CN108334873A - A kind of 3D four-dimension hand data discrimination apparatus - Google Patents
A kind of 3D four-dimension hand data discrimination apparatus Download PDFInfo
- Publication number
- CN108334873A CN108334873A CN201810300223.8A CN201810300223A CN108334873A CN 108334873 A CN108334873 A CN 108334873A CN 201810300223 A CN201810300223 A CN 201810300223A CN 108334873 A CN108334873 A CN 108334873A
- Authority
- CN
- China
- Prior art keywords
- characteristic
- hand
- data
- information
- point
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
Abstract
The present invention provides a kind of 3D four-dimension hand data discrimination apparatus, including：Hand-characteristic information collecting device, hand-characteristic 4 D data storage device and target person identity recognition device.Wherein hand-characteristic information collecting device includes：Image acquisition units, feature point extraction unit, point cloud generation unit and four-dimensional model construction unit.And it puts cloud generation unit and is made of characteristic point data collection, spatial depth information and feature point cloud data three parts.The acquisition of hand 4 D data can be completed using camera matrix, according to four dimension modules of several hand-characteristic picture construction hand-characteristics, to realize that human hands feature 4 D data acquires；Using identity of personage information as distinguishing mark, formation includes the database of a plurality of hand-characteristic 4 D data；The hand-characteristic 4 D data stored in database is found using the identity information of target person, and corresponding temmoku point cloud comparison method is to identify the identity of target person.
Description
Technical field
The present invention relates to technical field of biometric identification, especially a kind of 3D four-dimension hand data discrimination apparatus.
Background technology
Biological characteristic is the intrinsic physiology or behavioural characteristic of biology, such as fingerprint and palmmprint.Biological characteristic have it is certain only
One property and stability, i.e., the diversity ratio between certain biological characteristic of any two biology is larger, and biological characteristic generally will not be with
The time changes a lot, this allows for biological characteristic and is well suited for applying the certification in authentication or identifying system to believe
In the scenes such as breath.
Current biological attribute data is all the 2D data using plane, main to use by taking the biological characteristic of hand as an example
The mode of 2D is imitative according to the collected 2D pictures of hand to identify the feature of some or several hands, part criminal
2D hand-characteristics processed, part identifying system of out-tricking bring prodigious security risk to personal information security.Therefore, there is an urgent need for needles
Multidimensional data identification is carried out to hand-characteristic, improves safety, and support is provided for subsequent application.
Invention content
In view of the above problems, it is proposed that the present invention overcoming the above problem in order to provide one kind or solves at least partly
State the 3D four-dimension hand data discrimination apparatus of problem.
A kind of 3D four-dimension hand data discrimination apparatus, including following device：
Hand-characteristic information collecting device, for acquiring several hand-characteristic images of human body within given time, and root
According to four dimension modules of several hand-characteristic picture construction hand-characteristics, to realize four dimensions of the hand-characteristic of the human body
According to acquisition；
Hand-characteristic 4 D data storage device, for scanning or the identity information of typing human body (I1, I2 ... In) conduct
Distinguishing mark is associated storage to collected human hands feature 4 D data, and formation includes four dimension of a plurality of hand-characteristic
According to the database of (D1, D2 ... Dn)；
Target person identity recognition device, for identity information (I1, the I2 ... according to scanning or the target person of typing
In the hand-characteristic 4 D data (D1, D2 ... Dn) stored in the database) is found, and by the hand of the target person
Feature 4 D data (T1, T2 ... Tn) respectively with stored in the corresponding database hand-characteristic 4 D data (D1,
D2 ... Dn) it is compared, to identify the identity of target person.
Further, the hand-characteristic information collecting device includes：
Image acquisition units, the camera matrix for being formed using more cameras are acquired hand characteristic information, obtain
To several hand-characteristic images；
It is special to extract several described hands for handling several described hand-characteristic images for feature point extraction unit
Levy respective characteristic point in image；
Point cloud generation unit, for based on respective characteristic point in several hand-characteristic images described in extraction, generating hand
The feature point cloud data of portion's feature；
Four-dimensional model construction unit, four dimension modules for building hand-characteristic according to the feature point cloud data, with reality
The acquisition of existing hand-characteristic 4 D data.
Further, the feature point extraction unit includes image processor GPU and central processor CPU, and characteristic point carries
It takes unit that the image information of several hand-characteristic images is assigned in the block block of GPU and carries out operation, and combine CPU
Centralized dispatching and distribution function, calculate several described respective characteristic points of hand-characteristic image.
Further, in the image acquisition units, Visible Light Camera, laser camera, infrared camera, grating phase are used
Machine or light-field camera composition camera matrix are acquired hand characteristic information.
Further, the point cloud generation unit includes following information：
Characteristic point data collection carries out characteristic point according to respective characteristic point in several hand-characteristic images described in extraction
Matching, establish matched characteristic point data collection；
Spatial depth information calculates each camera relative to hand-characteristic spatially according to the optical information of camera
Relative position, and the space depth within given time of the characteristic point in the hand-characteristic image is calculated according to the relative position
Spend information；
Feature point cloud data is believed according to the spatial depth of matched characteristic point data collection and characteristic point within given time
Breath, generates the feature point cloud data of hand-characteristic.
Further, the feature of respective characteristic point is converted using scale invariant feature in several described hand-characteristic images
SIFT feature describes son to describe.
According to the optical information of camera, each camera is calculated relative to hand-characteristic spatially using light-stream adjustment
Relative position；The spatial depth information of characteristic point in several described hand-characteristic images includes：Spatial positional information and color
Information.
Further, the four-dimensional model construction unit, four dimension modules of hand-characteristic are built according to the point cloud data
The step of further comprise：
Set the reference dimension of four dimension modules to be built；
According to the spatial positional information of the reference dimension and the feature point cloud data, the feature point cloud data is determined
In each characteristic point bulk and time size, to build four dimension modules of hand-characteristic；The four of the hand-characteristic
Dimension module includes at least one following 4 D data：
Spatial form characteristic of four dimension modules on different time is described；
Surface texture feature data of four dimension modules on different time are described；
Facing material and light characteristic of four dimension modules on different time are described.
Further, it forms camera matrix using more cameras to be acquired hand characteristic information, in the following manner
It is laid out camera matrix：
Support construction is built, arc bearing structure is set in the support construction；
More cameras are arranged in the arc bearing structure；
Further, the identity information includes：It is one or more in name, gender, age and certificate number.Card
Piece number may include people in life in commonly used such as identification card number, passport No., license number, social security number or officer's identity card number
It is one or more.
Preferably, the identity information is obtained by scanning identity card, passport, driving license, social security card or officer's identity card, alternatively,
Manually or automatically the mode of typing obtains identity information from identity card, passport, driving license, social security card or officer's identity card.
Further, display is set in the arc bearing structure；
After structure obtains four dimension module of hand, four dimension modules are shown by visual means over the display；
Before the camera matrix formed using more cameras is acquired hand information, by display interfaces, if
The parameter of taking pictures of fixed each camera.
Further, the target person identity recognition device, when carrying out identification to target person, using temmoku point
Cloud matching identification method is to the product feature that is stored in the target product feature 4 D data (T1, T2 ... Tn) and the database
4 D data (D1, D2 ... Dn) is compared；The temmoku point cloud matching identification method includes the following steps：
S0301. characteristic point is fitted；
S0302. curved surface entirety best fit；
S0303. similarity calculation.
Further, the temmoku point cloud matching identification method comprises the following specific steps that：
Characteristic point fitting is carried out using based on spatial domain directly matched method, in the corresponding rigid region of two clouds,
It chooses three and features above point is used as fitting key point, pass through coordinate transform, directly carry out characteristic point Corresponding matching；
After characteristic point Corresponding matching, the alignment of data of the point cloud after whole curved surface best fit；
Similarity calculation is carried out using least square method.
The present invention provides a kind of 3D four-dimension hand data discrimination apparatus, which is specifically to utilize more camera compositions
Camera matrix is acquired hand characteristic information, obtains several hand-characteristic images on different moments；And then to several
Hand-characteristic image is handled, and respective characteristic point in several hand-characteristic images is extracted；Subsequently, based on several hands of extraction
Respective characteristic point in portion's characteristic image generates the feature point cloud data of hand-characteristic；Later, it is built according to feature point cloud data
Four dimension modules of hand-characteristic, to realize the acquisition of hand-characteristic 4 D data.It can be seen that the embodiment of the present invention uses more
Camera control technology carries out the acquisition of hand-characteristic information, can significantly improve the collecting efficiency of hand-characteristic information；Also, this
Inventive embodiments completely restore hand-characteristic spatially each using collecting the characteristic information of hand-characteristic spatially
Item feature, the application for subsequent hand-characteristic data provide unlimited possibility.To identify the identity information identification of target
4 D data, it is not necessary to the mass data in the data and database of target person be compared one by one, improve matching identification
Efficiency greatly improves the speed of identification, special using being carried out based on the directly matched temmoku point cloud matching identification method in spatial domain
Sign point fitting, the Fast Fitting for realizing hand-characteristic point compares, and then realizes the identification of identity rapid authentication.
Description of the drawings
By reading the detailed description of hereafter preferred embodiment, various other advantages and benefit are common for this field
Technical staff will become clear.Attached drawing only for the purpose of illustrating preferred embodiments, and is not considered as to the present invention
Limitation.And throughout the drawings, the same reference numbers will be used to refer to the same parts.In the accompanying drawings：
Fig. 1 shows 3D four-dimension hand data discrimination apparatus structural schematic diagram according to an embodiment of the invention；
Fig. 2 shows the structural schematic diagrams of 3D 4 D datas harvester according to an embodiment of the invention；
Fig. 3 shows the structural schematic diagram of 4 D data harvester according to another embodiment of the present invention；
Fig. 4, which is shown, to be shown according to one embodiment of the invention based on the hand-characteristic 4 D data identifying system that visible light is taken pictures
It is intended to；
Fig. 5 shows the hand-characteristic 4 D data identifying system signal based on laser scanning according to one embodiment of the invention
Figure；
Fig. 6, which is shown, to be known according to one embodiment of the invention based on the hand-characteristic 4 D data that depth infrared camera is taken pictures
Other system schematic；
Fig. 7 shows the hand-characteristic 4 D data identifying system signal based on raster scanning according to one embodiment of the invention
Figure；And
Fig. 8 shows the hand-characteristic 4 D data identifying system taken pictures based on light-field camera according to one embodiment of the invention
Schematic diagram.
Specific implementation mode
The exemplary embodiment of the disclosure is more fully described below with reference to accompanying drawings, although showing the disclosure in attached drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
It is limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure
Completely it is communicated to those skilled in the art.
It should be noted that the 3D 4 D datas in the present invention refer to three-dimensional space data binding time dimension data institute shape
At data, three dimensions binding time dimension refers to：Multiple same time intervals or different time intervals, different angle, no
With the data acquisition system of image or image formation situations such as orientation or different conditions.
In order to solve the above-mentioned technical problem, an embodiment of the present invention provides a kind of 3D four-dimension hand data discrimination apparatus.Fig. 1
Show 3D four-dimension hand data discrimination apparatus structural schematic diagram according to the ... of the embodiment of the present invention, in Fig. 1,3D four-dimension hand numbers
It can specifically include according to harvester 100：Hand-characteristic information collecting device 110, hand-characteristic 4 D data storage device
120, target person identity recognition device 130；
Hand-characteristic information collecting device 110, for acquiring several hand-characteristic images of human body within given time, and
According to four dimension modules of several hand-characteristic picture construction hand-characteristics, to realize the four-dimension of the hand-characteristic of the human body
Data acquire；
Hand-characteristic 4 D data storage device 120, for scanning or the identity information of typing human body (I1, I2 ... In)
Storage is associated to collected human hands feature 4 D data as distinguishing mark, formation includes a plurality of hand-characteristic four
The database of dimension data (D1, D2 ... Dn)；
Target person identity recognition device 130, for according to scanning or typing target person identity information (I1,
I2 ... In) find the hand-characteristic 4 D data (D1, D2 ... Dn) stored in the database, and by the target person
Hand-characteristic 4 D data (T1, T2 ... Tn) respectively with the hand-characteristic 4 D data that is stored in the corresponding database
(D1, D2 ... Dn) is compared, to identify the identity of target person.
Fig. 2 shows the structural schematic diagrams of 3D 4 D datas harvester according to an embodiment of the invention.Such as Fig. 2 institutes
Show, which may include image acquisition units 210, feature point extraction unit 220, put cloud generation unit 230 and four-dimensional mould
Type construction unit 240.
Image acquisition units 210, the camera matrix for being formed using more cameras are acquired hand characteristic information,
Obtain several hand-characteristic images；
Feature point extraction unit 220 extracts several described hands for handling several described hand-characteristic images
Respective characteristic point in characteristic image；
Point cloud generation unit 230, for based on respective characteristic point in several hand-characteristic images described in extraction, generating
The feature point cloud data of hand-characteristic；
Four-dimensional model construction unit 240, four dimension modules for building hand-characteristic according to the feature point cloud data, with
Realize the acquisition of hand-characteristic 4 D data.
The present invention alternative embodiment in, in above-mentioned image acquisition units 210, using Visible Light Camera, laser camera,
Infrared camera, grating camera or light-field camera composition camera matrix are acquired hand characteristic information.
Preferably, feature point extraction unit 220 includes image processor GPU and central processor CPU, feature point extraction list
The image information of several hand-characteristic images is assigned in the block block of GPU and carries out operation by member, and combines the collection of CPU
Middle scheduling and distribution function calculate several described respective characteristic points of hand-characteristic image and can see, and the embodiment of the present invention is adopted
The acquisition that hand-characteristic information is carried out with more photographing camera control technologies can significantly improve the acquisition effect of hand-characteristic information
Rate.Also, parallel computation of the embodiment of the present invention based on central processing unit and graphics processor can efficiently realize that feature is believed
The processing of breath.
Preferably, GPU is double GPU, and every GPU has a multiple block, such as 56 block, the embodiment of the present invention to this not
It is restricted.
In the alternative embodiment of the present invention, above-mentioned cloud generation unit 230 includes following information：
Characteristic point data collection carries out characteristic point according to respective characteristic point in several hand-characteristic images described in extraction
Matching, establish matched characteristic point data collection；
Spatial depth information calculates each camera relative to hand-characteristic spatially according to the optical information of camera
Relative position, and the space depth within given time of the characteristic point in the hand-characteristic image is calculated according to the relative position
Spend information；
Feature point cloud data is believed according to the spatial depth of matched characteristic point data collection and characteristic point within given time
Breath, generates the feature point cloud data of hand-characteristic.
In the alternative embodiment of the present invention, the feature of respective characteristic point uses scale not in several hand-characteristic images
Become Feature Conversion SIFT feature and describes son to describe.
In the alternative embodiment of the present invention, above-mentioned cloud generation unit 230 is additionally operable to：
According to the optical information of camera, each camera is calculated relative to hand-characteristic spatially using light-stream adjustment
Relative position；The spatial depth information of characteristic point in several described hand-characteristic images includes：Spatial positional information and color
Information.
In the alternative embodiment of the present invention, above-mentioned four-dimension model construction unit 240 is additionally operable to：
Set the reference dimension of four dimension modules to be built；
According to the spatial positional information of the reference dimension and the feature point cloud data, the feature point cloud data is determined
In each characteristic point bulk and time size, to build four dimension modules of hand-characteristic；
In the alternative embodiment of the present invention, four dimension modules of hand-characteristic include at least one following four dimensions
According to：
Spatial form characteristic of four dimension modules on different time is described；
Surface texture feature data of four dimension modules on different time are described；
Facing material and light characteristic of four dimension modules on different time are described.
In the alternative embodiment of the present invention, as shown in figure 3, the device of figure 2 above displaying can also include：
Camera matrix layout unit 310, is coupled with image acquisition units 210, for being utilized in image acquisition units 210
Before the camera matrix of more cameras composition is acquired hand characteristic information, it is laid out more cameras in the following manner：
Support construction is built, arc bearing structure is set in the support construction；
More cameras are arranged in the arc bearing structure.
It can be seen that the embodiment of the present invention carries out the acquisition of hand-characteristic information using camera matrix majorization technology, it can be with
Significantly improve the collecting efficiency of hand-characteristic information.Further, when the different types of camera of use is arranged in arc carrying knot
Different camera matrixes can be formed on structure, will be described in detail respectively below.
Situation one, as shown in figure 4, the camera being arranged in arc bearing structure above can support visible light to take pictures
Camera matrix.The 3D four-dimension hand data discrimination apparatus includes：
Cabinet：The main connection structure and surface structure of whole equipment；
Visible Light Camera matrix：It is responsible for human hands information collection；
Light compensating lamp：Camera shoots the supplement of external lights；
Identification module：The processing and identification of human hands information；
Control and display module：The control of equipment and the realization of display function；
Hand fixed module：It is responsible for the fixation of human body hand position.
Situation two, as shown in figure 5, the camera being arranged in arc bearing structure above can be the phase for supporting laser scanning
Machine matrix.The 3D four-dimension hand data discrimination apparatus includes：
Cabinet：The main connection structure and surface structure of whole equipment；
Laser scanning module：It is responsible for human hands information collection；
Sliding rail：In cabinet, for making laser scanning module be moved to specified collection point；
Identification module：The processing and identification of human hands information；
Control and display module：The control of equipment and the realization of display function；
Hand fixed module：It is responsible for the fixation of human body hand position.
Situation three, as shown in fig. 6, the camera being arranged in arc bearing structure above can support depth infrared camera
The camera matrix of scanning.The 3D four-dimension hand data discrimination apparatus includes：
Cabinet：The main connection structure and surface structure of whole equipment；
Infrared camera module：It is responsible for human hands depth distance information collection；
Visible Light Camera：According to the size of finger, camera lens automatic telescopic to suitable visual angle and clarity；
Runing rest：Support rotary mechanism；
Rotating mechanism：To control and mould is shown with fixed speed rotation and simultaneous transmission camera picture and Range finder data
Block；
Identification module：The processing and identification of human hands information；
Control and display module：The control of equipment and the realization of display function；
Hand fixed module：It is responsible for the fixation of human body hand position.
Situation four, as shown in fig. 7, the camera being arranged in arc bearing structure above can be the phase for supporting raster scanning
Machine matrix.The 3D four-dimension hand data discrimination apparatus includes：
Cabinet：The main connection structure and surface structure of whole equipment；
Grating camera model：It is responsible for human hands information collection；
Sliding rail：In cabinet, for making grating camera model be moved to specified collection point；
Identification module：The processing and identification of human hands information；
Control and display module：The control of equipment and the realization of display function；
Hand fixed module：It is responsible for the fixation of human body hand position.
Situation five, as shown in figure 8, the camera being arranged in arc bearing structure above can be the camera square of light field scanning
Battle array.The 3D four-dimension hand data discrimination apparatus includes：
Cabinet：The main connection structure and surface structure of whole equipment；
Light-field camera module：It is responsible for human hands information collection；
Astral lamp light module：Positioned at light-field camera both sides inside cabinet；
Identification module：The processing and identification of human hands information；
Control and display module：The control of equipment and the realization of display function；
Hand fixed module：It is responsible for the fixation of human body hand position.
In an alternate embodiment of the invention, can also display be set in arc bearing structure：Human hands are obtained in structure
Four dimension modules after, hand 4 D data is shown by visual mode over the display.
In an alternative embodiment, can also be by display interfaces before information collection, each camera of setting is taken pictures
Parameter, such as sensitivity, shutter speed, zoom magnification, aperture, the embodiment of the present invention are without being limited thereto.
Preferably, in hand-characteristic 4 D data storage device 120,110 institute of storage hand-characteristic information collecting device
Collected hand 4 D data, and it is used as distinguishing mark to collected hand with identity of personage information (I1, I2 ... In)
Feature 4 D data is stored, and formation includes the database of a plurality of hand-characteristic 4 D data library (D1, D2 ... Dn), for example,
4 D data D1 is with identity of personage information I1 associated storages, and the 4 D data D2 of another personage is with the identity information I2 of the personage
It is associated storage, and so on, form the database for the 4 D data for including n sample.
Preferably, the identity information includes：It is one or more in name, gender, age and certificate number.
Preferably, when target person identity recognition device 130 identifies piece identity, using temmoku point cloud matching identification method
To stored in the target product feature 4 D data (T1, T2 ... Tn) and the database product feature 4 D data (D1,
D2 ... Dn) it is compared；The temmoku point cloud matching identification method includes the following steps：
S0301. characteristic point is fitted；
S0302. curved surface entirety best fit；
S0303. similarity calculation.
Preferably, temmoku point cloud matching identification method comprises the following specific steps that：
Characteristic point fitting is carried out using based on spatial domain directly matched method, in the corresponding rigid region of two clouds,
It chooses three and features above point is used as fitting key point, pass through coordinate transform, directly carry out characteristic point Corresponding matching；
After characteristic point Corresponding matching, the alignment of data of the point cloud after whole curved surface best fit；
Similarity calculation is carried out using least square method.
Temmoku point cloud matching identification method (Yare Eyes point cloud match recognition method) is known
Other process and operation principle are as follows：First, point cloud at a time is the basic element for forming four dimension modules, it includes space
Coordinate information (XYZ) and colouring information (RGB).The attribute of point cloud includes spatial resolution, positional accuracy, surface normal etc..
Its feature is not influenced by external condition, will not all be changed for translating and rotating.Reverse software can carry out a cloud
Editor and processing, such as：Imageware, geomagic, catia, copycad and rapidform etc..Temmoku point cloud, which compares, to be known
Other method is distinctive to include based on the directly matched method in spatial domain：Iteration closest approach method ICP (Iterative closest
Point), ICP methods are generally divided into two steps, the fitting of first step characteristic point, second step curved surface entirety best fit.First fitting alignment
The purpose of characteristic point is in order to which the shortest time is found and is aligned two clouds of fitting to be compared.But not limited to this.Such as it can be with
It is：
The first step chooses three and features above point is used as fitting key point in the corresponding rigid region of two clouds,
By coordinate transform, characteristic point Corresponding matching is directly carried out.
ICP is a very effective tool in 3D data reconstruction process, at certain for curve or the registration of curved surface segment
One moment gave the rough initial alignment condition of two 3D models, and ICP iteratively seeks rigid transformation between the two with minimum
Change alignment error, realizes the registration of the space geometry relationship of the two.
Given setWithSet element two model surfaces of expression
Coordinate points, ICP registration techniques iteratively solve apart from nearest corresponding points, establish transformation matrix, and implement transformation to one of,
Until reaching some condition of convergence, its coding of iteration stopping is as follows：
1.1ICP algorithm
Input .P1, P2.
P after output is transformed2
P2(0)=P2, l=0；
Do
For P2(l) each point in
A nearest point y is looked in P1i；
End For
It calculatesRegistration error E；
IfE is more than a certain threshold value
Calculate P2(l) the transformation matrix T (l) between Y (l)；
P2(l+1)=T (l) P2(l), l=l+1；
Else
Stop；
End If
While||P2(l+l)-P2(l)||>threshold；
Wherein registration error
1.2 matchings based on local feature region：
By taking the identification of hand information as an example, human hands finger joint is rigid region, and metacarpus is plastic region, and plastic deformation influences
The accuracy of alignment, and then influence similarity.Second of gathered data has local difference, a kind of solution to plasticity model for the first time
Approach is only in rigid region selected characteristic point, and characteristic point is extracted from an object, keeps stablizing under certain condition
Constant attribute is fitted alignment using iteration closest approach method ICP characteristic points.
Requirement to characteristic point：
1) completeness contains object information as much as possible, is allowed to be different from the object of other classifications；2) compactedness tables
It is as few as possible up to required data volume；3) feature is also required preferably to be remained unchanged under model rotation, translation, mirror transformation.
In 3D living things feature recognitions, using two 3D biological characteristic model point clouds of alignment, the similar of input model is calculated
Degree, wherein registration error is as difference measure.
Second step：After characteristic point best fit, the alignment of data of the point cloud after whole curved surface best fit.
Third walks, similarity calculation.Least square method
Least square method (also known as least squares method) is a kind of mathematical optimization techniques.It by minimize error quadratic sum
Find the optimal function matching of data.Unknown data can be easily acquired using least square method, and these are acquired
Data and real data between error quadratic sum be minimum.Least square method can also be used for curve matching.It is some other excellent
Change problem can also be expressed by minimizing energy or maximizing entropy with least square method.It is usually used in solving curve fit problem,
And then solve the complete fitting of curved surface.It can accelerate Data Convergence by iterative algorithm, quickly acquire optimal solution.
If 3D data models at a time are inputted with stl file format, pass through calculating point cloud and triangle
The distance of piece determines its deviation.Therefore, this method needs to establish plane equation to each tri patch, and deviation arrives flat for point
The distance in face.And be IGES or STEP models for 3D data models at a time, since free form surface expression-form is
The faces NURBS, so the distance calculating in point to face needs the method for using numerical optimization to be calculated.By in iterative calculation point cloud
Each point expresses deviation to the minimum range of nurbs surface, or that nurbs surface carried out specified scale is discrete, with point and point
Apart from approximate expression point deviation, or it is converted into STL formats and carries out deviation calculating.Different coordinate alignment and deviation calculating side
The testing result of method, acquisition is also different.The size of alignment error will directly affect the confidence level of accuracy of detection and assessment report.
Best fit alignment is that detection error averagely arrives entirety, is terminated in terms of iteration by ensureing the whole minimum condition of deviation
The alignment procedure of calculation carries out 3D analyses to registration result, generates result object in the form of the root mean square of error between two figures
Output, root mean square is bigger, and difference of two models of reflection at this is bigger.Vice versa.Judge according to registration ratio is compared
Whether it is to compare subject matter.
Equipment application method is as follows
A. starting device：After turning on the power switch, central processing unit, camera matrix, band-like light compensating lamp is respectively started.
B. parameter setting：By display interfaces, the parameters that camera matrix is taken pictures can be set.
C. information collection：After parameter setting, starts matrix camera and start to carry out information collection, information to human hands
Acquisition time is completed in 0.8 second, the signal of acquisition finally with the format of digital picture (.jpg) reach central processing module into
Row processing, central processing module core are made of following components：
C.1 CPU (Central Processing Unit, central processing unit)：It is responsible for the transmission of entire digital signal
Scheduling, task distribution, the single calculation processing of memory management and part；
C.2 GPU (Graphics Processing Unit, image processing unit)：The GPU for selecting special type, has
Outstanding image-capable and efficient computing capability；
C.3 DRAM (Dynamic Random Access Memory, i.e. dynamic random access memory)：As entire
The temporary storage center of Digital Signal Processing needs the operational capability for matching CPU and GPU, obtains best processing and calculates effect
Energy.
D. information processing：The signal that matrix camera has acquired is transmitted to central processing module and carries out signal processing.
D.1 the process of information processing is as follows
D.1.1 the filtering of image is acquired
Using the characteristic of GPU, in conjunction with the characteristic of matrix operation of image filtering, image filtering can be in certain algorithm
Under support, it is rapidly completed.
D.1.2 the feature point extraction of image is acquired
The GPU to match using CPU and with overall performance, because the format of the various information of this equipment is all image pane
The various information contents of jpg can be evenly distributed to GPU's by formula in conjunction with the GPU with outstanding image-capable
In block, since this equipment has 56 block using double GPU, every GPU itself, so the jpg that acquisition information scratching arrives
Image can be evenly distributed to carry out operation above 112 block, and combine centralized dispatching and the distribution function of CPU, can be with
The characteristic point that every photo has rapidly is calculated, arranges in pairs or groups the GPU's of other common models relative to independent CPU or CPU
Operation, whole arithmetic speed time are the 1/10 or shorter of the latter.
D.1.3 the calculating of the matching and spatial depth information of acquisition image
The extraction of image characteristic point uses pyramidal hierarchical structure and the particular algorithm of space scale invariance, this
Two kinds of special algorithms are all the special tectonics of the GPU in conjunction with this equipment choosing, play the calculated performance of system to the greatest extent,
Realize the characteristic point in rapid extraction image information.
The Feature Descriptor of this process has 128 feature descriptions using SIFT feature description, SIFT feature description
Vector can describe the feature of 128 aspects of any characteristic point on direction and scale, significantly improve the essence to feature description
Degree, while Feature Descriptor has independence spatially.
The particular image that this equipment uses handles GPU, the calculating with excellent independent vector and processing capacity, for adopting
For SIFT feature vector with 128 special description, it is to be most suitable for only to be handled under conditions of special GPU in this way
, the specific calculations ability of the GPU can be given full play to, compares and is arranged in pairs or groups other common specifications using common CP U or CPU
The match time of GPU, characteristic point can reduce by 70%.
Feature Points Matching finishes, and system can use the algorithm of light-stream adjustment calculate camera relative to hand spatially
Relative position, according to the space coordinate of this relative position, GPU can rapidly calculate the depth information of hand-characteristic point.
D.1.4 the generation of feature point cloud data
According to D.1.3 calculating depth information of the hand-characteristic point in space, due to the vectorial computing capability that GPU has,
Spatial position and the colouring information of hand-characteristic point cloud can be rapidly matched, the model foundation needs of a standard are formed
Point cloud information.
E. characteristic size is demarcated：By the standard of characteristic point cloud size, initial reference is set for the size of entire model
Size.
By the special calibration in information collection, which there is space to determine size, due to hand-characteristic point
Cloud has spatially consistency of scale, is sized really by the special calibration, the size between any characteristic point of hand can
To be calculated from the spatial position coordinate of cloud.
F. the subsequent processing of data：It can by the way that point cloud data is further processed based on the size demarcated in E
To obtain the 3D data of hand at a time.
The format of 3D data has following several files：
.obj --- the spatial form feature of description 3D models
.jpg --- the surface texture feature of description 3D models
.mtl --- the Facing material and light feature of description 3D models
G. hand 4 D data is shown over the display by visualization method.
Identification module in central processing module finds the data according to the identity information (I1, I2 ... In) of target person
The biological characteristic 4 D data (D1, D2 ... Dn) stored in library, and by the hand-characteristic 4 D data of the target person (T1,
T2 ... Tn) it is compared respectively with the hand-characteristic 4 D data (D1, D2 ... Dn) stored in the corresponding database, to know
The identity of other target person, and over the display by recognition result output display.
An embodiment of the present invention provides one kind being based on 3D four-dimension hand data discrimination apparatus, is specifically using more in a device
The camera matrix of the different types of camera composition of platform is acquired hand characteristic information, obtains several hands in given time
Characteristic image；And then several hand-characteristic images are handled, extract respective characteristic point in several hand-characteristic images；With
Afterwards, respective characteristic point in several hand-characteristic images based on extraction generates hand-characteristic point cloud data；Later, according to spy
Four dimension modules for levying point cloud data structure hand-characteristic, to realize the acquisition of hand-characteristic 4 D data.It can be seen that the present invention
Embodiment carries out the acquisition of hand-characteristic information using different camera matrix majorization technologies, can significantly improve hand-characteristic letter
The collecting efficiency of breath；Also, the embodiment of the present invention is completely restored using the characteristic information of hand-characteristic spatially is collected
The various features of human hands spatially, the application for subsequent hand-characteristic data provide unlimited possibility.
Further, parallel computation of the embodiment of the present invention based on central processing unit and graphics processor, can be rapidly and efficiently
It realizes the processing of characteristic information and puts the generation of cloud in ground.Also, using scale invariant feature conversion SIFT feature description son knot
The computation capability for closing special graph processor, can fast implement the generation of the matching and space characteristics point cloud of characteristic point.
In addition, using unique sizing calibration method, can quickly and accurately extract any characteristic point of hand-characteristic bulk and
Time size generates four dimension modules of hand-characteristic, to realize the acquisition of 4 D data.
In the instructions provided here, numerous specific details are set forth.It is to be appreciated, however, that the implementation of the present invention
Example can be put into practice without these specific details.In some instances, well known method, structure is not been shown in detail
And technology, so as not to obscure the understanding of this description.
Similarly, it should be understood that in order to simplify the disclosure and help to understand one or more of each inventive aspect,
Above in the description of exemplary embodiment of the present invention, each feature of the invention is grouped together into single implementation sometimes
In example, figure or descriptions thereof.However, the method for the disclosure should be construed to reflect following intention：It is i.e. required to protect
Shield the present invention claims the more features of feature than being expressly recited in each claim.More precisely, as following
Claims reflect as, inventive aspect is all features less than single embodiment disclosed above.Therefore,
Thus the claims for following specific implementation mode are expressly incorporated in the specific implementation mode, wherein each claim itself
All as a separate embodiment of the present invention.
Those skilled in the art, which are appreciated that, to carry out adaptively the module in the equipment in embodiment
Change and they are arranged in the one or more equipment different from the embodiment.It can be the module or list in embodiment
Member or component be combined into a module or unit or component, and can be divided into addition multiple submodule or subelement or
Sub-component.Other than such feature and/or at least some of process or unit exclude each other, it may be used any
Combination is disclosed to all features disclosed in this specification (including adjoint claim, abstract and attached drawing) and so to appoint
Where all processes or unit of method or equipment are combined.Unless expressly stated otherwise, this specification (including adjoint power
Profit requires, abstract and attached drawing) disclosed in each feature can be by providing the alternative features of identical, equivalent or similar purpose come generation
It replaces.
In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments
In included certain features rather than other feature, but the combination of the feature of different embodiments means in of the invention
Within the scope of and form different embodiments.For example, in detail in the claims, embodiment claimed it is one of arbitrary
It mode can use in any combination.
The all parts embodiment of the present invention can be with hardware realization, or to run on one or more processors
Software module realize, or realized with combination thereof.It will be understood by those of skill in the art that can use in practice
Microprocessor or digital signal processor (DSP) are according to the ... of the embodiment of the present invention biological special based on Visible Light Camera to realize
Levy some or all functions of some or all components in 4 D data harvester.The present invention is also implemented as using
In executing some or all equipment or program of device of method as described herein (for example, computer program and meter
Calculation machine program product).It is such to realize that the program of the present invention may be stored on the computer-readable medium, or can have one
The form of a or multiple signals.Such signal can be downloaded from internet website and be obtained, or above be carried in carrier signal
For, or provide in any other forms.
It should be noted that the present invention will be described rather than limits the invention for above-described embodiment, and ability
Field technique personnel can design alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference mark between bracket should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" before element does not exclude the presence of multiple such
Element.The present invention can be by means of including the hardware of several different elements and being come by means of properly programmed computer real
It is existing.In the unit claims listing several devices, several in these devices can be by the same hardware branch
To embody.The use of word first, second, and third does not indicate that any sequence.These words can be explained and be run after fame
Claim.
So far, although those skilled in the art will appreciate that present invention has been shown and described in detail herein multiple shows
Example property embodiment still without departing from the spirit and scope of the present invention, still can according to the present disclosure directly
Determine or derive many other variations or modifications consistent with the principles of the invention.Therefore, the scope of the present invention is understood that and recognizes
It is set to and covers other all these variations or modifications.
Claims (10)
1. a kind of 3D four-dimension hand data discrimination apparatus, which is characterized in that including following device：
Hand-characteristic information collecting device, for acquiring several hand-characteristic images of human body within given time, and according to institute
Four dimension modules for stating several hand-characteristic picture construction hand-characteristics, to realize that the 4 D data of the hand-characteristic of the human body is adopted
Collection；
Hand-characteristic 4 D data storage device, is used to scan or the identity information of typing human body (I1, I2 ... In) is as identification
Mark is associated storage to collected human hands feature 4 D data, and formation includes a plurality of hand-characteristic 4 D data
The database of (D1, D2 ... Dn)；
Target person identity recognition device, for being looked for according to the identity information (I1, I2 ... In) of scanning or the target person of typing
To the hand-characteristic 4 D data (D1, D2 ... Dn) stored in the database, and by the hand-characteristic of the target person four
Dimension data (T1, T2 ... Tn) respectively with the hand-characteristic 4 D data (D1, D2 ... Dn) that is stored in the corresponding database
It is compared, to identify the identity of target person.
2. equipment according to claim 1, which is characterized in that the hand-characteristic information collecting device includes：
Image acquisition units, the camera matrix for being formed using more cameras are acquired hand characteristic information, obtain more
Width hand-characteristic image；
Feature point extraction unit extracts several described hand-characteristic figures for handling several described hand-characteristic images
The respective characteristic point as in；
Point cloud generation unit, for based on respective characteristic point in several hand-characteristic images described in extraction, it is special to generate hand
The feature point cloud data of sign；
Four-dimensional model construction unit, four dimension modules for building hand-characteristic according to the feature point cloud data, to realize hand
The acquisition of portion's feature 4 D data.
3. equipment according to claim 2, which is characterized in that the feature point extraction unit includes image processor GPU
And the image information of several hand-characteristic images is assigned to the block of GPU by central processor CPU, feature point extraction unit
Operation is carried out in block, and combines centralized dispatching and the distribution function of CPU, and it is respective to calculate several described hand-characteristic images
Characteristic point.
4. equipment according to claim 2, which is characterized in that in the image acquisition units, using Visible Light Camera,
Laser camera, infrared camera, grating camera or light-field camera composition camera matrix are acquired hand characteristic information.
5. equipment according to claim 2, which is characterized in that the point cloud generation unit includes following information：
Characteristic point data collection carries out of characteristic point according to respective characteristic point in several hand-characteristic images described in extraction
Match, establishes matched characteristic point data collection；
Spatial depth information calculates each camera relative to hand-characteristic spatially opposite according to the optical information of camera
Position, and the spatial depth within given time of the characteristic point in the hand-characteristic image is calculated according to the relative position and is believed
Breath；
Feature point cloud data, it is raw according to the spatial depth information of matched characteristic point data collection and characteristic point within given time
At the feature point cloud data of hand-characteristic.
6. equipment according to claim 2, which is characterized in that respective characteristic point in several described hand-characteristic images
Feature describes son to describe using scale invariant feature conversion SIFT feature；According to the optical information of camera, using bundle adjustment
Method calculates relative position of each camera relative to hand-characteristic spatially；Characteristic point in several described hand-characteristic images
Spatial depth information include：Spatial positional information and colouring information.
7. equipment according to claim 2, which is characterized in that the four-dimension model construction unit, according to described cloud number
Further comprise according to the step of four dimension module for building hand-characteristic：
Set the reference dimension of four dimension modules to be built；
According to the spatial positional information of the reference dimension and the feature point cloud data, determine each in the feature point cloud data
The bulk and time size of a characteristic point, to build four dimension modules of hand-characteristic；The four-dimensional mould of the hand-characteristic
Type includes at least one following 4 D data：
Spatial form characteristic of four dimension modules on different time is described；
Surface texture feature data of four dimension modules on different time are described；
Facing material and light characteristic of four dimension modules on different time are described.
8. equipment according to claim 2, which is characterized in that form camera matrix using more cameras and believe hand-characteristic
Breath is acquired, and is laid out camera matrix in the following manner：
Support construction is built, arc bearing structure is set in the support construction；
More cameras are arranged in the arc bearing structure；
The identity information includes：It is one or more in name, gender, age and certificate number.
9. equipment according to claim 8, which is characterized in that
Display is set in the arc bearing structure；
After structure obtains four dimension module of hand, four dimension modules are shown by visual means over the display；
Before the camera matrix formed using more cameras is acquired hand information, by display interfaces, setting is each
The parameter of taking pictures of platform camera.
10. equipment according to claim 1, which is characterized in that the target person identity recognition device, to target person
When carrying out identification, using temmoku point cloud matching identification method to the target product feature 4 D data (T1, T2 ... Tn) and
The product feature 4 D data (D1, D2 ... Dn) stored in the database is compared；The temmoku point cloud matching identification method
Include the following steps：
S0301. characteristic point is fitted；
S0302. curved surface entirety best fit；
S0303. similarity calculation；
The temmoku point cloud matching identification method comprises the following specific steps that：
Characteristic point fitting is carried out using based on the directly matched method in spatial domain, in the corresponding rigid region of two clouds, is chosen
Three and features above point conduct fitting key point, pass through coordinate transform, directly carry out characteristic point Corresponding matching；
After characteristic point Corresponding matching, the alignment of data of the point cloud after whole curved surface best fit；
Similarity calculation is carried out using least square method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810300223.8A CN108334873A (en) | 2018-04-04 | 2018-04-04 | A kind of 3D four-dimension hand data discrimination apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810300223.8A CN108334873A (en) | 2018-04-04 | 2018-04-04 | A kind of 3D four-dimension hand data discrimination apparatus |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108334873A true CN108334873A (en) | 2018-07-27 |
Family
ID=62933346
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810300223.8A Withdrawn CN108334873A (en) | 2018-04-04 | 2018-04-04 | A kind of 3D four-dimension hand data discrimination apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108334873A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110543871A (en) * | 2018-09-05 | 2019-12-06 | 天目爱视（北京）科技有限公司 | point cloud-based 3D comparison measurement method |
| WO2021129539A1 (en) * | 2019-12-24 | 2021-07-01 | 深圳市万普拉斯科技有限公司 | Physical appearance recognition method, apparatus, terminal, and computer readable storage medium |
-
2018
- 2018-04-04 CN CN201810300223.8A patent/CN108334873A/en not_active Withdrawn
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110543871A (en) * | 2018-09-05 | 2019-12-06 | 天目爱视（北京）科技有限公司 | point cloud-based 3D comparison measurement method |
| CN110543871B (en) * | 2018-09-05 | 2022-01-04 | 天目爱视（北京）科技有限公司 | Point cloud-based 3D comparison measurement method |
| WO2021129539A1 (en) * | 2019-12-24 | 2021-07-01 | 深圳市万普拉斯科技有限公司 | Physical appearance recognition method, apparatus, terminal, and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108549873A (en) | Three-dimensional face identification method and three-dimensional face recognition system | |
| CN108416312B (en) | A kind of biological characteristic 3D data identification method taken pictures based on visible light | |
| CN111316293A (en) | Method for object recognition | |
| US20140172377A1 (en) | Method to reconstruct a surface from oriented 3-d points | |
| EP3382645B1 (en) | Method for generation of a 3d model based on structure from motion and photometric stereo of 2d sparse images | |
| SE528068C2 (en) | Three dimensional object recognizing method for e.g. aircraft, involves detecting image features in obtained two dimensional representation, and comparing recovered three dimensional shape with reference representation of object | |
| CN109766876A (en) | Contactless fingerprint acquisition device and method | |
| CN108520230A (en) | A kind of 3D four-dimension hand images data identification method and equipment | |
| CN108446596A (en) | Iris 3D 4 D datas acquisition system based on Visible Light Camera matrix and method | |
| CN108319939A (en) | A kind of 3D four-dimension head face data discrimination apparatus | |
| CN108446597B (en) | A kind of biological characteristic 3D collecting method and device based on Visible Light Camera | |
| CN110807828A (en) | Oblique photography three-dimensional reconstruction matching method | |
| CN107506358A (en) | Message processing device and information processing method | |
| KR20230152149A (en) | Systems and methods for dynamic identification of surgical trays and items contained thereon | |
| CN108334874A (en) | A kind of 3D four-dimension iris image identification equipment | |
| Jiang et al. | Efficient match pair selection for oblique UAV images based on adaptive vocabulary tree | |
| CN108550184A (en) | A kind of biological characteristic 3D 4 D datas recognition methods and system based on light-field camera | |
| CN108334873A (en) | A kind of 3D four-dimension hand data discrimination apparatus | |
| EP1810216B1 (en) | 3d object recognition | |
| CN108259751A (en) | A kind of polyphaser data acquisition control system | |
| Galantucci et al. | Coded targets and hybrid grids for photogrammetric 3D digitisation of human faces | |
| Bianconi et al. | Robust color texture features based on ranklets and discrete Fourier transform | |
| CN108470150A (en) | A kind of biological characteristic 4 D data acquisition method and device based on Visible Light Camera | |
| CN108537236A (en) | A kind of polyphaser data control system for identifying | |
| CN110147773A (en) | A kind of remote sensing images recognition methods |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180727 |
|
| WW01 | Invention patent application withdrawn after publication |