provides, is used to reduce the operation complexity of three-dimensional reconstruction, improved
The precision and efficiency of three-dimensional reconstruction, the device include：
Image acquisition unit, the image information for obtaining target object；
Multiple luminescence units, are arranged on around described image acquiring unit；
Synchronous control unit, the light on and off for controlling the multiple luminescence unit are synchronous when each luminescence unit is lighted
Control described image acquiring unit photographic subjects object, obtains the image information of the lower target object of each luminescence unit irradiation；
Reconstruction unit, for the luminous model according to described image information and luminescence unit, rebuilds to target object；
The luminous model is output intensity, the angle of emergence on output intensity and each exit direction on each luminescence unit primary optical axis
The relation of degree.
Compared with laser scanning of the prior art and structure light scan reconstructing device, provided in an embodiment of the present invention three
Dimension reconstructing device, with following Advantageous Effects：
Firstly, since the three-dimensional reconstruction apparatus that invention is provided include an image acquisition unit and multiple luminescence unit structures
Into system architecture is simple, low cost, and by synchronous control unit, when controlling each luminescence unit to irradiate, control image is obtained
Unit sync pulse jamming target object is taken, using luminescence unit and the Strategy For Synchronization Control of image acquisition unit, it is possible to achieve quick
Scanning process, accurate three-dimensional reconstruction can be completed on closely, reduce the operation complexity of three-dimensional reconstruction.
Secondly, the present invention provide three-dimensional reconstruction apparatus consider luminescence unit distance objective it is nearer when, light source is irradiated to
Body surface is uneven, its incident light direction that cannot be assumed to be each point on directional light or spot light, and object
It is also different problems, by reconstruction unit, the image letter that luminous model and image acquisition unit according to luminescence unit are obtained
Breath, rebuilds, it is contemplated that the luminous model of luminescence unit to target object, improves the precision and efficiency of three-dimensional reconstruction.
On the other hand, based on same inventive concept, the embodiment of the present invention additionally provides a kind of three-dimensional rebuilding method, is used to drop
The operation complexity of low three-dimensional reconstruction, improves the precision and efficiency of three-dimensional reconstruction, and the method includes：
The light on and off of the multiple luminescence units of control, when each luminescence unit is lighted, Synchronization Control image acquisition unit shoots
Target object, obtains the image information of the lower target object of each luminescence unit irradiation；Multiple luminescence units are arranged on image acquisition
Around unit；
According to image information and the luminous model of luminescence unit, target object is rebuild；The luminous model is every
Output intensity, the relation of shooting angle on output intensity and each exit direction on individual luminescence unit primary optical axis.
Compared with prior art, three-dimensional rebuilding method provided in an embodiment of the present invention, it is first, multiple luminous by controlling
The light on and off successively of unit, when each luminescence unit is lighted, Synchronization Control image acquisition unit photographic subjects object obtains each
Image information under luminescence unit irradiation；Multiple luminescence units are evenly provided on around described image acquiring unit, can be with
Quick scanning process is realized, accurate three-dimensional reconstruction can be completed on closely, reduce the operation complexity of three-dimensional reconstruction；
Secondly, three-dimensional rebuilding method provided in an embodiment of the present invention consider luminescence unit distance objective it is nearer when, light source is irradiated to thing
Body surface face is uneven, its incident light direction that cannot be assumed to be each point on directional light or spot light, and object
It is different problems, the image information obtained by reconstruction unit, luminous model and camera according to luminescence unit, to object
Body is rebuild, it is contemplated that the luminous model of luminescence unit, improves the precision and efficiency of three-dimensional reconstruction.
Brief description of the drawings
Accompanying drawing described herein is used for providing a further understanding of the present invention, constitutes the part of the application, not
Constitute limitation of the invention.In the accompanying drawings：
Fig. 1 is the structural representation of three-dimensional reconstruction apparatus in the embodiment of the present invention；
Fig. 2 be the embodiment of the present invention luminous model in the schematic diagrames that change with the change of the effective half-angles of LED of g；
Fig. 3 is the schematic diagram of each parameter during three-dimensional reconstruction in the embodiment of the present invention；
Fig. 4 is to rebuild gained coordinate and actual coordinate transition diagram in the embodiment of the present invention during three-dimensional reconstruction；
Fig. 5 is the schematic flow sheet of three-dimensional rebuilding method in the embodiment of the present invention；
Fig. 6 is the schematic flow sheet of scaling method in three-dimensional rebuilding method in the embodiment of the present invention.
Specific embodiment
It is right with reference to implementation method and accompanying drawing to make the object, technical solutions and advantages of the present invention become more apparent
The present invention is described in further details.Here, exemplary embodiment of the invention and its illustrating for explaining the present invention, but simultaneously
It is not as a limitation of the invention.
Inventor has found：Existing light measurement three-dimensional reconstruction employ spot light or source of parallel light it is assumed that
Do not consider the characteristics of luminescence of light source itself, when object distance light source farther out or light source meets directional light or spot light in itself
During characteristic, existing method can obtain preferable three-dimensional reconstruction effect, but this is higher to light source oneself requirement, and parallel to obtain
Light, light source needs larger volume in itself, such as uses LED array device；When LED distance objectives are nearer, LED light source now
It is uneven to be irradiated to body surface, therefore cannot be assumed to be the incident light direction of each point in source of parallel light, and object
It is also different.Therefore, using the existing three-dimensional reconstruction apparatus based on light measurement, it is difficult to obtain good three-dimensional reconstruction effect.
The existing closely light measurement three-dimensional reconstruction based on single source, simply in image processing stage to image
Illuminance uniformity correct, for improving three-dimensional reconstruction effect, luminous model not in view of light source itself and enter
The factors such as firing angle degree, irradiation distance, therefore be also difficult to obtain preferable three-dimensional reconstruction effect.
Because inventor is found that above mentioned problem, inventor proposes a kind of three-dimensional reconstruction apparatus, method by the present invention
And scaling method, wherein, the reconstructing device can be using single luminescence unit (for example：Single led lamp bead is (rather than array, battle array
Row volume is big, not easy of integration)) as the method for light measurement one direction light source, and then realize that the accurate three-dimensional closely gone up is rebuild.
Due to the characteristic of single luminescence unit (LED light source), single luminescence unit (LED light source) is difficult to obtain the uniform photograph of larger area
Degree field, the luminous model of itself is related to its Illumination Distribution to single luminescence unit (LED light source), and this factor is existing
Light measurement three-dimensional rebuilding method in be not difficult to obtain preferable three-dimensional reconstruction effect it is considered that thus resulting in.It is of the invention then
This factor is dissolved into whole light measurement three-dimensional reconstruction framework, for example：By the modeling to light source, Exact recovery goes out photograph
The incident intensity of each point and direction in degree field, and then realize the calculating of accurate three-dimensional normal direction and process of reconstruction.Due to adopting
With single luminescence unit (for example：Single led lamp bead) as one direction light source, therefore whole device compact, it is easy to collect
Into cost is lower.
Specifically, the present invention proposes one kind according to luminescence unit (for example for the weak point in background above technology：
LED luminous model) solves the accurate light measurement Problems of Reconstruction under the irradiation of closely non-collimated light sources, is situated between in detail below
Continue explanation.
Fig. 1 is the structural representation of three-dimensional reconstruction apparatus in the embodiment of the present invention；As shown in figure 1, the device includes：
Image acquisition unit 1, the image information for obtaining target object；
Multiple luminescence units 2, are arranged on around image acquisition unit；
Synchronous control unit 3, the light on and off for controlling multiple luminescence units, when each luminescence unit is lighted, synchronous control
Imaged acquiring unit photographic subjects object, obtains the image information of the lower target object of each luminescence unit irradiation；
Reconstruction unit 4, for the luminous model according to image information and luminescence unit, rebuilds to target object；Institute
It is output intensity, the shooting angle on output intensity and each exit direction on each luminescence unit primary optical axis to state luminous model
Relation.
Compared with laser scanning of the prior art and structure light scan reconstructing device, provided in an embodiment of the present invention three
Dimension reconstructing device, with following Advantageous Effects：
Firstly, since the three-dimensional reconstruction apparatus that invention is provided include an image acquisition unit and multiple luminescence unit structures
Into system architecture is simple, and cost is extremely low, by synchronous control unit, when controlling each luminescence unit to irradiate, controls image
Acquiring unit sync pulse jamming target object, it is possible to achieve quick scanning process, can complete accurate Three-dimensional Gravity on closely
Build, reduce the operation complexity of three-dimensional reconstruction.
Secondly, the present invention provide three-dimensional reconstruction apparatus consider luminescence unit distance objective it is nearer when, light source is irradiated to
Body surface is uneven, its incident light direction that cannot be assumed to be each point on directional light or spot light, and object
It is also different problems, the image information obtained by reconstruction unit, luminous model and camera according to luminescence unit, to target
Object is rebuild, it is contemplated that the luminous model of luminescence unit, improves the precision and efficiency of three-dimensional reconstruction.
During specific implementation, image acquisition unit 1 can be camera, and camera, can when the image information of target object is obtained
With by the way of offline storage.Offline storage mode can be mutually used, shooting speed can be so improved, it is ensured that camera and luminous
Unit 2 is (for example：LED lamp bead) precise synchronization, the property treatment again of photo needed for having shot.It is of course also possible to use online pass
Defeated mode.
During specific implementation, luminescence unit 2 can be LED lamp bead, and LED lamp bead can be evenly provided on and obtain single with image
Unit is on the circumference in the center of circle, the number of LED lamp bead is no less than 4.
During specific implementation, the embodiment of the present invention is by taking the LED for commonly using as an example, it would however also be possible to employ other classification light sources, but needs pre-
Know the luminous model of the light source；The embodiment of the present invention can use all kinds of LED, such as visible ray, non-visible light, wherein, non-visible light
(such as infrared light), by taking the collection of face three-dimensional information as an example, dazzling problem can be avoided using infrared LED light source, and Consumer's Experience is more
It is close friend.
During specific implementation, such as luminescence unit 2 uses non-visible light source, in order to obtain the color and vein information of target object,
An image acquisition unit can be further added by, dedicated for the shooting of coloured image, color texture three-dimensional reconstruction result is mapped to,
Realize the acquisition of colorful three-dimensional model.
During specific implementation, luminescence unit 2 is (for example：LED lamp bead) quantity and position can be in optimized selection, no less than 4
Individual, the lighting angle of LED lamp bead can be in optimized selection according to the size of target object and operating distance.
It is flat without being simulated using area array light source due to only having used single led lamp bead as single arbitrary source
Row light, thus whole system compact, it is easier to operate, accurate three-dimensional information can be completed on closely and are obtained.
To sum up, three-dimensional reconstruction apparatus provided in an embodiment of the present invention, due to one figure of three-dimensional reconstruction apparatus that invention is provided
As acquiring unit (for example：Camera) and some LED lamp beads composition, single led chip is employed as single source, system architecture
Simply, cost is extremely low, by a synchronous control unit 3, camera sync pulse jamming figure when controlling each LED lamp bead to irradiate
Piece, using LED and image acquisition unit (for example：Camera) Strategy For Synchronization Control, be capable of achieving quick image acquisition process, no
Need using larger, the distant source of parallel light of volume, accurate three can be completed on closely using several LED lamp beads
Dimension information reconstruction, so as to realize the miniaturization of whole device, the need for meeting the collection of Table top type three-dimensional information, reduces Three-dimensional Gravity
Build the volume of device.Therefore, the three-dimensional reconstruction apparatus that the present invention is provided have that simple structure, speed is fast, working stability, precision
The high, significant advantage of compact, can be used for the quick three-dimensional digitlization of face, personal belongings, and low cost can be commonly to disappear
Expense person is received, with wide market prospects.
In one embodiment, LED lamp bead is infrared LED lamp bead, and image acquisition unit uses infrared fileter.It is specific real
Shi Shi, such as uses infrared LED light source, camera to load corresponding infrared fileter, improves the antijamming capability of ambient light.
During specific implementation, the main working process of reconstruction unit 4 is as follows：
(1) initial alignment of systematic parameter and estimation, including camera parameter are demarcated and each LED light source coordinate and its key light
Direction of principal axis is estimated；
(2) source modeling：According to the LED type specification for being used, the luminous moulds of LED are set up by luminous half angle value of LED
Type；
(3) three-dimensional normal direction is rebuild：Initialization each point coordinates of body surface, tries to achieve body surface each according to the luminous models of LED
Point incident intensity and direction, based on given reflection model, such as Lambert's model is rebuild to body surface；
(4) optimization of three-dimensional normal direction value：The object dimensional coordinate back substitution that will be tried to achieve, iteration is calculated, until iteration bar
Part terminates, and further improves the normal direction precision recovered；
(5) 3D shape is rebuild：According to acquired three-dimensional normal direction information, graded using surface area and obtain by general categories
Take accurate surface three dimension shape information.
The specific work process to reconstruction unit 4 describes in detail below.
Before being introduced to reconstruction unit 4, important models and calculation that the embodiment of the present invention is used are introduced first
Method：
The several important models used in the present invention program are introduced first：
(1) reflection model mentioned in the embodiment of the present invention can be Lambertian reflection model：
【Model 1】；
Wherein, IrReflective light intensity is represented, l represents light source exit direction, and | L | represents light source output intensity, and d represents eye point
To the distance of incidence point, n then represents body surface normal direction.During three-dimensional modeling, Lambert's model is a kind of conventional reflection mould
Type, the present invention is also illustrated as example.
(2) model that lights can be the luminous models of LED：
Iθ=I0cosg-1θ【Model 2】
Wherein, I0Represent the output intensity on primary optical axis, IθEach exit direction l is expressed as with primary optical axis into θ angles direction
Output intensity, then θ=arccos (lmain* l), wherein,It is LED key light direction of principal axis, j represents that j-th LED, l represent light
Source exit direction, its model of the LED of different size is also different, and the present invention is illustrated by example of this model.Wherein g is with LED
Effective half angle θharfChange and change,With shown in effective half-angle variation tendency Fig. 2.
(3) composite model of combination Lambert's model proposed by the present invention and the luminous models of LED【Model 3】
With reference to Lambert's modelModel I luminous with LEDθ=I0cosg-1θ.Then, when emergent light and primary optical axis
When the angle in direction is θ,So, and then
Secondly, the several important methods used in the present invention program are introduced with reference to Fig. 3：
(1) image coordinate is converted to camera coordinates (namely world coordinates of this example) algorithm【Algorithm 1】：
Assuming that known object point pixel coordinate in the picture be (u, v) and camera focus f, and the object point away from
It is Dist with a distance from camera plane Z0 planes.So according to camera perspective imaging model：
Wherein k in certain z to may be assumed that to be a constant in distance range, i.e. pixel distance and actual physics in the picture
The multiplication factor of distance, it is assumed that the space coordinates of object point correspondence camera coordinates system is (xo,yo,zo)。
(2) object each point incidence light direction is calculated【Algorithm 2】With light source to object each point distance【Algorithm 3】
Assuming that the coordinate of body surface every is (xo,yo,zo), it is known that the coordinate of each light source is(in Fig. 3
It is expressed as (xl,yl,zl)), j represents j-th light source (LED).So each incident light direction put isUnit direction vector is obtained after normalization【Algorithm 2】.L modulus can be obtained away from
From d=| L |【Algorithm 3】.Each parameter is as shown in Figure 3 in systems.
The important models and algorithm used in the embodiment of the present invention are described above, reconstruction unit 4 is carried out in detail below
Introduce.
In one embodiment, reconstruction unit 4 specifically for：
According to image information, the initial three-dimensional normal direction value of each reconstruction point of target object surface is estimated；
Luminous model and reflection model based on luminescence unit, are initial value with initial three-dimensional normal direction value, according to each reconstruction
The light source incidence light intensity and incident direction of point, find the optimal three-dimensional normal direction value of each reconstruction point；
According to optimal three-dimensional normal direction value, three-dimensional reconstruction is carried out to target object；
The reflection model be reflective light intensity and light source output intensity, light source exit direction, eye point to incidence point away from
From the relation of, target object surface normal direction value.
During specific implementation, it is nearer that three-dimensional reconstruction apparatus provided in an embodiment of the present invention consider luminescence unit distance objective
When, it is uneven that light source is irradiated to body surface, and it cannot be assumed to be each point on directional light or spot light, and object
Incident light direction be also different problems, by reconstruction unit 4, the figure that luminous model and camera according to luminescence unit are obtained
As information, target object is rebuild, that is, considers the factors such as the luminous model of distance, incident light direction and luminescence unit,
Improve the precision and efficiency of three-dimensional reconstruction.
During specific implementation, reconstruction unit 4 can be connected work with the equipment such as PC, and it is (personal that Image Information Processing is placed on into PC
Computer) end, also can built-in information processing module carry out real-time processing.Therefore, three-dimensional reconstruction dress provided in an embodiment of the present invention
Put, low cost, compact, can be used as independent terminal works, it is also possible to it is integrated with all kinds of intelligent terminals in, realize three
The function that dimension information gathering is rebuild.
In one embodiment, luminous model and reflection model based on luminescence unit, are first with initial three-dimensional normal direction value
Value, light source incidence light intensity and incident direction according to each reconstruction point find the optimal three-dimensional normal direction value of each reconstruction point, wrap
Include：
It is initial value with initial three-dimensional normal direction value based on light source composite model, iteration finds optimal three-dimensional normal direction value, for every
Individual iteration cycle is performed both by following operation：
Three-dimensional normal direction value, the space coordinates of each luminescence unit and key light direction of principal axis according to each reconstruction point, calculate every
The distance of the light source incidence direction of individual reconstruction point and the reconstruction point to each luminescence unit；
Based on light source composite model, light source incidence direction and the reconstruction point to each luminescence unit according to each reconstruction point
Distance, calculate the three-dimensional normal direction value of each reconstruction point, until find each reconstruction point depth information change in preset range
When interior, corresponding three-dimensional normal direction value, as optimal three-dimensional normal direction value；
Light source composite model is set up according to the luminous model and reflection model of luminescence unit, and the light source composite model is anti-
Output intensity, the exit direction on primary optical axis, each exit direction, the object penetrated on light intensity and each luminescence unit primary optical axis
Body surface normal value, the relation of the distance of eye point to incidence point.
Current light measurement method assumes light source for source of parallel light or ideal point light source, and this hypothesis is closely
It is invalid under LED light source irradiation, thus is difficult to obtain preferable three-dimensional reconstruction effect.And the present invention is using light measurement three
While dimension reconstruction technique, source model characteristic is more considered, realize the independent calculating intensity of illumination to each point on object
With incident light direction, and then its accurate normal direction value is obtained by optimization method, whole model is more accurate, it is thus possible to
Obtain more preferably three-dimensional reconstruction result.
In one embodiment, the model that lights is the luminous models of LED, and light source composite model is LED light source composite model,
LED light source composite model is：
Wherein, IrIt is LED lamp bead light source reflective light intensity, I0It is the output intensity on LED lamp bead primary optical axis, lmainIt is LED
Pearl key light direction of principal axis, l is LED lamp bead light source exit direction, g with the luminous models of LED effective half angle θharfChange and change,N for target object surface reconstruction point three-dimensional normal direction value, d be reconstruction point to each luminescence unit away from
From.
How to implement for a better understanding of the present invention, with reference to Fig. 3 and Fig. 4, to the operation principle of reconstruction unit 4
And process describes in detail.
(1) normal estimation and iteration optimization
By after demarcating steps, we can obtain the system core parameter of optimization (for example：Each luminescence unit it is optimal
Space coordinates and key light direction of principal axis), below we the reconstruction of three-dimensional normal direction can be carried out to target object to be measured.
A) every distance to camera Z0 planes of target object surface to be measured, is initialized as into Dist first (will object
It is placed in and demarcates near distance in advance, and do not consider the object just influence of change in itself first).So according to pixel coordinate and phase
The relation of machine coordinate【Algorithm 1】, body surface coordinate can be initialized as (ku, kv, Dist)；
B), the middle scaling method for providing optimizes parameters obtained according to embodiments of the present invention：Luminescence unit (such as LED lamp bead)
Space coordinatesLED key light direction of principal axisIt is utilized respectively light source direction calculating method【Algorithm 2】With light source objects away from
From algorithm【Algorithm 3】, calculate the light source direction l of every point (reconstruction point) with the point to light source apart from d.
C), the key light direction of principal axis with reference to obtained by calculatingWith each point incidence light direction l of initial calculation in above-mentioned steps b)
With apart from d, light source direction composite model is substituted into【Model 3】, can be to the method for each point of body surface
Initial estimation is carried out to n.
D), the normal direction value according to initial estimation, by surface Integration Strategy, can obtain the relative dimensional shape of object each point
(u,v,Zdep), wherein ZdepRepresent depth information of the object each point relative to the plane of delineation.
E) required relative coordinate, is converted into true three-dimension coordinate
As shown in figure 4, ZdepApparently not one real depth information, i.e. (u, v, Zdep) can not reflect that object is truly big
Small, it can only reflect the relative shape of object.Assuming that the real depth information of the point is Zreal, then object with true shape
Shape is represented by (x, y, Zreal).It is with (u, v, Zdep) there is a scaling relation, zoom factor
Simultaneously as the real depth information of the point is Zreal, then according to Z in the present inventionrealDefinition, the dot image
Three-dimensional coordinate (xo,yo,zo) it is represented by (x, y, Zreal+Dist).By【Algorithm 1】Understand, the space coordinates and pixel coordinate
Relation is represented byWherein f is the F+Z in Fig. 4dep, F represents focal length.
Obvious k1=k2, there is relationIn this way, object each point true coordinate can be tried to achieve
(xo′,yo′,zo′)；
F), by required object each point true three-dimension coordinate (x in e) stepo′,yo′,zo′) again substitute into (b)~(e) carry out
Iterative calculation, until apart from z (depth information of each reconstruction point of body surface, Ke Yishi：Object each point is relative to the plane of delineation
Depth information Zdep；The real depth information that can also be depth information of the object each point relative to reference planes, i.e. each point is
Zreal) change tend to a smaller assigned error scope for stabilization, iterated conditional convergence.
G), the final three-dimensional normal direction information according to obtained by iteration, by surface Integration Strategy, can obtain final accurate
Object dimensional model information.
A kind of three-dimensional rebuilding method is additionally provided based on same inventive concept, in the embodiment of the present invention, such as following implementation
Example is described.Because the principle of three-dimensional rebuilding method solve problem is similar to three-dimensional reconstruction apparatus, thus three-dimensional rebuilding method reality
The implementation that may refer to three-dimensional reconstruction apparatus is applied, part is repeated and is repeated no more.It is used below, term " unit " or " mould
Block " can realize the combination of the software and/or hardware of predetermined function.Although the device described by following examples is preferably with soft
Part is realized, but hardware, or the realization of the combination of software and hardware is also that may and be contemplated.
Fig. 5 is the schematic flow sheet of three-dimensional rebuilding method in the embodiment of the present invention, as shown in figure 5, the method is including as follows
Step：
Step 101：The light on and off of the multiple luminescence units of control, when each luminescence unit is lighted, Synchronization Control image is obtained
Unit photographs target object, obtains the image information of the lower target object of each luminescence unit irradiation；Multiple luminescence units are arranged on
Around image acquisition unit；
Step 102：According to image information and the luminous model of luminescence unit, target object is rebuild；It is described luminous
Model is output intensity, the pass of shooting angle on output intensity and each exit direction on each luminescence unit primary optical axis
System.
In one embodiment, according to image information and the luminous model of luminescence unit, target object is rebuild, wraps
Include：
According to image information, the initial three-dimensional normal direction value of each reconstruction point of target object surface is estimated；
Luminous model and reflection model based on luminescence unit, are initial value with initial three-dimensional normal direction value, according to each reconstruction
The light source incidence light intensity and incident direction of point, find the optimal three-dimensional normal direction value of each reconstruction point；
According to optimal three-dimensional normal direction value, three-dimensional reconstruction is carried out to target object；The reflection model is reflective light intensity and light
Source output intensity, light source exit direction, the distance of eye point to incidence point, the relation of target object surface normal direction value.
In one embodiment, luminous model and reflection model based on luminescence unit, are first with initial three-dimensional normal direction value
Value, light source incidence light intensity and incident direction according to each reconstruction point find the optimal three-dimensional normal direction value of each reconstruction point, wrap
Include：
It is initial value with initial three-dimensional normal direction value based on light source composite model, iteration finds optimal three-dimensional normal direction value, for every
Individual iteration cycle is performed both by following operation：
Three-dimensional normal direction value, the space coordinates of each luminescence unit and key light direction of principal axis according to each reconstruction point, calculate every
The distance of the light source incidence direction of individual reconstruction point and the reconstruction point to each luminescence unit；
Based on light source composite model, light source incidence direction and the reconstruction point to each luminescence unit according to each reconstruction point
Distance, calculate the three-dimensional normal direction value of each reconstruction point, until find each reconstruction point depth information change in preset range
When interior, corresponding three-dimensional normal direction value, as optimal three-dimensional normal direction value；
Light source composite model is set up according to the luminous model and reflection model of luminescence unit, and the light source composite model is anti-
Output intensity, the exit direction on primary optical axis, each exit direction, the object penetrated on light intensity and each luminescence unit primary optical axis
Body surface normal value, the relation of the distance of eye point to incidence point.
Compared with prior art, three-dimensional rebuilding method provided in an embodiment of the present invention, it is first, multiple luminous by controlling
The light on and off successively of unit, when each luminescence unit is lighted, Synchronization Control image acquisition unit photographic subjects object obtains each
Image information under luminescence unit irradiation；Multiple luminescence units are evenly provided on around described image acquiring unit, can be with
Quick scanning process is realized, accurate three-dimensional reconstruction can be completed on closely, reduce the operation complexity of three-dimensional reconstruction；
Secondly, three-dimensional rebuilding method provided in an embodiment of the present invention consider luminescence unit distance objective it is nearer when, light source is irradiated to thing
Body surface face is uneven, its incident light direction that cannot be assumed to be each point on directional light or spot light, and object
It is different problems, the image information obtained by reconstruction unit, luminous model and camera according to luminescence unit, to object
Body is rebuild, it is contemplated that the luminous model of luminescence unit, improves the precision and efficiency of three-dimensional reconstruction.
Fig. 6 is the schematic flow sheet to three-dimensional reconstruction apparatus scaling method in the embodiment of the present invention, as shown in fig. 6, the mark
Determine method, comprise the following steps：
Step 201：Space coordinates and key light direction of principal axis of each luminescence unit relative to image acquisition unit according to a preliminary estimate；
Step 202：One planar object is placed on image acquisition unit front, synchronous control unit control is multiple luminous single
The light on and off successively of unit, when each luminescence unit is lighted, Synchronization Control image acquisition unit camera plane object obtains each hair
Image information under light unit irradiation；Optical axis of the planar object perpendicular to image acquisition unit；
Step 203：Luminous model and reflection model based on luminescence unit, with space coordinates according to a preliminary estimate and primary optical axis
Direction is initial value, according to the light source incidence light intensity and incident direction of each reconstruction point on image information, planar object, finds each
The optimal spatial coordinate and key light direction of principal axis of luminescence unit.
In one embodiment, luminous model and reflection model based on luminescence unit, with space coordinates and primary optical axis
Exit direction be initial value, according to the light source incidence light intensity and incident direction of each reconstruction point on image information, planar object, look for
To the optimal spatial coordinate and key light direction of principal axis of each luminescence unit, including：
It is initial value with space coordinates according to a preliminary estimate and key light direction of principal axis based on light source composite model, iteration finds each
The optimal spatial coordinate and key light direction of principal axis of luminescence unit, following operation is performed both by for each iteration cycle：
The space coordinates and key light direction of principal axis of image information, each luminescence unit according to planar object, calculate object
The distance of the light source incidence direction of each reconstruction point of body and the reconstruction point to each luminescence unit；
The light source incidence direction and the reconstruction of luminous model and reflection model based on luminescence unit according to each reconstruction point
Point arrives the distance of each luminescence unit, the three-dimensional normal direction value of all reconstruction points of Calculation Plane body surface；
Three-dimensional normal direction value to all reconstruction points carries out mean square deviation calculating, the three-dimensional normal direction value until finding all reconstruction points
Mean square deviation it is minimum when, the space coordinates and key light direction of principal axis of corresponding each luminescence unit, as optimal spatial coordinate and master
Optical axis direction.
How embodiment is implemented for a better understanding of the present invention, and Fig. 3 and Fig. 4 is may be referred to below, and the present invention is implemented
The operation principle and process of the scaling method of the three-dimensional reconstruction apparatus that example is provided describe in detail.
A), first, camera internal parameter is demarcated, the inner parameter of camera can be obtained using conventional method, such as
The parameters such as focal length F, center position.
B), as shown in figure 3, being world coordinate system origin with camera photocentre with camera coordinates system as world coordinate system, it is assumed that
Output intensity on single led lamp primary optical axis is I0, according to LED on device with relative position (the original hypothesis LED master of camera
Optical axis direction encapsulates plane perpendicular to LED, and attitude of the LED module in whole device, angle be can reconcile it is measurable),
Substantially space coordinates of all LED relative to camera can be estimatedKey light direction of principal axisJ represents j-th LED.
C), light source coordinates and primary optical axis direction calculating and optimization：
I places it in camera front using a planar object (such as blank sheet of paper) with lambertian reflection characteristics, and attitude is hung down
Directly in camera optical axis, it is assumed that the plane is Dist with the distance of camera Z0 planes.Control LED flickers successively, and utilizes camera
The image of the module photograph plane under different LED illuminations, obtains image Ij(u,v)。
Ii uses above-mentioned pixel coordinate to world coordinate system transformation approach【Algorithm 1】, each picture in image can be tried to achieve
Corresponding space coordinates (the x of vegetarian refreshmentso,yo,zo) it is utilized respectively light source direction calculating method【Algorithm 2】With light source objects distance algorithm【Calculate
Method 3】, calculate the light source direction l of each pixel with the point to light source apart from d.
Iii estimates and the key light direction of principal axis obtained by calculating by more thanWith each point incident direction l light source is substituted into apart from d
Direction composite model【Model 3】, plane normal direction n can be estimated.
Iv is plane, i.e. each point normal direction all same thereon due to known object of experiment, when plane is perpendicular to camera optical axis,
Its normal direction is directed to the opposite direction of the Z in camera optical axis direction, therefore, our settable minimum energy equations (can be to each
Point normal direction n does deviation integration), with initial estimation parameter as parameter to be optimized, the minimum desired value of mean square deviation is calculated with normal direction, no
It is disconnected that above-mentioned initial estimation parameter is optimized.Required plane closest to true planar is represented when energy equation is minimized.
Now estimate each value of gained close to actual value.
【Explanation】：Parameter to be estimated includes：Each LED space coordinatesLED key light direction of principal axisPass through
The above method, tries to achieve optimal spatial coordinate and key light direction of principal axis.
The scaling method of three-dimensional reconstruction apparatus provided in an embodiment of the present invention, is obtained by the way that a planar object is placed on into image
Unit front is taken, synchronous control unit controls the light on and off successively of the multiple luminescence unit, when each luminescence unit is lighted, together
Step control described image acquiring unit shoots the planar object, obtains the image information under the irradiation of each luminescence unit；It is described
Planar object perpendicular to image acquisition unit optical axis, luminous model and reflection model based on luminescence unit, with according to a preliminary estimate
Space coordinates and key light direction of principal axis be initial value, according to the light source incidence of each reconstruction point on described image information, planar object
Light intensity and incident direction, find the optimal spatial coordinate and key light direction of principal axis of each luminescence unit, and the system to initial estimation is closed
Bond parameter is optimized estimation, improves the accuracy of three-dimensional reconstruction apparatus parameter, it is ensured that three-dimensional reconstruction apparatus follow-up three
Tie up the precision and efficiency rebuild.
In sum, technical scheme provided in an embodiment of the present invention, with following Advantageous Effects：
1) current light measurement method assumes light source for source of parallel light or ideal point light source, and this hypothesis is closely
It is invalid under LED light source irradiation, thus is difficult to obtain preferable three-dimensional reconstruction effect.And the present invention is using light measurement three
While dimension reconstruction technique, source model characteristic is more considered, realize the independent calculating intensity of illumination to each point on object
With incident light direction, whole model is more accurate, it is thus possible to obtain more preferably three-dimensional reconstruction result；
2) due to only having used single led light source as single arbitrary source, without being simulated using area array light source
Directional light, thus whole system compact, it is easier to operate, accurate three-dimensional information can be completed on closely and are obtained；
3) on the basis of initial parameter estimation, optimization and iterative strategy are introduced, is capable of achieving more accurately system core
Parameter Estimation and three-dimensional normal direction are calculated；
4) using LED and the Strategy For Synchronization Control of camera, it is capable of achieving quick image acquisition process；
5) system cost is low, compact, can be used as independent terminal works, it is also possible to integrated to be set with all kinds of intelligent terminals
In standby, the function of three-dimensional information collection is realized.
Present invention focuses on proposing a kind of light measurement three-dimensional reconstruction apparatus and method based on luminous model, Neng Gou
Under the conditions of the non-standard light source of non-parallel light/non-spot light, realize accurate object each point incident intensity and direction
Calculating, obtain more preferable 3D than existing methods and rebuild effect, so as to the high-precision 3D collections for also providing low cost set
It is standby.
Obviously, those skilled in the art should be understood that each module, device or each step of the above-mentioned embodiment of the present invention
Can be realized with general computing device, they can be concentrated on single computing device, or be distributed in multiple calculating
On the network that device is constituted, alternatively, they can be realized with the executable program code of computing device, it is thus possible to
It is stored in being performed by computing device in storage device, and in some cases, can be with suitable different from herein
Sequence performs shown or described step, or they are fabricated to each integrated circuit modules respectively, or by them
Multiple modules or step are fabricated to single integrated circuit module to realize.So, the embodiment of the present invention is not restricted to any specific
Hardware and software combine.
The preferred embodiments of the present invention are these are only, is not intended to limit the invention, for those skilled in the art
For member, the embodiment of the present invention can have various modifications and variations.It is all within the spirit and principles in the present invention, made it is any
Modification, equivalent, improvement etc., should be included within the scope of the present invention.
Claims (12)
1. a kind of three-dimensional reconstruction apparatus, it is characterised in that including：
Image acquisition unit, the image information for obtaining target object；
Multiple luminescence units, are arranged on around described image acquiring unit；
Synchronous control unit, the light on and off for controlling the multiple luminescence unit, when each luminescence unit is lighted, Synchronization Control
Described image acquiring unit photographic subjects object, obtains the image information of the lower target object of each luminescence unit irradiation；
Reconstruction unit, for the luminous model according to described image information and luminescence unit, rebuilds to target object；It is described
Luminous model is output intensity, the shooting angle on output intensity and each exit direction on each luminescence unit primary optical axis
Relation.
2. three-dimensional reconstruction apparatus as claimed in claim 1, it is characterised in that the reconstruction unit specifically for：
According to described image information, the initial three-dimensional normal direction value of each reconstruction point of target object surface is estimated；
Luminous model and reflection model based on luminescence unit, with the initial three-dimensional normal direction value as initial value, according to each reconstruction
The light source incidence light intensity and incident direction of point, find the optimal three-dimensional normal direction value of each reconstruction point；
According to the optimal three-dimensional normal direction value, three-dimensional reconstruction is carried out to target object；
The reflection model is reflective light intensity and light source output intensity, light source exit direction, the distance of eye point to incidence point, mesh
Mark the relation of body surface normal direction value.
3. three-dimensional reconstruction apparatus as claimed in claim 2, it is characterised in that luminous model and reflection mould based on luminescence unit
Type, with the initial three-dimensional normal direction value as initial value, light source incidence light intensity and incident direction according to each reconstruction point find each
The optimal three-dimensional normal direction value of reconstruction point, including：
Based on light source composite model, with the initial three-dimensional normal direction value as initial value, iteration finds optimal three-dimensional normal direction value, for every
Individual iteration cycle is performed both by following operation：
Three-dimensional normal direction value, the space coordinates of each luminescence unit and key light direction of principal axis according to each reconstruction point, calculate each weight
The distance of the light source incidence direction laid foundations and the reconstruction point to each luminescence unit；
Based on light source composite model, light source incidence direction and the reconstruction point according to each reconstruction point to each luminescence unit away from
From, the three-dimensional normal direction value of each reconstruction point is calculated, when the depth information for finding each reconstruction point changes within a preset range,
Corresponding three-dimensional normal direction value, as optimal three-dimensional normal direction value；
The light source composite model is set up according to the luminous model and reflection model of luminescence unit, and the light source composite model is anti-
Output intensity, the exit direction on primary optical axis, each exit direction, the object penetrated on light intensity and each luminescence unit primary optical axis
Body surface normal value, the relation of the distance of eye point to incidence point.
4. three-dimensional reconstruction apparatus as described in claims 1 to 3 is any, it is characterised in that described image acquiring unit is camera,
The camera is stored when the image information of target object is obtained by the way of offline storage.
5. three-dimensional reconstruction apparatus as described in claims 1 to 3 is any, it is characterised in that the luminescence unit is LED lamp bead,
The LED lamp bead is evenly provided on around described image acquiring unit, and the number of the LED lamp bead is no less than 4.
6. three-dimensional reconstruction apparatus as claimed in claim 5, it is characterised in that the luminous model is that LED lights model, described
Light source composite model is LED light source composite model, and the LED light source composite model is：
Wherein, IrIt is LED lamp bead light source reflective light intensity, I0It is the output intensity on LED lamp bead primary optical axis, lmainIt is LED lamp bead master
Optical axis direction, l is LED lamp bead light source exit direction, g with the luminous models of LED effective half angle θharfChange and change,N for target object surface reconstruction point three-dimensional normal direction value, d be reconstruction point to each luminescence unit away from
From.
7. three-dimensional reconstruction apparatus as claimed in claim 5, it is characterised in that the LED lamp bead is infrared LED lamp bead；It is described
Image acquisition unit uses infrared fileter.
8. a kind of three-dimensional rebuilding method, it is characterised in that including：
The light on and off of the multiple luminescence units of control, when each luminescence unit is lighted, Synchronization Control image acquisition unit photographic subjects
Object, obtains the image information of the lower target object of each luminescence unit irradiation；The multiple luminescence unit is arranged on described image
Around acquiring unit；
According to described image information and the luminous model of luminescence unit, target object is rebuild；The luminous model is every
Output intensity, the relation of shooting angle on output intensity and each exit direction on individual luminescence unit primary optical axis.
9. three-dimensional rebuilding method as claimed in claim 8, it is characterised in that according to described image information and the hair of luminescence unit
Light model, rebuilds to target object, including：
According to described image information, the initial three-dimensional normal direction value of each reconstruction point of target object surface is estimated；
Luminous model and reflection model based on luminescence unit, with the initial three-dimensional normal direction value as initial value, according to each reconstruction
The light source incidence light intensity and incident direction of point, find the optimal three-dimensional normal direction value of each reconstruction point；
According to the optimal three-dimensional normal direction value, three-dimensional reconstruction is carried out to target object；
The reflection model is reflective light intensity and light source output intensity, light source exit direction, the distance of eye point to incidence point, mesh
Mark the relation of body surface normal direction value.
10. three-dimensional rebuilding method as claimed in claim 9, it is characterised in that luminous model and reflection based on luminescence unit
Model, with the initial three-dimensional normal direction value as initial value, light source incidence light intensity and incident direction according to each reconstruction point find every
The optimal three-dimensional normal direction value of individual reconstruction point, including：
Based on light source composite model, with the initial three-dimensional normal direction value as initial value, iteration finds optimal three-dimensional normal direction value, for every
Individual iteration cycle is performed both by following operation：
Three-dimensional normal direction value, the space coordinates of each luminescence unit and key light direction of principal axis according to each reconstruction point, calculate each weight
The distance of the light source incidence direction laid foundations and the reconstruction point to each luminescence unit；
Based on light source composite model, light source incidence direction and the reconstruction point according to each reconstruction point to each luminescence unit away from
From, the three-dimensional normal direction value of each reconstruction point is calculated, when the depth information for finding each reconstruction point changes within a preset range,
Corresponding three-dimensional normal direction value, as optimal three-dimensional normal direction value；
The light source composite model is set up according to the luminous model and reflection model of luminescence unit, and the light source composite model is anti-
Output intensity, the exit direction on primary optical axis, each exit direction, the object penetrated on light intensity and each luminescence unit primary optical axis
Body surface normal value, the relation of the distance of eye point to incidence point.
11. three-dimensional rebuilding methods as claimed in claim 10, it is characterised in that in the three-dimensional normal direction according to each reconstruction point
Value, the space coordinates of each luminescence unit and key light direction of principal axis, calculate light source incidence direction and reconstruction point of each reconstruction point
To the distance of each luminescence unit, also include before：
Space coordinates and key light direction of principal axis of each luminescence unit relative to image acquisition unit according to a preliminary estimate；
One planar object is placed on image acquisition unit front, synchronous control unit controls the multiple luminescence unit successively
Light on and off, when each luminescence unit is lighted, Synchronization Control described image acquiring unit shoots the planar object, obtains each hair
Image information under light unit irradiation；Optical axis of the planar object perpendicular to image acquisition unit；
Luminous model and reflection model based on luminescence unit, are initial value with space coordinates according to a preliminary estimate and key light direction of principal axis,
According to the light source incidence light intensity and incident direction of each reconstruction point on described image information, planar object, each luminous list is found
The optimal spatial coordinate and key light direction of principal axis of unit.
12. three-dimensional rebuilding methods as claimed in claim 11, it is characterised in that luminous model and reflection based on luminescence unit
Model, with the exit direction in the space coordinates and primary optical axis as initial value, according on described image information, planar object each
The light source incidence light intensity and incident direction of reconstruction point, find the optimal spatial coordinate and key light direction of principal axis of each luminescence unit, bag
Include：
It is initial value with space coordinates according to a preliminary estimate and key light direction of principal axis based on light source composite model, iteration finds each and lights
The optimal spatial coordinate and key light direction of principal axis of unit, following operation is performed both by for each iteration cycle：
The space coordinates and key light direction of principal axis of image information, each luminescence unit according to planar object, calculate target object
The distance of the light source incidence direction of each reconstruction point and the reconstruction point to each luminescence unit；
Luminous model and reflection model based on luminescence unit, light source incidence direction and the reconstruction point according to each reconstruction point are arrived
The distance of each luminescence unit, the three-dimensional normal direction value of all reconstruction points of Calculation Plane body surface；
Three-dimensional normal direction value to all reconstruction points carries out mean square deviation calculating, until find all reconstruction points three-dimensional normal direction value it is equal
When variance is minimum, the space coordinates and key light direction of principal axis of corresponding each luminescence unit, as optimal spatial coordinate and primary optical axis
Direction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510927095.6A CN106875468B (en) | 2015-12-14 | 2015-12-14 | Three-dimensional reconstruction device and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510927095.6A CN106875468B (en) | 2015-12-14 | 2015-12-14 | Three-dimensional reconstruction device and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106875468A true CN106875468A (en) | 2017-06-20 |
| CN106875468B CN106875468B (en) | 2020-05-22 |
Family
ID=59178020
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510927095.6A Active CN106875468B (en) | 2015-12-14 | 2015-12-14 | Three-dimensional reconstruction device and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106875468B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109084700A (en) * | 2018-06-29 | 2018-12-25 | 上海摩软通讯技术有限公司 | The three dimensional local information acquisition methods and system of article |
| CN109147029A (en) * | 2018-06-25 | 2019-01-04 | 西安电子科技大学 | A kind of monocular polarization three-dimensional rebuilding method |
| CN109191560A (en) * | 2018-06-29 | 2019-01-11 | 西安电子科技大学 | Monocular based on scattered information correction polarizes three-dimensional rebuilding method |
| CN109377547A (en) * | 2018-09-20 | 2019-02-22 | 东南大学 | A kind of the photometric stereo three-dimensional reconstruction system and method for reconstructing of self-calibration light source direction |
| CN109410316A (en) * | 2018-09-21 | 2019-03-01 | 深圳前海达闼云端智能科技有限公司 | Method, tracking, relevant apparatus and the storage medium of the three-dimensional reconstruction of object |
| CN109472853A (en) * | 2018-11-16 | 2019-03-15 | 厦门大学 | A kind of lambert's body microcosmic surface reconstructing method based on image irradiation intensity |
| WO2019140778A1 (en) * | 2018-01-17 | 2019-07-25 | 深圳中科飞测科技有限公司 | Three-dimensional reconstruction system and three-dimensional reconstruction method |
| CN111951376A (en) * | 2020-07-28 | 2020-11-17 | 中国科学院深圳先进技术研究院 | Three-dimensional object reconstruction method fusing structural light and photometry and terminal equipment |
| CN112294453A (en) * | 2020-10-12 | 2021-02-02 | 浙江未来技术研究院(嘉兴) | Microsurgery surgical field three-dimensional reconstruction system and method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102074044A (en) * | 2011-01-27 | 2011-05-25 | 深圳泰山在线科技有限公司 | System and method for reconstructing surface of object |
| CN103729881A (en) * | 2013-12-28 | 2014-04-16 | 北京工业大学 | Tongue body dynamic three-dimensional reconstruction system oriented to tongue inspection in traditional Chinese medicine |
| CN103886642A (en) * | 2014-04-04 | 2014-06-25 | 北京科技大学 | Method for achieving three-dimensional reconstruction of steel plate surface |
-
2015
- 2015-12-14 CN CN201510927095.6A patent/CN106875468B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102074044A (en) * | 2011-01-27 | 2011-05-25 | 深圳泰山在线科技有限公司 | System and method for reconstructing surface of object |
| CN103729881A (en) * | 2013-12-28 | 2014-04-16 | 北京工业大学 | Tongue body dynamic three-dimensional reconstruction system oriented to tongue inspection in traditional Chinese medicine |
| CN103886642A (en) * | 2014-04-04 | 2014-06-25 | 北京科技大学 | Method for achieving three-dimensional reconstruction of steel plate surface |
Non-Patent Citations (1)
| Title |
|---|
| 谢利民 等: "基于不同光照条件的三维重建算法", 《华中科技大学学报（自然科学版）》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019140778A1 (en) * | 2018-01-17 | 2019-07-25 | 深圳中科飞测科技有限公司 | Three-dimensional reconstruction system and three-dimensional reconstruction method |
| US11448498B2 (en) | 2018-01-17 | 2022-09-20 | Skyverse Limited | Three-dimensional reconstruction system and three-dimensional reconstruction method |
| JP2021510836A (en) * | 2018-01-17 | 2021-04-30 | スカイヴァース・テクノロジー・カンパニー・リミテッド | 3D reconstruction system and 3D reconstruction method |
| JP7073532B2 (en) | 2018-01-17 | 2022-05-23 | スカイヴァース・テクノロジー・カンパニー・リミテッド | 3D reconstruction system and 3D reconstruction method |
| CN109147029B (en) * | 2018-06-25 | 2022-11-08 | 西安电子科技大学 | Monocular polarization three-dimensional reconstruction method |
| CN109147029A (en) * | 2018-06-25 | 2019-01-04 | 西安电子科技大学 | A kind of monocular polarization three-dimensional rebuilding method |
| CN109084700B (en) * | 2018-06-29 | 2020-06-05 | 上海摩软通讯技术有限公司 | Method and system for acquiring three-dimensional position information of article |
| CN109084700A (en) * | 2018-06-29 | 2018-12-25 | 上海摩软通讯技术有限公司 | The three dimensional local information acquisition methods and system of article |
| CN109191560A (en) * | 2018-06-29 | 2019-01-11 | 西安电子科技大学 | Monocular based on scattered information correction polarizes three-dimensional rebuilding method |
| CN109191560B (en) * | 2018-06-29 | 2021-06-15 | 西安电子科技大学 | Monocular polarization three-dimensional reconstruction method based on scattering information correction |
| CN109377547A (en) * | 2018-09-20 | 2019-02-22 | 东南大学 | A kind of the photometric stereo three-dimensional reconstruction system and method for reconstructing of self-calibration light source direction |
| CN109410316A (en) * | 2018-09-21 | 2019-03-01 | 深圳前海达闼云端智能科技有限公司 | Method, tracking, relevant apparatus and the storage medium of the three-dimensional reconstruction of object |
| CN109472853A (en) * | 2018-11-16 | 2019-03-15 | 厦门大学 | A kind of lambert's body microcosmic surface reconstructing method based on image irradiation intensity |
| CN111951376A (en) * | 2020-07-28 | 2020-11-17 | 中国科学院深圳先进技术研究院 | Three-dimensional object reconstruction method fusing structural light and photometry and terminal equipment |
| CN112294453B (en) * | 2020-10-12 | 2022-04-15 | 浙江未来技术研究院(嘉兴) | Microsurgery surgical field three-dimensional reconstruction system and method |
| CN112294453A (en) * | 2020-10-12 | 2021-02-02 | 浙江未来技术研究院(嘉兴) | Microsurgery surgical field three-dimensional reconstruction system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106875468B (en) | 2020-05-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106875468A (en) | Three-dimensional reconstruction apparatus and method | |
| CN109816703B (en) | Point cloud registration method based on camera calibration and ICP algorithm | |
| CN108510583B (en) | The generation method of facial image and the generating means of facial image | |
| CN105608734B (en) | A kind of image rebuilding method using three-dimensional image information acquisition device | |
| Roth et al. | Adaptive 3D face reconstruction from unconstrained photo collections | |
| CN111009007B (en) | Finger multi-feature comprehensive three-dimensional reconstruction method | |
| Wu et al. | Fusing multiview and photometric stereo for 3d reconstruction under uncalibrated illumination | |
| US9317973B2 (en) | Augmented reality method applied to the integration of a pair of spectacles into an image of a face | |
| CN106228507A (en) | A kind of depth image processing method based on light field | |
| CN106067190B (en) | A kind of generation of fast face threedimensional model and transform method based on single image | |
| CN104952063B (en) | For indicating the method and system of virtual objects in the view of true environment | |
| WO2021042208A1 (en) | Dynamically estimating light-source-specific parameters for digital images using a neural network | |
| CN102999942B (en) | Three-dimensional face reconstruction method | |
| Higo et al. | A hand-held photometric stereo camera for 3-d modeling | |
| Papadhimitri et al. | Uncalibrated near-light photometric stereo | |
| CN107506714A (en) | A kind of method of face image relighting | |
| CN109919876A (en) | A kind of true face model building of three-dimensional and three-dimensional true face photographic system | |
| CN109697688A (en) | A kind of method and apparatus for image procossing | |
| CN109285215A (en) | A kind of human 3d model method for reconstructing, device and storage medium | |
| CN108470370A (en) | The method that three-dimensional laser scanner external camera joint obtains three-dimensional colour point clouds | |
| CN109712172A (en) | A kind of pose measuring method of initial pose measurement combining target tracking | |
| CN108629294A (en) | Human body based on deformation pattern and face net template approximating method | |
| WO2006049147A1 (en) | 3d shape estimation system and image generation system | |
| Haefner et al. | Photometric depth super-resolution | |
| CN109903377A (en) | A kind of three-dimensional face modeling method and system without phase unwrapping |
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 |