CN107358185A - Palm print and palm vein image-recognizing method and device based on topological analysis - Google Patents
Palm print and palm vein image-recognizing method and device based on topological analysis Download PDFInfo
- Publication number
- CN107358185A CN107358185A CN201710534623.0A CN201710534623A CN107358185A CN 107358185 A CN107358185 A CN 107358185A CN 201710534623 A CN201710534623 A CN 201710534623A CN 107358185 A CN107358185 A CN 107358185A
- Authority
- CN
- China
- Prior art keywords
- palm
- veins
- arteries
- palm print
- 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
- 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
- G06V40/12—Fingerprints or palmprints
- G06V40/1365—Matching; Classification
- G06V40/1376—Matching features related to ridge properties or fingerprint texture
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Collating Specific Patterns (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention provides a kind of palm print and palm vein image-recognizing method based on topological analysis, comprise the following steps：S1：Obtain palm print and palm vein image；S2：Palm print and palm vein image is pre-processed to obtain palm print and palm vein distribution map；S3：The node of demarcation palm arteries and veins, node include bifurcation and end points in palm print and palm vein distribution map；S4：Establish each palm arteries and veins and the topological relation slapped between the node on arteries and veins；S5：According to topological relation identification palm print and palm vein information.Accordingly, present invention also offers a kind of palm print and palm vein pattern recognition device based on topological analysis.The topological structure of this method and device based on palm arteries and veins is characterized and stored to the feature for slapping arteries and veins, processing procedure is simple, the operand of Digital Image Processing is small, effectively reduce data processing amount and memory data output in palm arteries and veins identification process, identification process is not limited by shooting angle, shooting distance and picture sampler model, versatile, and device uses MCU, cost is low, low in energy consumption.
Description
Technical field
The present invention relates to palm print and palm vein image identification technical field, more particularly to a kind of palm print and palm vein based on topological analysis
Image-recognizing method and device.
Background technology
The hand vein recognition (identification of palm arteries and veins) of palm is a kind of living body biological identification technique of high safety.Technique is in pole
It is identified under the system of the accurate safety of degree.It is that the distribution of palm vein is obtained using near infrared ray inductor with principle
Figure, model is stored, and then establish everyone unique palm vein image database in a computer, then pass through computer
Log in and store everyone palm vein data, can also be managed in computer.Arteries and veins identification is slapped by being then based on live body
The high security of bio-identification, uniqueness, have broad application prospects in safety-security area.
In current palm print and palm vein identification technology, as Application No. 201010276401.7 application for a patent for invention in, its
Disclosed palm print and palm vein identifying system is by directly gathering palmprint image and palm arteries and veins image, and by the palm print and palm vein image of collection
Carry out aspect ratio pair after pretreatment with the sample that pre-saves, its content specifically compared palmmprint and does not slap the distribution of arteries and veins itself yet
Image information (distribution of such as palm print and palm vein and trend), its pretreatment carried out still only resides within existing image procossing
And compare technical.
In addition, during the identification of existing palm print and palm vein, due to based on the extraction to the graphic feature in image in itself, thus,
The image collected for camera, due to deformation (such as wide-angle lens collection image in image border caused by some cameras itself
Deformation, the anamorphose rates of different camera lenses is different), the versatility for causing image collecting device is poor, to system unit
Uniformity requires higher.Thus prior art is controlled by different application occasion and different collection lenses, different images
Identification equipment can not be compatible, and unique identification is poor, does not have effective identification.
Therefore, the existing position for needing to fix hand in identification process, when collected hand range image harvester away from
From it is different when, pretreated image result will be had a great influence.Particularly, after hand is offset, or plane angle where palm
During degree transformation, characteristics of image changes greatly.When now carrying out palm print and palm vein feature recognition, system obtains the palm arteries and veins of palm in processing
Distribution characteristics, to avoid identification error, influence journey of the palm inclination to identification process is often obtained by the way of statistics
For degree to carry out de-redundancy, the amount of calculation of progress image preprocessing is also very big required for this mode, heavier to system burden, whole
Individual system power dissipation is big, therefore it is required that system has stronger operational capability to perform the Digital Image Processing of complexity.Thus its structure
During into product, the requirement of operational capability is limited to, small product size is larger, is unfavorable for being widely popularized in safety-security area.
The content of the invention
It is an object of the invention to provide a kind of palm print and palm vein image-recognizing method and device based on topological analysis, with solution
Required operand present in certainly existing palm print and palm vein recognition methods is larger, fault-tolerance is poor, is unable to good conformity in difference
The problem of image recognition of the palm print and palm vein of angle.
The second object of the present invention is to provide a kind of palm print and palm vein image-recognizing method and device based on topological analysis,
To solve, the existing security device poor universality based on palm print and palm vein identification, power consumption are high, spacing requirement is high (to the position of palm
Put, angle requirement it is higher), the problem of cost is high, volume is larger.
To achieve the above object, the invention provides a kind of palm print and palm vein image-recognizing method based on topological analysis, bag
Include following steps：
S1：Obtain palm print and palm vein image；
S2：The palm print and palm vein image is pre-processed to obtain palm print and palm vein distribution map；
S3：The node of demarcation palm arteries and veins, the node include bifurcation and end points in the palm print and palm vein distribution map；
S4：Establish each palm arteries and veins and the topological relation slapped between the node on arteries and veins；
S5：Palm print and palm vein information is identified according to the topological relation.
It is preferred that the step S1 further comprises：The palm print and palm vein image is gathered using picture sampler.
It is preferred that the pretreatment in the step S2 includes：After wavelet de-noising processing being carried out to the palm print and palm vein image,
Binary conversion treatment is carried out again, obtains digitized palm print and palm vein distribution map.
It is preferred that the step S3 further comprises：
First, according to palm natural quality demarcation cog region in palm print and palm vein image；
Then the node in the cog region according to nodal community demarcation palm arteries and veins.
It is preferred that include according to palm natural quality demarcation cog region：Identification finger and palm are bordered by position, feeding refer to
The interface point of middle finger and nameless as a reference point with the interface point of little finger of toe, using the line it is side in palm after two reference point lines
On take square area as the cog region；Or feeding refers to and the interface point of middle finger, nameless interface point and thumb with little finger of toe
Refer to the interface point with forefinger as three reference points, take the inscribed circle of three reference points as the cog region.
It is preferred that the nodal community includes：End points, triradius, right-angled intersection point.
It is preferred that the step S4 includes：
Rectangular coordinate system is established in the cog region；
Extract the end points of the palm arteries and veins in the cog region；
According to two end points of the same palm arteries and veins Distance Judgment of the two coordinate directions palm arteries and veins is fastened in the rectangular co-ordinate
Trend；
Initial code is carried out to palm arteries and veins according to the trend of palm arteries and veins；
By the node on each bar palm arteries and veins and carry out secondary coding for node；
The initial code for slapping arteries and veins is combined with the secondary coding of corresponding node, obtains slapping arteries and veins topological structure coding schedule
To characterize the topological relation of palm arteries and veins.
It is preferred that the step S5 also includes：Preset matching weight is set for the different pieces of information in the topological relation, known
During other palm print and palm vein information, it is then match cognization success that matching weight, which reaches preset matching weight,.
Present invention also offers a kind of palm print and palm vein pattern recognition device based on topological analysis, including：
Image collection module, for obtaining palm print and palm vein image；
Image pre-processing module, for being pre-processed to obtain palm print and palm vein distribution map to the palm print and palm vein image；
Topological analysis module, for the node of the demarcation palm arteries and veins in the palm print and palm vein distribution map, and establish each palm arteries and veins
And the topological relation between the node on palm arteries and veins；
Identification module, for identifying palm print and palm vein information according to the topological relation.
It is preferred that also include：Memory module, for storing the topological relation for being authorized to the personnel to prestore and slapping arteries and veins, so that knowledge
The topological relation stored in topological relation and memory module that other module obtains topological analysis module carries out matching and is identified.
Present invention also offers a kind of palm print and palm vein pattern recognition device based on topological analysis, including：
Picture sampler, for gathering palm print and palm vein image；
Micro-control unit, for being pre-processed to obtain palm print and palm vein distribution map to the palm print and palm vein image, and in institute
The node of demarcation palm arteries and veins in palm print and palm vein distribution map is stated, and then establishes each palm arteries and veins and the topological relation slapped between the node on arteries and veins, and
Palm print and palm vein information is identified according to the topological relation.
The invention has the advantages that：
(1) topological structure based on palm arteries and veins is characterized and stored to the feature for slapping arteries and veins, and processing procedure is simple, digital picture
The operand of combing is small, so as to effectively reduce data processing amount and memory data output in palm arteries and veins identification process；
(2) palm arteries and veins based on topological analysis identifies that its identification process is related only to slapping the topological relation of arteries and veins, and identification process is not
Limited by palm arteries and veins image taking angle, shooting distance and picture sampler model, as long as extractor is opened up after photographing palm arteries and veins image
Flutterring the feature of relation can be identified, and versatility is stronger；
(3) relation on each point and side is only characterized due to slapping the topological relation of arteries and veins, data processing amount reduces, Jin Erjin
The processor requirement of row processing reduces, and using MCU, cost is low, low in energy consumption, is very suitable for extensive use.
Brief description of the drawings
Fig. 1 is the palm print and palm vein image-recognizing method flow chart based on topological analysis of preferred embodiment；
Fig. 2 is the palm print and palm vein image schematic diagram that a preferred embodiment obtains；
Fig. 3 is the cog region schematic diagram divided in Fig. 2 palm print and palm vein image；
Fig. 4 is topology diagram corresponding to the palm arteries and veins in Fig. 3 cog region；
Fig. 5 be Fig. 4 topology diagram carry out slap arteries and veins move towards judgement reference chart；
Fig. 6 is the composition figure of the palm print and palm vein pattern recognition device based on topological analysis of the preferred embodiment of the present invention.
Embodiment
Below with reference to the accompanying drawing of the present invention, clear, complete description is carried out to the technical scheme in the embodiment of the present invention
And discussion, it is clear that as described herein is only a part of example of the present invention, is not whole examples, based on the present invention
In embodiment, the every other implementation that those of ordinary skill in the art are obtained on the premise of creative work is not made
Example, belongs to protection scope of the present invention.
For the ease of the understanding to the embodiment of the present invention, make further by taking specific embodiment as an example below in conjunction with accompanying drawing
Illustrate, and each embodiment does not form the restriction to the embodiment of the present invention.
Present embodiments provide a kind of palm print and palm vein image-recognizing method based on topological analysis, shown in Figure 1, the party
Method comprises the following steps：
S1：Obtain palm print and palm vein image；
S2：The palm print and palm vein image is pre-processed to obtain palm print and palm vein distribution map；
S3：The node of demarcation palm arteries and veins, the node include bifurcation and end points in the palm print and palm vein distribution map；
S4：Establish each palm arteries and veins and the topological relation slapped between the node on arteries and veins；
S5：Palm print and palm vein information is identified according to the topological relation.
This method by grow naturally the palm arteries and veins topological relation (unique biological characteristic) carry out extraction and analysis and then
It is identified based on the topological relation, not only realizes effective identification of personnel identity, while the institute of the identification based on topological relation
The data volume of processing greatly reduces, while recognition efficiency effectively improves, it is not necessary to carries out the computing and analysis of complexity, effectively
Ground reduces the pressure of processor, and then can reduce the volume and cost of the processor for performing this method.
Below with the specific implementation procedure of the example in detail recognition methods.
Step S1 is first carried out to carry out obtaining palm print and palm vein image.Here palm print and palm vein figure is gathered by picture sampler
Picture, light filling can be carried out according to ambient brightness by LED during collection, and LED as desired by specific wavelength etc. is carried out
The imaging radiation of blood vessel, the infrared radiation blood vessel of infrared band is such as used, realize that blood vessel develops.Specifically adopted in the present embodiment
Above-mentioned palm arteries and veins image is gathered with ccd image sensor, while palmprint image is shot by camera.The ccd image senses
Device can launch infrared ray and form image by the weak reflection of the ferroheme of the red blood cell in vein blood vessel, so that camera is adopted
The vein image data of blood-collecting tube.Wherein, the sensor can be acquired according to the change of the wavelength of the infrared ray of transmission, be adopted
The vein blood vessel data of collection are the picture signals reflected positioned at skin of palm of hand layer 2-5mm vein blood vessel, obtain palm print and palm vein figure
Picture.
After obtaining palm print and palm vein image, step S2 is performed, palm print and palm vein image is pre-processed to obtain palm print and palm vein point
Butut.Here pretreatment specifically includes：After first carrying out wavelet de-noising processing to palm print and palm vein image, arteries and veins is clearly slapped
Image.Binary conversion treatment is carried out to the palm arteries and veins image after noise reduction process again, obtains digitized palm print and palm vein distribution map, this implementation
The palm arteries and veins figure of example is shown in Figure 2, now slaps arteries and veins image and is able to be indicated and store by data signal, is easy to follow-up number
Wordization processing and analysis.
Then, the node that step S3 demarcates palm arteries and veins in palm print and palm vein distribution map is performed, node here includes bifurcation
And end points.
Method when demarcating node specifically uses, and obtains each bar palm arteries and veins image, is obtained according to Trendline corresponding to palm arteries and veins trend
The Curvature varying of palm arteries and veins is taken, when Curvature varying is mutated (as become big suddenly or diminishing), corresponding to bending/bifurcated position of blood vessel
Put, then catastrophe point now is bifurcation.In addition, node also includes the end points of blood vessel line.
In further preferred embodiment, before node is demarcated, first, belong to naturally according to palm in palm print and palm vein image
Property demarcation cog region；Then the node in the cog region according to nodal community demarcation palm arteries and veins.
Wherein, include according to palm natural quality demarcation cog region：Identification finger is bordered by position with palm, takes food therein
The interface point B and the third finger of finger and middle finger are as a reference point with the interface point A of little finger of toe, using the line as side after two reference point lines
Square area is taken on palm, and as PIO characteristic areas (cog region), the cog region after extraction is shown in Figure 3.It is identified
For ease of subsequent treatment behind area, it is also necessary to carry out the simplification of topological structure using the basic theory of graph theory, a width is obtained after simplifying
It is the topology diagram on multiple sides comprising multiple nodes, it is shown in Figure 4.Here, when taking finger to be bordered by position with palm, ginseng
The position being bordered by according to each finger of Curvature varying acquirement of the curve of palm profile with palm, the biological property according to human hand are true
Determine cog region, so, the human hand of human hand or different angle for different characteristic, its characteristic area obtained is according to current
The cog region that the reference point of the palm pattern of sampling is made, this processing mode are not limited by the shooting angle of picture sampler
System, namely do not limited by the placed angle of human hand.For the image of the palm of the gathered person of different angle, it can pass through
Processing obtains specific reference point biologically, and then identifies specific PIO characteristic areas, is easy to subsequently to quiet in the area
Arteries and veins trend is identified and topological analysis.
Wherein, nodal community includes：End points, triradius, right-angled intersection point.Wherein, triradius and right-angled intersection point belong to
In bifurcation.Corresponding to as the angle value of node is 1,3,4 in topology diagram.Due to for blood vessel, while have one
The situation of 5 branches is less on individual point, thus the present embodiment does not consider the node that angle value is 5.But in other embodiment, if any need
Will, can also nodal community setting and identification be carried out to nodes more than 5 branches.
Step S4 is performed again, establishes each palm arteries and veins and the topological relation slapped between the node on arteries and veins.
Specifically, step S4 further comprises：
Rectangular coordinate system is established in cog region；
Extract the end points of the palm arteries and veins in cog region；
According to two end points of the same palm arteries and veins Distance Judgment of the two coordinate directions palm arteries and veins is fastened in the rectangular co-ordinate
Trend；
Initial code is carried out to palm arteries and veins according to the trend of palm arteries and veins；
By the node on each bar palm arteries and veins and carry out secondary coding for node；
The initial code for slapping arteries and veins is combined with the secondary coding of corresponding node, obtains slapping arteries and veins topological structure coding schedule
To characterize the topological relation of palm arteries and veins.
Corresponding diagram 5, the coordinate system of foundation forms x-axis for reference point A, B, and the direction on the vertical side of line is y-axis therewith,
Now, end points D, E on same palm arteries and veins are identified, then takes palm arteries and veins end points D, E accounting for relative to established coordinate system where it
With rectangular region a1, understood after comparing its length of side in the direction of the x axis and the length of side in the y-axis direction, on its x-axis direction
The length of side is more than the length of side in the y-axis direction, then may determine that to be horizontal, initial code is assigned for the palm arteries and veins for the trend of the palm arteries and veins
H1.In the case where there is the horizontal palm arteries and veins of Article 2, then continue as it and assign initial code H2, continue to assign H3 by that analogy,
H4 etc..Have been assigned horizontal initial code each palm arteries and veins can according to its in the cog region leftmost side or rightmost side end points relative to this
The fore-and-aft distance of A points (two summits of B points or lower section also can) in the point such as Fig. 5 of one jiao of cog region is sequenced successively, is assigned
Give initial code adjacent successively.For example, the end points that arteries and veins low order end is laterally slapped in the present embodiment is located at the imparting numbering of lower side
For H1, H2, H3, H4 etc. are followed successively by from left to right.
Similarly, for such as any one palm arteries and veins, extracting the end points of the palm arteries and veins first by said process, existing according to end points
The distance of different directions compares the size of its length of side in the direction of the x axis and the length of side in the y-axis direction in cog region.Work as x-axis
The length of side on direction is more than the length of side in the y-axis direction, then judges that for longitudinal direction, initial compile is assigned for the palm arteries and veins for the trend of the palm arteries and veins
Code Z1, if any longitudinal palm arteries and veins more than Article 2, then continue as it and assign initial code Z2, if any more than 2, then with such
Push away and continue to assign Z3, Z4 etc..Have been assigned each palm arteries and veins of longitudinal initial code, similarly can according to its in cog region most
Upside or lower side end points relative to the point such as A points (two summits of B points or lower section also can) of one jiao of the cog region laterally away from
From being sequenced successively, initial code adjacent successively is assigned, for example, the end points of longitudinal direction palm arteries and veins bottom is located at the tax of the leftmost side
It is Z1 to give numbering, is followed successively by Z2, Z3, Z4 etc. from left to right.
After carrying out initial code, continue to record all nodes on the palm arteries and veins, with continued reference to Fig. 5, to slapping arteries and veins H1, its left and right
Include 3 nodes between end points D, E, from right to left respectively triradius I, four crunode J (namely cross node), four crunode K, then
Respectively H1-A-C, H1-B-D, H1-C-D are encoded for three node distributions are secondary, wherein, H1 represents palm arteries and veins belonging to node, in
Between letter represent which node, A is first, and B is second, and last letter represents the attribute of the node, and C is trident
Point, D are four crunodes.The initial code for slapping arteries and veins H1 and the secondary coding of the node on the palm arteries and veins are combined again, obtain slapping arteries and veins
Topological structure encodes：H1=(3, AC, BD, CD), wherein, AC, BD, CD correspond to the secondary volume of three nodes of above-mentioned distribution respectively
Code.
The like, obtain H2, H3 ..., Hn coding, and Z1~Zm coding, n are the number of laterally palm arteries and veins, m is
The number of longitudinal direction palm arteries and veins.
N is laterally slapped the coding of arteries and veins and the coding composition palm arteries and veins topological structure coding schedule of m longitudinal direction palm arteries and veins, as identification
Topological relation data.The table directly characterizes the topological structure and its distribution trend of palm arteries and veins in an encoded form, so as to ten
Divide clear.After palm arteries and veins topological structure coding schedule is stored as identification reference data, identification every time only needs to be compiled
The comparison of code table, the amount of storage and processing speed of data can greatly improve.
Finally, step S5 is performed, palm print and palm vein information is identified according to above-mentioned topological relation.Including：For in topological relation
Different pieces of information set preset matching weight, identify palm print and palm vein information when, matching weight reach preset matching weight then for
With identifying successfully.Here weight refers to, it is contemplated that shoots shooting distance when slapping arteries and veins, shooting angle and video camera shooting level
Deviation, arteries and veins information is slapped in part can have small difference after topologizing, then when the data according to topological relation are matched, is
Avoid for the same hand slap coding schedule that it prestores it is with the data transformation of the coding schedule of currently processed generation consistent without
Identification, using weights, it is considered as successful match after the data rate of coincideing reaches matching weight (the alternatively referred to as threshold value).
Certainly, due to for topology diagram, the token state of other forms also be present, such as laterally palm rapid pulse amount, bus are slapped
Rapid pulse amount etc., it also may make up one group of coding, for the topological characteristic of the palm arteries and veins of identification different people.And above-mentioned coded system
Also it is not limited to characterize nodes encoding using letter, can directly use the numerical value of binary zero/1 progress data encoding, can also use
Other letter, words etc. are assigned in different nodes to be distinguished from each other.
In the embodiment of another deformation, above-mentioned initial code is also included for being slapped corresponding to node D, E for arteries and veins, will
It on the basis of the first order vessel that identification obtains, continues to mark the two level blood vessel being attached thereto, this implementation as first order vessel
Two level blood vessel example in the example blood vessel with reference to two nodes I, L in figure 5 corresponding to, it is one to be connected with first order vessel
The branch vessel of level blood vessel, therefore two level blood vessel is used as, H1-I L are encoded to for it, i.e., being added again in H1 basis of coding should
The name of two level blood vessel or coding.Branch vessel on two level blood vessel such as also be present, then can continue to encode two level blood as needed
Three-level blood vessel on pipe, three-level blood vessel here is the blood vessel for continuing shunting on two level blood vessel using two level blood vessel as trunk.According to
It is secondary to analogize, it can continue to encode the topological structures such as the level Four blood vessel on three-level blood vessel, the Pyatyi blood vessel on level Four blood vessel as needed
Vessel branch in the presence of figure.After being separately encoded to vessel branches at different levels, then the node on the blood vessel (palm arteries and veins) is marked successively,
The secondary coding based on the blood vessel is carried out to it, realizes the sign to the overall structure coding of the blood vessel.Certainly, here
The exemplary only coded systems of numbering H1-I L, those skilled in the art can use the volume of other forms herein as needed
Code, such as directly it is indicated with 0/1 yard can be distinguished from each other on different bits at one group, or rower is only entered with numeral
Know.
Then in a preferred embodiment accordingly, step S5 weight can also be defined as the power on different token state influences
Again, it is different that coding and its influence to final structure of the coding of two level blood vessel of first order vessel are for, is entered in step s 5
During row match cognization, the weight for being arranged to the coding of first order vessel is 90%, and the weight of the coding of two level blood vessel is 80%.This
It is meant that in final identification, the identical rate of matching to first order vessel requires higher, and the matching to the coding of two level blood vessel is kissed
Conjunction rate requires relatively low, and this mode can fully take into account branch vessel may be due to shooting angle in sampling relative to Major Vessels
Degree or shooting distance influence and partial loss, avoid the gap of initial pictures during due to this sampling from causing to last matching degree
Influence, and then avoid invalid identification.
In another preferred embodiment, the method according to palm natural quality demarcation cog region can also be：Feeding refer to
The interface point of middle finger, nameless and little finger of toe interface point and the interface point of thumb and forefinger take three ginsengs as three reference points
The inscribed circle of examination point is as cog region.Now, equally can be with forefinger and middle finger when the establishment of coordinate system in area is identified
As transverse axis reference direction, direction normal thereto is then the longitudinal axis for interface point, the line direction of the nameless interface point with little finger of toe
Reference direction.According to this direction setting, can equally progress be moved towards to the palm arteries and veins of all directions in the cog region of circle
Judge and encode, then the node on palm arteries and veins is encoded.Finally obtain the topological relation of palm arteries and veins in the circular cog region.
The exemplary only coding of above-mentioned coding form, those skilled in the art can be as needed using other forms
Coding structure, as long as it is analyzed and encoded based on topological structure, it is all contained in the scope of the present invention.And the application
Accompanying drawing be only a kind of exemplary schematic diagram corresponding with vascular distribution, actual topology diagram can moreover be only simply
A side-graph of a relation (relation between points only being identified, while being form of straight lines).In addition, above-mentioned various pieces are not being rushed
Can be bonded to each other in the case of prominent, improve after apply, it is belonged in above-mentioned example scope of disclosure.
Present invention also offers a kind of palm print and palm vein pattern recognition device based on topological analysis, the device is used to perform
The palm print and palm vein image-recognizing method stated.As shown in fig. 6, the device that the present embodiment provides specifically includes：
Image collection module 601, for obtaining palm print and palm vein image；
Image pre-processing module 602, for being pre-processed to obtain palm print and palm vein distribution map to the palm print and palm vein image；
Topological analysis module 603, for the node of the demarcation palm arteries and veins in the palm print and palm vein distribution map, and establish each palm
The topological relation between node on arteries and veins and palm arteries and veins；
Identification module 604, for identifying palm print and palm vein information according to the topological relation.
Here image collection module 601 includes picture sampler, or the combination including picture sampler and LED.Figure
As samplers sample palm print and palm vein image, light filling can be carried out according to ambient brightness by the LED of setting during collection, and according to
The imaging radiation by the progress blood vessel such as LED of specific wavelength is needed, the infrared radiation blood vessel of infrared band is such as used, realizes
Blood vessel develops.Wherein, picture sampler can be set to ccd image sensor as needed, with dedicated for collection palm vessel graph.
And image pre-processing module, topological analysis module and identification module are respectively three functional modules pre-seted on controller,
Corresponding function is performed by default single-chip microcomputer configuration processor respectively.Due in the application, identification palm print and palm vein information is
What the topological relation (data) based on palm arteries and veins was carried out, the required data volume for storing and handling is compared with traditional palm arteries and veins identification in amount
The reduction of level, thus, the controller can use MCU (micro-control unit), the overall miniaturization of realization device.
Wherein, the further preferred device also includes：Memory module, the module are used to store the personnel for being authorized to and prestoring
The topological relation of arteries and veins is slapped, so as to be stored in topological relation and memory module that identification module 604 obtains topological analysis module 603
Topological relation carry out matching and be identified.The memory module can be the internal memory that MCU is carried, or the storage device of peripheral hardware, this hair
Bright concrete form when being applied to does not limit.
In another preferred embodiment, present invention also offers a kind of palm print and palm vein image recognition dress based on topological analysis
Put, the device includes：Picture sampler, for gathering palm print and palm vein image；And
Micro-control unit, for being pre-processed to obtain palm print and palm vein distribution map to the palm print and palm vein image, and in institute
The node of demarcation palm arteries and veins in palm print and palm vein distribution map is stated, and then establishes each palm arteries and veins and the topological relation slapped between the node on arteries and veins, and
According to topological relation identification palm print and palm vein information.During concrete application, according to topological relation identification palm print and palm vein information after, by its with
The palm print and palm vein information to prestore is compared, to realize the identification to personnel identity.
The palm print and palm vein pattern recognition device based on topological analysis can be used in security device as needed, such as intelligence
Energy door lock, is identified by efficient palm print and palm vein, realizes that personnel identity identifies, and then carry out as needed according to personnel identity
Authorization unlocking or not authorization unlocking.
The foregoing is only a specific embodiment of the invention, but protection scope of the present invention is not limited thereto, any
Those skilled in the art the invention discloses technical scope in, to the present invention deformation or replacement done, should all cover
Within protection scope of the present invention.Therefore, protection scope of the present invention should be defined by described scope of the claims.
Claims (11)
1. a kind of palm print and palm vein image-recognizing method based on topological analysis, it is characterised in that comprise the following steps：
S1：Obtain palm print and palm vein image；
S2：The palm print and palm vein image is pre-processed to obtain palm print and palm vein distribution map；
S3：The node of demarcation palm arteries and veins, the node include bifurcation and end points in the palm print and palm vein distribution map；
S4：Establish each palm arteries and veins and the topological relation slapped between the node on arteries and veins；
S5：Palm print and palm vein information is identified according to the topological relation.
2. the palm print and palm vein image-recognizing method according to claim 1 based on topological analysis, it is characterised in that the step
Rapid S1 further comprises：The palm print and palm vein image is gathered using picture sampler.
3. the palm print and palm vein image-recognizing method according to claim 1 based on topological analysis, it is characterised in that the step
Pretreatment in rapid S2 includes：After carrying out wavelet de-noising processing to the palm print and palm vein image, then binary conversion treatment is carried out, obtained
Digitized palm print and palm vein distribution map.
4. the palm print and palm vein image-recognizing method according to claim 1 based on topological analysis, it is characterised in that the step
Rapid S3 further comprises：
First, according to palm natural quality demarcation cog region in palm print and palm vein image；
Then the node in the cog region according to nodal community demarcation palm arteries and veins.
5. the palm print and palm vein image-recognizing method according to claim 4 based on topological analysis, it is characterised in that according to hand
Palm natural quality demarcation cog region includes：Identification finger is bordered by position with palm, and feeding refers to the interface point and the third finger with middle finger
It is as a reference point with the interface point of little finger of toe, after two reference point lines on palm square area is taken as institute using the line as side
State cog region；Or feeding refers to the interface point conduct with the interface point of middle finger, nameless interface point and thumb and forefinger with little finger of toe
Three reference points, the inscribed circle of three reference points is taken as the cog region.
6. the palm print and palm vein image-recognizing method according to claim 4 based on topological analysis, it is characterised in that the section
Point attribute includes：End points, triradius, right-angled intersection point.
7. the palm print and palm vein image-recognizing method based on topological analysis according to claim 4 or 5, it is characterised in that institute
Stating step S4 includes：
Rectangular coordinate system is established in the cog region；
Extract the end points of the palm arteries and veins in the cog region；
The trend of the Distance Judgment of the two coordinate directions palm arteries and veins is fastened in the rectangular co-ordinate according to two end points of same palm arteries and veins；
Initial code is carried out to palm arteries and veins according to the trend of palm arteries and veins；
By the node on each bar palm arteries and veins and carry out secondary coding for node；
The initial code for slapping arteries and veins is combined with the secondary coding of corresponding node, obtains slapping arteries and veins topological structure coding schedule with table
The topological relation of sign palm arteries and veins.
8. the palm print and palm vein image-recognizing method according to claim 1 based on topological analysis, it is characterised in that the step
Rapid S5 also includes：For the different pieces of information in the topological relation, preset matching weight is set, when identifying palm print and palm vein information, matching
It is then match cognization success that weight, which reaches preset matching weight,.
A kind of 9. palm print and palm vein pattern recognition device based on topological analysis, it is characterised in that including：
Image collection module, for obtaining palm print and palm vein image；
Image pre-processing module, for being pre-processed to obtain palm print and palm vein distribution map to the palm print and palm vein image；
Topological analysis module, for the node of the demarcation palm arteries and veins in the palm print and palm vein distribution map, and establish each palm arteries and veins and the palm
The topological relation between node on arteries and veins；
Identification module, for identifying palm print and palm vein information according to the topological relation.
10. the palm print and palm vein pattern recognition device according to claim 9 based on topological analysis, it is characterised in that also wrap
Include：Memory module, for storing the topological relation for being authorized to the personnel that prestore and slapping arteries and veins, so that identification module is by topological analysis module
The topological relation stored in obtained topological relation and memory module carries out matching and is identified.
A kind of 11. palm print and palm vein pattern recognition device based on topological analysis, it is characterised in that including：
Picture sampler, for gathering palm print and palm vein image；
Micro-control unit, for being pre-processed to obtain palm print and palm vein distribution map to the palm print and palm vein image, and in the palm
The node of demarcation palm arteries and veins in line palm arteries and veins distribution map, and then each palm arteries and veins and the topological relation slapped between the node on arteries and veins are established, and foundation
The topological relation identifies palm print and palm vein information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710534623.0A CN107358185B (en) | 2017-07-03 | 2017-07-03 | Palm print and palm vein image-recognizing method and device based on topological analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710534623.0A CN107358185B (en) | 2017-07-03 | 2017-07-03 | Palm print and palm vein image-recognizing method and device based on topological analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107358185A true CN107358185A (en) | 2017-11-17 |
| CN107358185B CN107358185B (en) | 2019-10-25 |
Family
ID=60292813
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710534623.0A Active CN107358185B (en) | 2017-07-03 | 2017-07-03 | Palm print and palm vein image-recognizing method and device based on topological analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107358185B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108073916A (en) * | 2018-01-24 | 2018-05-25 | 四川政安通科技有限公司 | Palm print information collecting device |
| CN108960187A (en) * | 2018-07-20 | 2018-12-07 | 珠海格力电器股份有限公司 | Permission determines method and device |
| CN110598661A (en) * | 2019-09-19 | 2019-12-20 | 京麦智能科技(苏州)有限公司 | Palmprint and palmar vein acquisition system |
| CN112560710A (en) * | 2020-12-18 | 2021-03-26 | 北京曙光易通技术有限公司 | Method for constructing finger vein recognition system and finger vein recognition system |
| CN112668412A (en) * | 2020-12-15 | 2021-04-16 | 深兰科技（上海）有限公司 | Two-dimensional code generation method and device, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080123908A1 (en) * | 2003-06-21 | 2008-05-29 | Waldman David A | Acquisition of High Resolution Biometric Images |
| CN103559489A (en) * | 2013-11-19 | 2014-02-05 | 沈阳工业大学 | Method for extracting features of palm in non-contact imaging mode |
| CN105426821A (en) * | 2015-11-04 | 2016-03-23 | 浙江工业大学 | Palm vein feature extracting and matching method based on eight neighborhood and secondary matching |
| CN106022218A (en) * | 2016-05-06 | 2016-10-12 | 浙江工业大学 | Palm print palm vein image layer fusion method based on wavelet transformation and Gabor filter |
-
2017
- 2017-07-03 CN CN201710534623.0A patent/CN107358185B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080123908A1 (en) * | 2003-06-21 | 2008-05-29 | Waldman David A | Acquisition of High Resolution Biometric Images |
| CN103559489A (en) * | 2013-11-19 | 2014-02-05 | 沈阳工业大学 | Method for extracting features of palm in non-contact imaging mode |
| CN105426821A (en) * | 2015-11-04 | 2016-03-23 | 浙江工业大学 | Palm vein feature extracting and matching method based on eight neighborhood and secondary matching |
| CN106022218A (en) * | 2016-05-06 | 2016-10-12 | 浙江工业大学 | Palm print palm vein image layer fusion method based on wavelet transformation and Gabor filter |
Non-Patent Citations (2)
| Title |
|---|
| 张晨: ""基于复杂网络与多小波的掌纹识别方法"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 张环: ""掌纹掌脉及其融合识别技术研究"", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108073916A (en) * | 2018-01-24 | 2018-05-25 | 四川政安通科技有限公司 | Palm print information collecting device |
| CN108073916B (en) * | 2018-01-24 | 2021-12-17 | 山东汇佳软件科技股份有限公司 | Palm print information collecting device |
| CN108960187A (en) * | 2018-07-20 | 2018-12-07 | 珠海格力电器股份有限公司 | Permission determines method and device |
| CN110598661A (en) * | 2019-09-19 | 2019-12-20 | 京麦智能科技(苏州)有限公司 | Palmprint and palmar vein acquisition system |
| CN112668412A (en) * | 2020-12-15 | 2021-04-16 | 深兰科技（上海）有限公司 | Two-dimensional code generation method and device, electronic equipment and storage medium |
| CN112560710A (en) * | 2020-12-18 | 2021-03-26 | 北京曙光易通技术有限公司 | Method for constructing finger vein recognition system and finger vein recognition system |
| CN112560710B (en) * | 2020-12-18 | 2024-03-01 | 北京曙光易通技术有限公司 | Method for constructing finger vein recognition system and finger vein recognition system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107358185B (en) | 2019-10-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107358185B (en) | Palm print and palm vein image-recognizing method and device based on topological analysis | |
| CN104239769B (en) | Personal identification method and system based on finger vein features | |
| Zhang et al. | A robust iris segmentation scheme based on improved U-net | |
| CN108009520A (en) | A kind of finger vein identification method and system based on convolution variation self-encoding encoder neutral net | |
| CN102542281B (en) | Non-contact biometric feature identification method and system | |
| CN103870808B (en) | Finger vein identification method | |
| CN107066942A (en) | A kind of living body faces recognition methods and system | |
| CN107862282A (en) | A kind of finger vena identification and safety certifying method and its terminal and system | |
| CN107832739A (en) | A kind of palm vein identification system and method based on ORB feature extractions | |
| CN108921019A (en) | A kind of gait recognition method based on GEI and TripletLoss-DenseNet | |
| CN103793692A (en) | Low-resolution multi-spectral palm print and palm vein real-time identity recognition method and system | |
| CN106991380A (en) | A kind of preprocess method based on vena metacarpea image | |
| CN107944435A (en) | A kind of three-dimensional face identification method, device and processing terminal | |
| CN102163283A (en) | Method for extracting face characteristic based on local three-value mode | |
| Raghavendra et al. | A low cost wrist vein sensor for biometric authentication | |
| CN105138974B (en) | A kind of multi-modal Feature fusion of finger based on Gabor coding | |
| CN110390282A (en) | A kind of finger vein identification method and system based on the loss of cosine center | |
| CN111178130A (en) | Face recognition method, system and readable storage medium based on deep learning | |
| CN105760841A (en) | Identify recognition method and identify recognition system | |
| CN111914643A (en) | Human body action recognition method based on skeleton key point detection | |
| CN109766873A (en) | A kind of pedestrian mixing deformable convolution recognition methods again | |
| CN110555380A (en) | Finger vein identification method based on Center Loss function | |
| CN107798308A (en) | A kind of face identification method based on short-sighted frequency coaching method | |
| CN115830652B (en) | Deep palm print recognition device and method | |
| CN111985332A (en) | Gait recognition method for improving loss function based on deep learning |
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 |