CN1346116A - Method for identifying human body biological characteristics - Google Patents
Method for identifying human body biological characteristics Download PDFInfo
- Publication number
- CN1346116A CN1346116A CN00124954.1A CN00124954A CN1346116A CN 1346116 A CN1346116 A CN 1346116A CN 00124954 A CN00124954 A CN 00124954A CN 1346116 A CN1346116 A CN 1346116A
- Authority
- CN
- China
- Prior art keywords
- image
- human body
- conversion
- border
- regularization
- 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
- 238000000034 method Methods 0.000 title claims abstract description 71
- 238000006243 chemical reaction Methods 0.000 claims abstract description 44
- 238000004458 analytical method Methods 0.000 claims abstract description 14
- 238000000926 separation method Methods 0.000 claims abstract description 7
- 230000008569 process Effects 0.000 claims abstract description 3
- 230000008859 change Effects 0.000 claims description 8
- 230000009471 action Effects 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 238000013459 approach Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 230000009021 linear effect Effects 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 4
- 238000009499 grossing Methods 0.000 claims description 3
- 230000004888 barrier function Effects 0.000 claims description 2
- 230000009466 transformation Effects 0.000 claims description 2
- 230000007246 mechanism Effects 0.000 description 7
- 230000035479 physiological effects, processes and functions Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 230000001815 facial effect Effects 0.000 description 2
- 238000013316 zoning Methods 0.000 description 2
- 229930183419 Irisone Natural products 0.000 description 1
- UZFLPKAIBPNNCA-BQYQJAHWSA-N alpha-ionone Chemical compound CC(=O)\C=C\C1C(C)=CCCC1(C)C UZFLPKAIBPNNCA-BQYQJAHWSA-N 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 229940079593 drug Drugs 0.000 description 1
- 230000008570 general process Effects 0.000 description 1
- 238000003706 image smoothing Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000013011 mating Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 210000001747 pupil Anatomy 0.000 description 1
- 210000001525 retina Anatomy 0.000 description 1
- 210000003786 sclera Anatomy 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Collating Specific Patterns (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The identification method of biological characteristics of human body includes the following steps: 1. image separation: detecting edge, contour line or zone of analytical object from biolgoical tissue image of human body; 2. zone standardization conversion; establishing slandard grid of meshes on the separated image zone; 3. making regularization conversion of texture in the standard mesh; 4. equal-length maximum entropy coding; coding regularized characteristics of texture on every small zone; 5. analysis of characterictic code: analysizing two groups of codes obtained by using the above-mentioned steps to process two different images of same human body biological tissue, if said two groups of codes are similar, these two image are same, otherwise they are different. Said method is applicable to identification of biological tissue of human body by using image, and its error identification rate is lower than that of existent method.
Description
The present invention relates to method for identifying human body biological characteristics.
Human body biological characteristics is meant the Human Physiology tissue that can be used for differentiating personal identification, for example fingerprint, retina, iris, face, face type, DNA etc.Compare with non-human body biometric identity authentication such as password, password, seal, human body biological characteristics is exclusive, constant, the inherent authentication foundation that does not abandon of individual, and the human body biological characteristics authentication is natural rational identity identifying method.In the human body biological characteristics recognition technology, the existing long developing history of digital finger-print recognition system in recent years, has some other digital recognition system to occur successively again.Take a broad view of existing human body biological characteristics recognition technology, can see that following characteristics are arranged:
1. independence
Characteristics of existing human body biological characteristics recognition technology are that different physiological tissues are had different recognition methods.The technology of developing according to these methods also just constitutes different autonomous systems respectively naturally with product, image capture device, independently algorithm, independently software and chip are separately separately arranged separately independently, to such an extent as to independently application and environment for use separately.
2. employing method for mode matching
Another characteristics of prior art are to adopt method for mode matching mostly, these methods or be the numerical approach of foundational development with artificial comparison method, or serve as the method that the basis forms with pattern texture image treatment technology.Because method therefor has empirical or can not analyze image on micro-scale, thereby the accuracy of these class methods and poor reliability.In addition, the code that produces with these class methods is generally not equal-length code, is unsuitable for mating, and has both increased the identification difficulty, has reduced accuracy of identification again.
At the problem that prior art exists, the present invention will provide a kind of new recognition methods, and this method is organized as identifying object with specific Human Physiology, use collecting device to obtain image, come authenticating identity by the special disposal to image.This method is applicable to the various identifications that can collect the biological tissue of human body of usable image by image capture device, it provides general recognition methods and technology to different human body biological tissue, and on micro-scale, image is analyzed, make misclassification rate far below existing method and technology.
Task of the present invention is achieved through the following technical solutions:
A kind of method for identifying human body biological characteristics that uses image capture device and computing machine to carry out the steps include:
(1) separation of images
Gather the biological tissue of human body image with image capture device, from giving biological tissue of human body image, detect edge, outline line or the zone of particular analysis object, so that separation, extraction and analysis object with computing machine;
(2) regional standard conversion
Is to set up the standard grid on the analyzed area with computing machine at isolated image-region, carries out the regional standard conversion;
(3) textural characteristics regularization conversion
With computing machine texture in the normal grid is carried out the regularization conversion;
(4) isometric, maximum entropy coding
With computing machine the regularization sign of texture on each zonule is weaved into code, as the digital representation of whole analyzed area epigraph texture;
(5) condition code analysis
Two groups of codes after with computing machine two different images of same biological tissue of human body being handled with the method in above-mentioned (1)-(4) are analyzed, to determine that this two image is whether from same people's same biological tissue of human body, promptly carry out authentication, two groups of codes after with computing machine two different images of same biological tissue of human body being handled with the method in above-mentioned (1)-(4) are analyzed, to determine that this two image is whether from same people's same biological tissue of human body, promptly carry out authentication, the correspondence code position that value is different in two groups of codes in whole sign indicating number positions shared number percent smaller or equal to the setting person, then this two image is same, otherwise is different.
This method has following characteristics:
1. unitarity: different human body biological tissue is provided general recognition methods and technology, make in this way and can develop comprehensive human body biological characteristics identification product with technology;
2. reliability: on micro-scale, image is analyzed, made misclassification rate far below existing method and technology;
3. practicality: the image capture device of different human body biological tissue can be integrated into single equipment, computer software is integrated into one chip, different human body biological tissue recognition system is integrated into single product, under unified environment, use.
Describe in detail below in conjunction with embodiment.
Fig. 1 is the iris image of human body;
Fig. 2 is the fingerprint image of human body;
Fig. 3 is the face-image of human body.
The present invention is a kind of method for identifying human body biological characteristics that uses image capture device and computing machine to carry out, and it is applicable to the identification that can obtain the human body biological characteristics of its image by certain means.Because discern specific biological tissue of human body and image thereof, at first will from background, separate the image of this part biological tissue.Then, texture feature extraction from the image of separating, again to these features analyze and same file store (database) in archives material compare differentiation, at last the main body of holding this physiological tissue is made the authentication conclusion.The present invention has proposed new implementation method to each subprocess in the above-mentioned general process, thereby becomes a kind of brand-new recognition technology.These new methods and new technology are as shown in the table:
| Purpose | Content | Method |
| Separation of images | Edge/zone is detected | Self-defined border/area detector |
| Signature analysis | The spatial domain is analyzed | Regional standard conversion mechanism |
| Time-domain analysis | Textural characteristics regularization conversion mechanism table method mechanism | |
| Feature identification | The feature statement | Isometric, maximum entropy scrambler |
| Feature identification | The condition code analyzer |
Specifying of the inventive method step is as follows:
(1) self-defined border/area detector
A. the problem that will solve
From giving image, detect edge, outline line or the zone of particular analysis object, with separation, extraction and analysis object.
B. method
Obtain the image of specific biological tissue of human body by image capture device.For sake of convenience, the two dimensional gray image of this biological tissue is with f (x) expression, the position of pixel in the x presentation video wherein, and f (x) is the gray scale of this point.We are called self-defined border/area detector to the border in the identifying/regional automatic detecting machine system, and this mechanism can realize by software or chip.Now the principle of work with self-defined border/area detector is described below:
The structure in a. self-defined border that will detect or zone:, just can define linear barrier's detecting device if the border of detecting is to be made of straight-line segment.In general, detecting device is not necessarily linear, and normally image f (x) is certain line integral (in order to detection boundaries) or the area branch (in order to surveyed area) of parameter with λ, and note is made F that (x, λ), then (x λ) is the function of f (x) to F.Wherein, λ is and the border that will detect or the regional relevant geometric parameter of shape, as angle, radius, radius-of-curvature etc.
B. detecting device is carried out smoothly: be the level and smooth notion of explanation detecting device, our explanation earlier is level and smooth to picture point.
Level and smooth to picture point: should not treat each point (current point) in the image, the point around should calculating is to its influence isolatedly, that is to say to carry out level and smooth.Need consider that the point of its influence is called the application point of current point to current point, the zone that application point constituted is called action scope.If only consider linear effect, the line segment that action scope can be is mid point with current point, otherwise, can be the center with current point rectangle or other shape.Each application point can be different to the size that influences of current point, need by smooth function given.If consider that some influence nearby is big, some influence at a distance is little, and (x, σ) as smooth function, wherein σ is a scale factor generally to choose Gaussian function G.
Detecting device level and smooth: to detecting device F (x λ) determines its action scope, and to F (x, λ) each point on (check point, promptly current point) with G (x, σ) on action scope, carry out level and smooth.
C. λ is differentiated or calculate the change amount: detecting device is parameter with λ.The physical significance that with λ is parameter is: have the many groups of pixels that can constitute the defined boundary shape of detecting device (one group of point be called may border) in image, but in general have only wherein one group to be real border.There is different λ values on different possible borders.In order from the possibility border, to search out real border, (for example, total gray-scale value on possible border) variation that is to say the general characteristic on each possibility border when needing to calculate the λ variation, to be that variable may be differentiated on the border to each with λ, or calculate the change amount.
D. calculate the extreme value of derivative or change amount: from image property, all there is such character on real border: its change amount might reach maximum in the border in institute.Therefore, can determine real border by the extreme value of calculating derivative or change amount.
(2) regional standard conversion mechanism
A. the problem that will solve
Even to same a kind of physiological tissue of same individual, under different time, varying environment, all may obtain the image of different sizes, other character of image also might be different.In addition, the original-shape of image may bring many inconvenience to analyzing and processing.For the different images of handling same physiological tissue and the shape that analyzed area is had be suitable for handling, must carry out the standardization conversion to analyzed area.We still use the image of the specific physiological tissue of f (x) expression, and conversion requires to be summarized as follows (may only require a part that satisfies wherein):
A. another zone is arrived in the image transformation of a certain analyzed area on the plane;
B. the border with a certain analyzed area in the image is another kind of shape from a kind of shape conversion;
C. the textural characteristics to analytic target inside carries out standardization;
D. image is transformed to another coordinate space from a coordinate space.
B. method
Generally speaking, and do not require whole conversion of carrying out in the above-mentioned requirements.To the d item in the above-mentioned requirements, after the coordinate system before and after conversion is given, can realize by common coordinate transform, will not discuss here.Now above-mentioned a, b, three of c are provided following general approach:
A. set up normal grid in institute on to analyzed area according to the characteristics of practical problems; Set up graticule mesh and will follow two principles: the one, to setting up with a kind of graticule mesh with a kind of physiological tissue, the 2nd, divide grid in the graticule mesh according to the geometric configuration in zone.If the geometric configuration of analyzed area has been suitable for analyzing and processing, normal grid can no longer be carried out other conversion after setting up.
B. if the geometric configuration of analyzed area is carried out conversion, then the line segment in the graticule mesh must be transformed to the target area one by one.In the case, the former boundary contour (being made as C1) and the another kind of outline line (being made as C2) that will convert to are all known.Conversion is carried out in the following manner: just establishing, the intersection point of the line segment a in conversion and boundary contour C1 is x1, x2, a (x) is the point on the line segment a, then after the conversion value of image f (x) in the target area by two known function h1 (x, x1), (x x2) determines h2, h1 (x, x1), (x x2) is called the standardization function to h2.
(3) textural characteristics regularization conversion mechanism
A. the problem that will solve
The different texture structure of two classes is arranged.One class is the pattern texture, has texture primitive in this texture structure, and texture primitive or roughly make even repeated arrangement in a wider context by the partial structurtes that primitive is formed; Another kind of is the texture of non-structure rule, and this class texture has randomness, atypism (not having tangible texture primitive), unevenness (non-homogeneous repeated arrangement), uncertain characteristics such as (same class physiology or physical organization have uncertain texture structure).The biological tissue of human body texture all can be counted as the texture of non-structure rule.For this class texture, it is very difficult to extract geometric properties.The present invention has provided the completely new approach that makes this class textural characteristics regularization, can unify to handle different human body biological characteristics tissue texture, this just makes the present invention be different from prior art such as fingerprint identification technology fully, and can unite the identification of different human body biological characteristic.
B. method 1
We do through the note of the analyzed area after the provincial characteristics conversion R (x, y), i the zonule of Ri for being divided into by the graticule mesh after the conversion, (x0 y0) is the geometric center of this zonule.So that (x0 y0) sets up coordinate system for the center on each zonule, (xm yn) is the coordinate of any point among the Ri, and (xm yn) is point (xm, gray scale yn) to g.We will carry out the regularization conversion to the texture on each zonule under this coordinate system.Textural characteristics regularization conversion mechanism is as follows:
A. image is carried out smoothly: in the implementation method of " self-defined border/area detector ", we once carried out smoothly with detection boundaries or zone image.Here, we carry out image smoothing once more so that the textural characteristics regularization.Method is as follows: to gray-scale value g (xm, the yn) weighting (promptly multiply by a certain factor, or claim weight coefficient) of each point.These weight coefficients can also be chosen from sine function, cosine function, exponential function, hyperbolic function or other function except that coordinate xm, yn.
B. to given point, ask with different weight functions to g (xm, yn) carry out repeatedly level and smooth and, the even number kind is generally adopted in promptly multi-time weighted summation.
C. among the Ri ask the weighted sum ci described in above-mentioned a., the b. a little, then ci is that the regularization that region R i goes up this point texture feature characterizes.
C. method 2
Regularization table method: can be summarized in one group of numerical operation to the weighting smoothing algorithm of textural characteristics regularization conversion, on the zonule after the division, implement the numeric factors of spectrum analysis and be compiled into table to draw, during use if in table appropriate area choose numeric factors and do simple arithmetical operation, the regularization that just can realize the biological tissue of human body textural characteristics on this zonule characterizes, and this method is called regularization table method.Regularization table method only is a kind of simple calculating method of method 1, there is no essential distinction.
(4) isometric, maximum entropy coding
A. the problem that will solve
The regularization of the textural characteristics on each zonule characterized weave into code, as the digital representation of whole analyzed area epigraph texture, and as the foundation of feature identification.Because of analyzed area has been realized standardization, so the code of compiling is an equal-length code.
B. method
A. the weight function (sine function, cosine function etc.) in textural characteristics regularization conversion B.a. can be shone upon (in particular cases can be positive and negative two) each other, and regards as the regularization of mapping each other sign identical.
B. for satisfying above-mentioned requirements, the invariant f in the mapping must be got and make weighted value ci≤f that coding threshold value: region R i goes up each point, then correspondence code position value is 0, otherwise value is 1.
C. to definite biological tissue of human body, in textural characteristics regularization conversion B., all adopt the smoothing processing of same number, identical weight function, draw fixed-length coding thus.Can prove that this coding method is the maximum entropy coding, includes maximum fault information.
(5) condition code analysis
A. the problem that will solve
Two different images of given same biological tissue of human body, with obtaining two groups of codes (condition code) after the processing of the method in the above-mentioned steps (1) to (4), must analyze these two groups of codes, to determine that whether these two images are from same biological tissue of human body, that is to say, carry out identity and differentiate.
B. method
A. calculate the hamming distance h d of two groups of codes, i.e. the correspondence code position that value is different in two groups of codes shared number percent in whole sign indicating number positions;
B. determine authentication threshold value δ according to statistical nature, when hd≤δ, think same, otherwise think different.
Be to use the instantiation that the inventive method is discerned biological tissue of human body below:
1. iris recognition
Iris image is as shown in Figure 1: the ring-type among the figure between black pupil and white sclera is organized as iris, and yellow's iris is dark-brown.The iris recognition process is as follows:
(1) (x, λ) relevant with the value of f (x) on circular arc AB, parameter lambda is the radius of this circular arc to F.Use self-defined border/method for detecting area to obtain the inner and outer boundary of iris, determine analyzed area;
(2) on the ring-type analyzed area, set up graticule mesh in the following manner: outside diameter is divided into the k equal portions in the part of iris one side (left side black line segment among the figure), excircle is divided into l equal portions (among the figure shown in the white line segment), on iris, set up k
*The l graticule mesh is as the standardization analyzed area;
(3) the Ri regularization of going up textural characteristics in zoning characterizes ci, and ci is by w[m, n]
*G[m, n] try to achieve, wherein, and w[m, n] be weight function, g[m, n] be the gray scale of pixel [m, n];
(4) it is level and smooth to carry out two kinds or four kinds, and all iris images are kept same level and smooth order and same zonule (Ri) in proper order.Under two kinds or four kinds of level and smooth situations, code length is respectively 2
*K
*L and 4
*K
*L, K, L are determined by k, l respectively.Calculate hd and authentication threshold value δ, carry out authentication.
2. fingerprint recognition
Fingerprint image as shown in Figure 2, identifying is as follows:
(1) locating template that provides with reference to image capture device, and F (x, λ) relevant with the value of f (x) on the ellipse that with focal length λ is parameter, use self-defined border/method for detecting area to obtain the outer boundary of fingerprint.Oval interior zone is an analyzed area shown in the figure;
(2) on the ellipsometric analysis zone, set up graticule mesh in the following manner: transverse is divided into the k equal portions, minor axis is divided into the l equal portions, on the fingerprint image analyzed area, set up graticule mesh, as the standardization analyzed area;
(3) the Ri regularization of going up textural characteristics in zoning characterizes ci, and ci is by w[m, n]
*G[m, n] try to achieve, wherein, and w[m, n] be weight function, g[m, n] be the gray scale of pixel [m, n];
(4) it is level and smooth to carry out twice or four times, and all fingerprint images are kept same level and smooth order and same Ri order, under twice level and smooth and four level and smooth situation, all can obtain isometric code (four code lengths when level and smooth are twice when level and smooth two times).Calculate hd and authenticate threshold value δ and carry out authentication.
3. face recognition
Can suitably choose the image of facial a certain portion of tissue and discern, as shown in Figure 3, identifying is as follows:
(1) F (x, λ) relevant with the value of f (x) on the arc that with radius-of-curvature λ is parameter, use self-defined border/method for detecting area to obtain white unique point shown in the figure.The polygonal region of being delimited by unique point is an analyzed area.
(2) on the polygon analyzed area, set up graticule mesh in the following manner: the upper end horizontal line section is divided into the k equal portions, the line segment between the unique point of two ends about the image middle part is divided into the l equal portions, on face-image, set up graticule mesh, as the standardization analyzed area.
(3), (4) two parts are same as fingerprint recognition.
The present invention will be used for human body biological characteristics identification to applicant's an other invention " image microprocessing ", thereby provide the comprehensive recognition technology of human body biological characteristics such as iris, fingerprint, facial tissue.Use this comprehensive recognition technology can develop the comprehensive product of human body biological characteristics authentication.The appearance of this product has adapted to demand and the expectation of society to safe, practical authentication product; satisfied under network environment; in the no worker monitor mode; carry out the request for utilization of authentication fast and automatically, safely with high reliability; thereby can substitute existing authentication means; for society constructs new safety curtain one, for family and individual provide safeguard protection.
The application that the human body biological characteristics that uses the present invention to develop can be discerned product is summarised as three aspects: the control of 1. important gateway (is promptly controlled tangible door, or the passage on the physical significance), 2. information system access control (is promptly controlled invisible door, or the passage on the logical meaning), 3. be used in combination with the association area product.Following table provides some the concrete application items in the above-mentioned field:
| Application facet | Affiliated application | Application item for example |
| Important gateway control (passage on the control physical significance) | Social safety national security financial security | 1. gate control system, as: national treasury, arm store, drug storage, confidential office, proof box, antitheft door be public passage 2., as: attendance checking system, entry and exit port, subway inlet, college entrance examination examination hall |
| Information system access control (passage on the steering logic meaning) | Information security social safety financial security | 1. internet: Internet, Intranet, WAN, LAN is network application system 2.: MIS, the electronic banking system, ecommerce, government network application systems etc. are smart card 3., driver's license, I.D.s etc. are microcomputer 4., workstation, server start password etc. |
| Be used in combination with the association area product | Be used for mobile phone, household appliances, as the start password; Be used for the dynamic password device; Be used for automotive theft proof system etc. |
Claims (6)
1. a method for identifying human body biological characteristics that uses image capture device and computing machine to carry out the steps include:
(1) separation of images
Gather the biological tissue of human body image with image capture device, from this biological tissue of human body image, detect edge, outline line or the zone of particular analysis object with computing machine, so as to separate, the extraction and analysis object;
(2) regional standard conversion
Is to set up the standard grid on the analyzed area with computing machine at isolated image-region, carries out the regional standard conversion;
(3) textural characteristics regularization conversion
With computing machine texture in the normal grid is carried out the regularization conversion;
(4) isometric, maximum entropy coding
With computing machine the regularization sign of texture on each zonule is weaved into code, as the digital representation of whole analyzed area epigraph texture;
(5) condition code analysis
Two groups of codes after with computing machine two different images of same biological tissue of human body being handled with the method in above-mentioned (1)-(4) are analyzed, to determine that this two image is whether from same people's same biological tissue of human body, promptly carry out authentication, the correspondence code position that value is different in two groups of codes in whole sign indicating number positions shared number percent smaller or equal to the setting person, then this two image is same, otherwise is different.
2. method for identifying human body biological characteristics as claimed in claim 1 is characterized in that the method for described separation of images is:
With the two dimensional gray image of biological tissue of human body with F (x) expression, the position of picture element in the x presentation video wherein, f (x) is the gray scale of this point; Border in the identifying/regional automatic detecting machine system is called self-defined border/area detector, and the process of self-defined border/area detector is as follows:
The structure in a. self-defined border that will detect or zone: if the border of detecting is to be made of straight-line segment, just define linear barrier's detecting device F (x, λ), this detecting device normally image f (x) is the line integral or the area branch of parameter with λ, F (x, λ) be the function of f (x), λ is and the border that will detect or the regional relevant geometric parameter of shape;
B. detecting device is carried out smoothly: to edge detector F (x λ) determines its action scope, and to edge detector F (x, each point on λ) carries out on action scope smoothly with smooth function;
C. λ is differentiated or calculate the change amount: the variation of the general characteristic that each may the border when calculating λ and changing, with from searching out real border the border;
D. calculate the extreme value of derivative or change amount: determine real border by the extreme value of calculating derivative or change amount.
3. method for identifying human body biological characteristics as claimed in claim 1, the method that it is characterized in that described regional standard conversion are all or part of among the following method a-d:
A. another zone is arrived in the image transformation of a certain analyzed area on the plane;
B. the border with a certain analyzed area in the image is another kind of shape from a kind of shape conversion;
C. the textural characteristics to analytic target inside carries out standardization;
D. image is transformed to another coordinate space from a coordinate space;
Generally speaking, and do not require whole conversion of carrying out in the above-mentioned requirements, now provide the general approach of three of above-mentioned a, b, c:
A. set up normal grid in institute on to analyzed area, set up graticule mesh and will follow two principles: the one, to setting up with a kind of graticule mesh with a kind of physiological tissue, the 2nd, divide grid in the graticule mesh according to the geometric configuration in zone; If the geometric configuration of analyzed area has been suitable for analyzing and processing, normal grid can no longer be carried out other conversion after setting up;
B. if the geometric configuration of analyzed area is carried out conversion, then the line segment in the graticule mesh must be transformed to the target area one by one, conversion is carried out in the following manner: just establishing, the intersection point of the line segment a in conversion and boundary contour C1 is x1, x2, a (x) is the point on the line segment a, then after the conversion value of image f (x) in the target area by two known standardization function h1 (x, x1), (x x2) determines h2.
4. method for identifying human body biological characteristics as claimed in claim 1 is characterized in that the method for described textural characteristics regularization conversion is:
With the geometric center through each zonule in the analyzed area after the provincial characteristics conversion is that coordinate system is set up in the center on each zonule, under this coordinate system the texture on each zonule is being carried out following regularization conversion:
A. image is carried out smoothly: to the gray-scale value weighting of each point, these weight coefficients can also be chosen from sine function, cosine function, exponential function, hyperbolic function or other function except that coordinate;
B. to given point, ask with different weight functions to carry out repeatedly level and smooth and, the even number kind is generally adopted in promptly multi-time weighted summation;
C. in the zonule ask the weighted sum described in above-mentioned a., the b. a little, then this weighted sum is that the regularization of this point texture feature on this zonule characterizes.
5. method for identifying human body biological characteristics as claimed in claim 1 is characterized in that described isometric, maximum entropy Methods for Coding is:
A. the weight function in textural characteristics regularization conversion can be shone upon each other, and regards as the regularization of mapping each other sign identical;
B. for satisfying above-mentioned requirements, constant the measuring in the mapping made the coding threshold value, the weighted value of each point is smaller or equal to this coding threshold value on each zonule, and correspondence code position value is 0, otherwise value is 1;
C. to definite biological tissue of human body, all adopt the smoothing processing of same number, identical weight function in textural characteristics regularization conversion, draw fixed-length coding thus, this coding method is the maximum entropy coding, includes maximum fault information.
6. method for identifying human body biological characteristics as claimed in claim 1 is characterized in that the method for described condition code analysis is:
A. calculate correspondence code position shared number percent in whole sign indicating numbers position that value is different in two groups of codes;
B. determine the authentication threshold value according to statistical nature, when above-mentioned number percent authenticates threshold value smaller or equal to this, think same, otherwise think different.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB001249541A CN1144158C (en) | 2000-09-25 | 2000-09-25 | Method for identifying human body biological characteristics |
| PCT/CN2001/001452 WO2002080088A1 (en) | 2000-09-25 | 2001-09-25 | Method for biometric identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB001249541A CN1144158C (en) | 2000-09-25 | 2000-09-25 | Method for identifying human body biological characteristics |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1346116A true CN1346116A (en) | 2002-04-24 |
| CN1144158C CN1144158C (en) | 2004-03-31 |
Family
ID=4590768
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB001249541A Expired - Fee Related CN1144158C (en) | 2000-09-25 | 2000-09-25 | Method for identifying human body biological characteristics |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN1144158C (en) |
| WO (1) | WO2002080088A1 (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1294521C (en) * | 2004-06-28 | 2007-01-10 | 李剑华 | Outer shape structure of commercial guest greeting robot and identifying method |
| CN100353379C (en) * | 2003-07-23 | 2007-12-05 | 西北工业大学 | An image retrieval method based on image grain characteristic |
| CN1792060B (en) * | 2003-05-21 | 2011-05-25 | 皇家飞利浦电子股份有限公司 | Methd and system for authenticating physical object |
| CN101072100B (en) * | 2006-05-12 | 2012-03-28 | 联想(北京)有限公司 | Authenticating system and method utilizing reliable platform module |
| CN101159554B (en) * | 2006-10-04 | 2012-05-23 | 株式会社日立制作所 | Biometric authentication system, enrollment terminal, authentication terminal and authentication server |
| CN101520844B (en) * | 2002-05-09 | 2012-05-30 | 索尼株式会社 | Blood vessel authentication apparatus |
| CN102016918B (en) * | 2008-04-28 | 2014-04-16 | 公立大学法人大阪府立大学 | Method for creating image database for object recognition, processing device |
| CN101903891B (en) * | 2007-12-20 | 2014-12-17 | 皇家飞利浦电子股份有限公司 | Defining classification thresholds in template protection systems |
| CN106326869A (en) * | 2016-08-26 | 2017-01-11 | 广东欧珀移动通信有限公司 | Fingerprint identification method, device and mobile terminal |
| CN113569792A (en) * | 2021-08-05 | 2021-10-29 | 北京惠朗时代科技有限公司 | Intelligent safety box application method and device based on accurate fingerprint identification |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB9516611D0 (en) * | 1995-08-14 | 1995-10-18 | At & T Global Inf Solution | An authorisation system |
| US5815252A (en) * | 1995-09-05 | 1998-09-29 | Canon Kabushiki Kaisha | Biometric identification process and system utilizing multiple parameters scans for reduction of false negatives |
| US5995630A (en) * | 1996-03-07 | 1999-11-30 | Dew Engineering And Development Limited | Biometric input with encryption |
| AU3022897A (en) * | 1997-02-21 | 1998-09-09 | Kaba Schliesssysteme Ag | Method and device for automatic face recognition |
-
2000
- 2000-09-25 CN CNB001249541A patent/CN1144158C/en not_active Expired - Fee Related
-
2001
- 2001-09-25 WO PCT/CN2001/001452 patent/WO2002080088A1/en active Application Filing
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101520844B (en) * | 2002-05-09 | 2012-05-30 | 索尼株式会社 | Blood vessel authentication apparatus |
| CN1792060B (en) * | 2003-05-21 | 2011-05-25 | 皇家飞利浦电子股份有限公司 | Methd and system for authenticating physical object |
| CN100353379C (en) * | 2003-07-23 | 2007-12-05 | 西北工业大学 | An image retrieval method based on image grain characteristic |
| CN1294521C (en) * | 2004-06-28 | 2007-01-10 | 李剑华 | Outer shape structure of commercial guest greeting robot and identifying method |
| CN101072100B (en) * | 2006-05-12 | 2012-03-28 | 联想(北京)有限公司 | Authenticating system and method utilizing reliable platform module |
| CN101159554B (en) * | 2006-10-04 | 2012-05-23 | 株式会社日立制作所 | Biometric authentication system, enrollment terminal, authentication terminal and authentication server |
| CN101903891B (en) * | 2007-12-20 | 2014-12-17 | 皇家飞利浦电子股份有限公司 | Defining classification thresholds in template protection systems |
| CN102016918B (en) * | 2008-04-28 | 2014-04-16 | 公立大学法人大阪府立大学 | Method for creating image database for object recognition, processing device |
| CN106326869A (en) * | 2016-08-26 | 2017-01-11 | 广东欧珀移动通信有限公司 | Fingerprint identification method, device and mobile terminal |
| CN106326869B (en) * | 2016-08-26 | 2021-01-12 | Oppo广东移动通信有限公司 | Fingerprint identification method and device and mobile terminal |
| CN113569792A (en) * | 2021-08-05 | 2021-10-29 | 北京惠朗时代科技有限公司 | Intelligent safety box application method and device based on accurate fingerprint identification |
| CN113569792B (en) * | 2021-08-05 | 2023-12-01 | 北京惠朗时代科技有限公司 | Intelligent safe box application method and device based on accurate fingerprint identification |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1144158C (en) | 2004-03-31 |
| WO2002080088A1 (en) | 2002-10-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Dai et al. | Multifeature-based high-resolution palmprint recognition | |
| CN103324944B (en) | A False Fingerprint Detection Method Based on SVM and Sparse Representation | |
| CN101093539A (en) | Matching identification method by extracting characters of vein from finger | |
| Poursaberi et al. | A novel iris recognition system using morphological edge detector and wavelet phase features | |
| CN1760887A (en) | Robust Feature Extraction and Recognition Method for Iris Image | |
| CN1144158C (en) | Method for identifying human body biological characteristics | |
| Savithiri et al. | Performance analysis on half iris feature extraction using GW, LBP and HOG | |
| Al-Helali et al. | A Fast personal palm print authentication Based on 3d-multi–Wavelet Transformation | |
| Mokni et al. | Pre-processing and extraction of the ROIs steps for palmprints recognition system | |
| Priya et al. | Authentication of identical twins using tri modal matching | |
| Bansal et al. | FAR and FRR based analysis of iris recognition system | |
| CN105488460A (en) | Physiological feature based image processing method | |
| CN1246800C (en) | Synergic fingerprint identification method | |
| Samai et al. | Oriented Local Binary Pattern (LBP θ): A new scheme for an efficient feature extraction technique | |
| Praks et al. | On SVD-free Latent Semantic Indexing for iris recognition of large databases | |
| Mohammed et al. | Developing iris recognition system based on enhanced normalization | |
| CN1305001C (en) | Finger print characteristic matching method in intelligent card | |
| Hussein et al. | Palmprint verification using invariant moments based on wavelet transform | |
| Attallah et al. | Application of BSIF, Log-Gabor and mRMR transforms for iris and palmprint based Bi-modal identification system | |
| Palanikumar et al. | Advanced palmprint recognition using unsharp masking and histogram equalization | |
| Chirchi et al. | Enhanced isocentric segmentor and wavelet rectangular coder to iris segmentation and recognition | |
| Vélez et al. | Robust ear detection for biometric verification | |
| CN1924891A (en) | Differential operation based high-precision palm print recognition method | |
| Dehghani et al. | Retinal identification based on rotation invariant moments | |
| Daniel et al. | A data fusion technique designed for multimodal biometric systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C06 | Publication | ||
| PB01 | Publication | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| ASS | Succession or assignment of patent right |
Owner name: BEIJING CITY NUOJIN INFORMATION TECHNOLOGICAL CO. Free format text: FORMER OWNER: WANG JIESHENG Effective date: 20050422 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20050422 Address after: Two Zhongguancun North Street 100080 No. 4 Beijing city Haidian District shuiqingmuhua Park Building No. 4 room 1604 Patentee after: Beijing King Information Technology Co., Ltd. Address before: 100080 No. 952, building 607, South Road, Beijing Academy of Sciences, Zhongguancun Patentee before: Wang Jiesheng |
|
| ASS | Succession or assignment of patent right |
Owner name: BEIJING ARITHEN INFORMATION TECHNOLOGY CO., LTD. Free format text: FORMER OWNER: BEIJING CITY NUOJIN INFORMATION TECHNOLOGICAL CO.,LTD. Effective date: 20050729 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20050729 Address after: 100080, room 1501, Haidian new technology building, No. 65 West Fourth Ring Road, Haidian District, Beijing Patentee after: Kaipingaisen IT Tech Co., Ltd., Beijing Address before: Two Zhongguancun North Street 100080 No. 4 Beijing city Haidian District shuiqingmuhua Park Building No. 4 room 1604 Patentee before: Beijing King Information Technology Co., Ltd. |
|
| DD01 | Delivery of document by public notice |
Addressee: Beijing arithen Information Technology Co., Ltd. Intellectual Property Department Document name: Notification to Pay the Fees |
|
| DD01 | Delivery of document by public notice |
Addressee: Wang Qin Document name: Notification to Pay the Fees |
|
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20040331 Termination date: 20110925 |