CN102479328B - Identity verification device and method based on biological characteristics - Google Patents
Identity verification device and method based on biological characteristics Download PDFInfo
- Publication number
- CN102479328B CN102479328B CN201110141655.7A CN201110141655A CN102479328B CN 102479328 B CN102479328 B CN 102479328B CN 201110141655 A CN201110141655 A CN 201110141655A CN 102479328 B CN102479328 B CN 102479328B
- Authority
- CN
- China
- Prior art keywords
- comparison result
- biometric image
- processing unit
- image
- contraposition
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 77
- 238000012795 verification Methods 0.000 title abstract 2
- 238000012545 processing Methods 0.000 claims abstract description 95
- 238000000605 extraction Methods 0.000 claims abstract description 43
- 239000000284 extract Substances 0.000 claims description 5
- 238000012958 reprocessing Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 15
- 238000006073 displacement reaction Methods 0.000 description 3
- 239000012634 fragment Substances 0.000 description 2
- 210000001015 abdomen Anatomy 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Landscapes
- Collating Specific Patterns (AREA)
Abstract
The invention discloses an identity authentication device and method based on biological characteristics. The method comprises the following steps: storing the first biological characteristic image and the corresponding first characteristic point information in a storage unit; extracting a second biological characteristic image of the target to be verified by using an image extraction unit; analyzing a plurality of feature points of the second biological feature image and correspondingly providing second feature point information through a biological feature extraction unit; comparing the first characteristic point information with the second characteristic point information through the processing unit to obtain a first comparison result and alignment information; the reprocessing unit carries out alignment of the first biological characteristic image and the second biological characteristic image according to the alignment information and then carries out pattern comparison of the first biological characteristic image and the second biological characteristic image so as to obtain a second comparison result; and judging whether the object to be verified passes the identity verification or not by using the processing unit according to the first comparison result and the second comparison result.
Description
Technical field
The present invention relates to a kind of authentication means and method thereof, particularly relates to a kind of authentication means and method thereof based on biological characteristic.
Background technology
Use all the time key or personal identification number as the instrument of security protection easily because of thief, hacker attacks or forget password safety is started a leak and inconvenience.
Along with biometrics identification technology progress day by day, use biological characteristic quite universal as the mode of identification personal identification.At present, existing living things feature recognition method mostly is and utilizes pattern comparison (pattern matching) or unique point comparison (minutiae matching) these two kinds of modes to carry out identification.Yet utilize the biometrics identification technology of pattern comparison, its shortcoming is that the data volume of image is large, and in comparison process, also have the problems such as displacement, the anglec of rotation and figure deformation to need to process, therefore comparison speed is slow, easily waste the time that user waits for authentication.
On the other hand, utilize the biometrics identification technology of unique point comparison, though its advantage compared with pattern comparison fast, but shortcoming be when user's custom change (take fingerprint as example but be not limited to: the angle/speed of pressing or sliding) or environment change (such as but not limited to: in the time of humiture), may cause extracted unique point lazy weight, therefore the biometrics identification technology that is difficult to compare by unique point is obtained authentication, thus, not only cause user's inconvenience, more lost and utilized unique biological characteristic as the advantage of authentication.
In sum, existing biometrics identification technology cannot be identified target to be identified fast and effectively, to carry out the operation of authentication, therefore consider for these reasons, the present inventor thinks and designs a kind of authentication means and method thereof based on biological characteristic, to the defect for prior art, improved, and then the enforcement utilization in enhancement industry.
Summary of the invention
Because the problem of above-mentioned prior art, object of the present invention is just to provide a kind of authentication means and method thereof based on biological characteristic, to solve pattern, compares slow and data volume is large and unique point is very few and cannot utilize unique point Comparison Method by the problem of authentication.
According to object of the present invention, a kind of authentication means based on biological characteristic is proposed, it comprises storage element, Extraction of Image unit, biological characteristic extraction unit and a processing unit.Storage unit stores the first biometric image and corresponding First Characteristic dot information.Extraction of Image unit extracts the second biometric image to target to be verified.A plurality of unique points correspondence that biological characteristic extraction unit is analyzed the second biometric image provide Second Characteristic dot information.Processing unit is electrically connected to respectively storage element, Extraction of Image unit and biological characteristic extraction unit.Processing unit comparison First Characteristic dot information and Second Characteristic dot information, to draw the first comparison result and contraposition information, according to contraposition information, carry out the contraposition of the first biometric image and the second biometric image, carrying out the pattern of the first biometric image and the second biometric image compares again, to draw the second comparison result, then according to the first comparison result and the second comparison result, judge that whether target to be verified is by authentication.
Wherein, processing unit, more according to default-weight ratio, calculates the 3rd comparison result by the first comparison result and the second comparison result by expression formula, and according to the 3rd comparison result and threshold value, judges that whether target to be verified is by authentication.
Wherein, processing unit also draws the second comparison result according to the similarity of the first biometric image and the second biometric image, and by expression formula, calculate threshold value according to the second comparison result, again according to the first comparison result and threshold value, judge that whether target to be verified is by authentication, wherein expression formula makes the similarity corresponding with the second comparison result and threshold value be negative correlation.
Wherein, processing unit system is according to contraposition information contraposition the first biometric image and the second biometric image, feature extraction unit more identifies a singular point (Singular Point) of the second biometric image according to Second Characteristic dot information, processing unit launches a specific region scope to surrounding centered by singular point, and compare the first biometric image and the second biometric image within the scope of specific region in the mode of pattern comparison, and obtain the second comparison result.
Wherein, processing unit is according to contraposition information contraposition the first biometric image and the second biometric image, and compares with the pattern that the mode of contract drawing is carried out the first biometric image and the second biometric image, and obtains the second comparison result.
According to object of the present invention, a kind of auth method based on biological characteristic is also proposed.Auth method based on biological characteristic is applicable to the authentication means based on biological characteristic, and this authentication means based on biological characteristic comprises storage element, Extraction of Image unit, biological characteristic extraction unit and processing unit.Auth method based on biological characteristic comprises the following step: first at storage unit stores the first biometric image and corresponding First Characteristic dot information; Then utilize Extraction of Image unit to extract the second biometric image to target to be verified; A plurality of unique points the correspondence of by biological characteristic extraction unit, analyzing the second biometric image again provide Second Characteristic dot information; Then by processing unit, compare First Characteristic dot information and Second Characteristic dot information, to draw the first comparison result and contraposition information; Next at processing unit, according to contraposition information, carry out the contraposition of the first biometric image and the second biometric image, then the pattern that carries out the first biometric image and the second biometric image compares, to draw the second comparison result; And utilize processing unit to judge that according to the first comparison result and the second comparison result whether target to be verified is by authentication.
Wherein, the auth method based on biological characteristic also comprises the following step: according to default-weight ratio, utilize processing unit that the first comparison result and the second comparison result are calculated to the 3rd comparison result by expression formula; Then by processing unit according to the 3rd comparison result and threshold value, judge that whether target to be verified is by authentication.
Wherein, the auth method based on biological characteristic also comprises the following step: utilize processing unit to draw the second comparison result according to the similarity of the first biometric image and the second biometric image; And according to the second comparison result, by expression formula, calculate threshold value at processing unit; And by processing unit according to the first comparison result and threshold value, judge that whether target to be verified is by authentication.Wherein, expression formula makes the similarity corresponding with the second comparison result and threshold value be negative correlation.
Wherein, at processing unit, carry out, in the step of pattern comparison, also comprising the following step: utilize processing unit according to contraposition information contraposition the first biometric image and the second biometric image; Then in feature extraction unit, according to Second Characteristic dot information, identify the singular point (Singular Point) of the second biometric image; And utilize processing unit to surrounding, to launch specific region scope centered by singular point, and compare the first biometric image and the second biometric image within the scope of specific region in the mode of pattern comparison, and obtain the second comparison result.
Wherein, with processing unit, carry out, in the step of pattern comparison, also comprising the following step: utilize processing unit according to contraposition information contraposition the first biometric image and the second biometric image; And the pattern that carries out the first biometric image and the second biometric image in the mode of contract drawing by processing unit is compared, and obtains the second comparison result.
As mentioned above, according to authentication means and the method thereof based on biological characteristic of the present invention, it can have one or more in following advantage:
(1) authentication means that should be based on biological characteristic and method thereof can be compared resulting contraposition information by unique point and carry out two image contrapositions, because of the quantity of information of unique point fewer than the quantity of information of pattern, so faster than pattern (pattern) contraposition with unique point (minutiae) contraposition, can solve thus the problem consuming time that prior art is caused with pattern contraposition two images.
(2) this authentication means and method thereof based on biological characteristic can be by processing unit according to different weight proportions, bring the first comparison result and the second comparison result into expression formula to calculate the 3rd comparison result, can solving thus prior art, with unique point Comparison Method, to carry out unique point that authentication causes not enough and cannot be by the shortcoming of authentication.
(3) result that this authentication means based on biological characteristic and method thereof can be compared by pattern is dynamically adjusted the threshold value of unique point comparison, and due to above-mentioned both be negative correlation, can solve thus the very few problem being still able to by authentication of user characteristics point.
(4) this authentication means and method thereof based on biological characteristic can be by the pattern within the scope of comparison specific region, reduce in prior art pattern comparison the excessive problem of the data volume that causes, improve thus the speed of comparison, then compare with pattern and compare the precision that reaches authentication by unique point.
Accompanying drawing explanation
Fig. 1 is the calcspar of the first embodiment of the authentication means based on biological characteristic of the present invention.
Fig. 2 is the schematic diagram of the first embodiment of the authentication means based on biological characteristic of the present invention.
Fig. 3 is the calcspar of the second embodiment of the authentication means based on biological characteristic of the present invention.
Fig. 4 is the schematic diagram of the second embodiment of the authentication means based on biological characteristic of the present invention.
Fig. 5 is the process flow diagram of the auth method based on biological characteristic of the present invention.
Fig. 6 is another process flow diagram of the auth method based on biological characteristic of the present invention.
Fig. 7 is a process flow diagram again of the auth method based on biological characteristic of the present invention.
Fig. 8 is the process flow diagram of image of the comparison specific region scope of the auth method based on biological characteristic of the present invention.
Fig. 9 is the process flow diagram that carries out pattern comparison in contract drawing mode of the auth method based on biological characteristic of the present invention.
Primary clustering symbol description: 1 is the authentication means based on biological characteristic, 10 is storage element, 100 is the first biometric image, 101 is First Characteristic dot information, 11 is Extraction of Image unit, 110 is the second biometric image, 111 is Second Characteristic dot information, 12 is biological feature extraction unit, 13 is processing unit, 130 is the first comparison result, 131 is contraposition information, 132 is the second comparison result, 133 is the 3rd comparison result, 134 is threshold value, 2 is target to be verified, 30 is area formula fingerprint sensor, 31 is finger, 32 is main frame, 33 is fingerprint image, 330 is singular point, 331 is specific region scope, 40 is swipe fingerprint sensor, 41 is fingerprint image, S10～S15, S20～S21, S30～S32, S40～S42, S50～S51 is step.
Embodiment
With reference to Fig. 1, it is the calcspar of the first embodiment of the authentication means based on biological characteristic of the present invention.As shown in the figure, authentication means 1 based on biological characteristic comprises storage element 10, Extraction of Image unit 11, biological characteristic extraction unit 12 and processing unit 13, wherein, processing unit 13 can be central processing unit (Central Processing Unit, CPU) or microprocessor (Micro-Processing Unit), and be electrically connected respectively storage element 10, Extraction of Image unit 11 and biological characteristic extraction unit 12, storage element 10 can be various storage mediums, embedded internal memory for example, circumscribed memory card (body), disk sheet or disc etc., and store the first biometric image 100 and corresponding First Characteristic dot information 101, but the quantity of the First Characteristic dot information of the first stored biometric image and correspondence is not limited to this, that is to say, storage element 10 can store a plurality of the first biometric images and corresponding a plurality of First Characteristic dot informations.
When user carries out authentication with target 2 to be verified (as: fingerprint, iris, people's face or the lines of the hand), should the authentication means 1 based on biological characteristic first utilize 11 pairs of Extraction of Image unit target 2 to be verified to extract the second biometric image 110.Then, processing unit 13 is controlled a plurality of unique points that biological characteristic extraction unit 12 is analyzed the second biometric image 110, obtains thus the Second Characteristic dot information 111 corresponding with the second biometric image 110.Then, processing unit 13 is with unique point Comparison Method (minutiae matching) comparison First Characteristic dot information 101 and Second Characteristic dot information 111, to draw the first comparison result 130 and contraposition information 131.Next, processing unit 13 carries out the contraposition of the first biometric image 100 and the second biometric image 110 according to contraposition information 131, after contraposition completes, processing unit 13 is with pattern Comparison Method (pattern matching) comparison the first biometric image 100 and the second biometric image 110, to draw the second comparison result 132, and judge that according to the first comparison result 130 and the second comparison result 132 whether target 2 to be verified is by authentication.
As mentioned above, processing unit 13 is according to default-weight ratio, bring the first comparison result 130 and the second comparison result 132 into expression formula, after calculating, obtain the 3rd comparison result 133, processing unit 13 is set threshold value 134 again, and judges that with threshold value 134 whether target 2 to be verified is by authentication according to the 3rd comparison result 133.
Wherein, Extraction of Image unit 11 can be the various devices that extract image, such as: fingerprint sensor, iris sensor, video camera or camera etc.In addition, processing unit 13 and biological characteristic extraction unit 12 can be individually independently compute chip, also can be integrated into same compute chip.
With reference to Fig. 2, it is the schematic diagram of the first embodiment of the authentication means based on biological characteristic of the present invention.As shown in the figure, Extraction of Image unit is implemented by area formula fingerprint sensor 30, and storage element, biological characteristic extraction unit and processing unit are implemented by main frame 32.When user presses area formula fingerprint sensor 30 with finger 31, first area formula fingerprint sensor 30 extracts the fingerprint image 33 of finger 31, and transmits fingerprint image 33 to main frame 32.Then, biological characteristic extraction unit is analyzed a plurality of unique points of fingerprint image 33, and correspondence provides Second Characteristic dot information.
Storage element can store a plurality of the first fingerprint images and each First Characteristic dot information corresponding with the first fingerprint image, processing unit carries out the unique point of Second Characteristic dot information and First Characteristic dot information and compares (minutiae matching), and drawing the first comparison result and contraposition information, wherein contraposition information at least comprises the anglec of rotation and displacement.Processing unit draws the first comparison result according to the similarity of Second Characteristic dot information and First Characteristic dot information, and in the present embodiment, the first comparison result is comparison mark, and for example: similarity reaches 30%, comparing mark is 30 minutes.
When processing unit draws after the first comparison result, according to unique point, compare resulting contraposition information and carry out the contraposition of the first fingerprint image and fingerprint image 33, and because the data volume of characteristic point information is little compared with the data volume of image, that therefore with unique point Comparison Method contraposition image ratio, with pattern Comparison Method, bit image is come is quick, thus, by the present invention, with unique point, compare resulting contraposition information and carry out the contraposition of image, can reduce the image that carries out before pattern comparison to bit time.
Wherein, First Characteristic dot information and Second Characteristic dot information at least comprise type information, coordinate information, angle information, curvature information or the directional information of each unique point, the end points that type information comprises each unique point (Endings), bifurcation point (Bifurcations), core (Core) or trigpoint (Delta).
When processing unit completes after the contraposition of the first fingerprint image and fingerprint image 33, processing unit carries out pattern according to the first fingerprint image and fingerprint image 33 and compares (pattern matching).It is worth mentioning that, in carrying out the process of pattern comparison, the problem consuming time causing in order to reduce the pattern comparison of full fingerprint, the present inventor proposes to be identified according to Second Characteristic dot information by feature extraction unit the singular point 330 (Singular Point) of the second biometric image especially, , feature extraction unit can be determined singular point 330 according to a plurality of unique points of the type analysis of the unique points such as core or trigpoint, processing unit launches specific region scope 331 to surrounding again centered by singular point 330, and according to the fingerprint image 33 in specific region scope 331 and the first fingerprint image, carry out pattern and compare, and draw the second comparison result, like this, the speed that can shorten pattern comparison also can reduce the data volume of stored image, because of the deformation probability of the fingerprint image within the scope of this specific region, be less than again the deformation probability of full fingerprint image, if so validation problem that has figure deformation to cause, can lower full fingerprint image comparison time by the fingerprint image within the scope of comparison specific region.
Although it should be noted that aforesaid specific region scope is to take squarely concept of the present invention to be described as example, those skilled in the art can understand and may extend to circle, ellipse or polygon easily, end is seen user's setting.
Then, processing unit can draw the second comparison result according to the similarity of the first fingerprint image and fingerprint image 33, and in the present embodiment, the second comparison result is comparison mark, and for example: similarity reaches 80%, comparing mark is 80 minutes.
Next, the present inventor can rule of thumb give the first weighted value to compare resulting the first comparison result by unique point, and gives the second weighted value to compare resulting the second comparison result by pattern.In the present embodiment, processing unit sets the first weighted value and the second weighted value is all 0.5, but not as limit.In addition, it is 45 minutes that processing unit is also set threshold value, and calculates the 3rd comparison result by expression formula, as follows:
30×0.5+80×0.5＝55。
As above formula can draw the 3rd comparison result, be 55 minutes, then processing unit judges whether the 3rd comparison result is greater than threshold value again, because the 3rd comparison result is within 55 minutes, to be greater than threshold value 45 minutes, so processing unit judges that this user is by authentication.
As above-described embodiment, can inspect out in the prior art cannot be by the target to be verified of authentication with unique point Comparison Method, can be by the authentication means based on biological characteristic of the present invention according to unique point Comparison Method and the resulting comparison result of pattern Comparison Method, preset weight proportion, improve thus the accuracy rate of authentication, also can avoid (for example changing because of target use habit to be identified, but be not limited to: the fingerprint extracting when user stands mostly is finger tip, the fingerprint extracting when user is seated mostly is finger abdomen, excessive velocities or excessively slow during slip, angular deflection etc. during slip) or environment change (for example, but be not limited to: humiture, it is dirty or old etc. that sensor is crossed), cause extracted characteristic point information deficiency and cannot utilize unique point Comparison Method to carry out the problem of authentication.
With reference to Fig. 3, it is the calcspar of the second embodiment of the authentication means based on biological characteristic of the present invention.In figure, its framework is roughly identical with the first embodiment, both difference is, processing unit 13 according to the second comparison result 132 by expression formula, through calculating a threshold value 134, so threshold value 134 is dynamically adjusted along with the second comparison result 132, and this expression formula makes the similarity corresponding with the second comparison result 132 and threshold value 134 be negative correlation, that is to say when similarity higher, threshold value 134 is lower, anti-when similarity lower, threshold value 134 is higher.Then,, when processing unit 13 calculates after threshold value 134, processing unit 13 judges that with threshold value 134 whether target 2 to be verified is by authentication according to the first comparison result 130.
With reference to Fig. 4, it is the schematic diagram of the second embodiment of the authentication means based on biological characteristic of the present invention.As shown in the figure, Extraction of Image unit is implemented by swipe fingerprint sensor 40, but is not limited to this.When user slips over swipe fingerprint sensor 40 by finger 31 in the mode of sliding, swipe fingerprint sensor 40 extracts a plurality of fingerprint image fragments, and by processing unit, analyze the coincidence part of a plurality of fingerprint image fragments, synthesize complete fingerprint image 41.Difference that it should be noted that the present embodiment and the first embodiment is, processing unit is that the pattern that carries out the first fingerprint image and fingerprint image 41 in the mode of contract drawing is compared, and obtains the second comparison result.Wherein the mode of contract drawing can comprise the size of downscaled images and can arrange in pairs or groups and changes the resolution etc. of image, and with the advantage that the mode of contract drawing is carried out pattern comparison be to save storage element storage area and can by contract drawing effectively quickening figure by the speed of displacement in comparison process and rotation.
In addition, processing unit draws the second comparison result according to the first fingerprint image 100 and the similarity of fingerprint image 41, and in the present embodiment, the second comparison result is comparison mark, supposes that comparison mark is 70 minutes.Then, processing unit is brought the second comparison result into expression formula and is calculated a threshold value, obtains threshold value and is 21 minutes, as follows:
70×0.3＝21。
Whether the fingerprint that processing unit judges user according to the first comparison result and threshold value is by authentication,, when the first comparison result surpasses threshold value 21 timesharing, processing unit judges that this user is by authentication, otherwise, when the first comparison result is less than threshold value 21 timesharing, processing unit judges this subscriber authentication failure.
As mentioned above, expression formula makes the similarity corresponding with the second comparison result and threshold value be negative correlation, as shown in the table:
| The second comparison result (comparison mark) | Take advantage of (sign of operation) |
| Be greater than 90 | 0.0 |
| 70～90 | 0.3 |
| 60～70 | 0.35 |
| 50～60 | 0.4 |
| 40～50 | 0.7 |
| 30～40 | 0.9 |
| 20～30 | 1.2 |
| Lower than 20 | 1.4 |
It is as shown in the table, when the second comparison result higher, after expression formula is calculated, drawn threshold value is lower, , when higher with the similarity of pattern comparison, tolerable is lower with the similarity of unique point comparison, because the second comparison result and threshold value are negative correlation, so processing unit can be according to the second comparison result dynamic adjusted threshold value, except overcoming unique point lazy weight that prior art extracted when comparing with unique point and cannot be by the puzzlement of authentication, also can overcome the problem consuming time being caused with image contraposition by compare resulting contraposition information with unique point, and more by the mode of contract drawing, improve the speed of pattern comparison.
Although aforementioned in the process of the explanation authentication means based on biological characteristic of the present invention, the concept of the auth method based on biological characteristic of the present invention is described simultaneously, for ask clear for the purpose of, below still provide process flow diagram with detailed description.
With reference to Fig. 5, it is the process flow diagram of the auth method based on biological characteristic of the present invention.As shown in the figure, auth method based on biological characteristic of the present invention is applicable to the authentication means based on biological characteristic, it comprises storage element, Extraction of Image unit, biological characteristic extraction unit and processing unit, and the auth method based on biological characteristic comprises the following step:
S10: at storage unit stores the first biometric image and corresponding First Characteristic dot information;
S11: utilize Extraction of Image unit to extract the second biometric image to target to be verified;
S12: analyze a plurality of unique points of the second biometric image and Second Characteristic dot information is provided accordingly by biological characteristic extraction unit;
S13: compare First Characteristic dot information and Second Characteristic dot information by processing unit, to draw the first comparison result and contraposition information;
S14: the contraposition of carrying out the first biometric image and the second biometric image according to contraposition information at processing unit is carried out both patterns again and compared, to draw the second comparison result; And
S15: utilize processing unit to judge that according to the first comparison result and the second comparison result whether target to be verified is by authentication.
With reference to Fig. 6, it is another process flow diagram of the auth method based on biological characteristic of the present invention.As shown in the figure, the auth method based on biological characteristic of the present invention also comprises the following step:
S20: according to default-weight ratio, utilize processing unit that the first comparison result and the second comparison result are calculated to the 3rd comparison result by expression formula; And
S21: according to the 3rd comparison result and threshold value, judge that whether target to be verified is by authentication by processing unit.
With reference to Fig. 7, it is a process flow diagram again of the auth method based on biological characteristic of the present invention.As shown in the figure, the auth method based on biological characteristic of the present invention also comprises the following step:
S30: utilize processing unit to draw the second comparison result according to the similarity of the first biometric image and the second biometric image;
S31: calculate threshold value according to the second comparison result by expression formula at processing unit, wherein this expression formula makes the similarity corresponding with the second comparison result and threshold value be negative correlation; And
S32: according to the first comparison result and threshold value, judge that whether target to be verified is by authentication by processing unit.
With reference to Fig. 8, it is the process flow diagram of image of the comparison specific region scope of the auth method based on biological characteristic of the present invention.As shown in the figure, the auth method based on biological characteristic of the present invention carries out, in the step of pattern comparison, also comprising the following step in processing unit:
S40: utilize processing unit according to contraposition information contraposition the first biometric image and the second biometric image;
S41: identify the singular point (Singular Point) of the second biometric image in feature extraction unit according to Second Characteristic dot information; And
S42: utilize processing unit to launch specific region scope to surrounding centered by singular point, and compare the first biometric image and the second biometric image within the scope of specific region in the mode of pattern comparison, and obtain the second comparison result.
With reference to Fig. 9, it is the process flow diagram that carries out pattern comparison in contract drawing mode of the auth method based on biological characteristic of the present invention.As shown in the figure, the auth method based on biological characteristic of the present invention carries out, in the step of pattern comparison, also comprising the following step at processing unit:
S50: utilize processing unit according to contraposition information contraposition the first biometric image and the second biometric image; And
S51: the group that carries out the first biometric image and the second biometric image by the mode of contract drawing at processing unit compares, and obtains the second comparison result.
The detailed embodiment of each above-mentioned step had described in detail in the narration of the authentication means based on biological characteristic above, at this, just repeated no more.
In sum, authentication means based on biological characteristic and method thereof proposed by the invention, can be by the less characteristic point information of comparison data volume to draw contraposition information, and carry out the contraposition of image according to this contraposition information, solved thus the problem consuming time that in prior art, two image contrapositions cause; No matter compare again the image within the scope of specific region or carry out pattern comparison in the mode of contract drawing, all can improve the speed of comparison, lowering the resource consumption of system computing, and reaching the object of saving storage area.
In addition, according to adjusting threshold value with the drawn comparison result of pattern Comparison Method, can avoid single Comparison Method to cause the problem of identity None-identified, especially for user, be accustomed to changing or environment change, cause extracted unique point very few and be difficult to the problems such as checking, more can be smoothly through authentication by authentication means and the method thereof based on biological characteristic of the present invention, and can reach high accuracy rate.
It is worth mentioning that, the present inventor also pays close attention to security and the convenience issue of authentication.Generally speaking, those skilled in the art with false rejection rate (FRR, False Rejection Rate) and wrong acceptance rate (FAR, False Acceptance Rate) the recognition capability index as recognition system.False rejection rate (FRR) representative should be identified but not passing through of passing through, and another kind of wrong acceptance rate (FAR) representative can not but having passed through by identification.Prior art is the standard of only whether passing through as identity by single Comparison Method and single threshold value, sometimes in order to consider user's convenience, just reduce threshold value, though can allow user more easily by checking, increase also wrong acceptance rate (FAR), allow also passing through of should not passing through, cause on the contrary the leak in safety; On the other hand, region for height control, for considering its security, just improve threshold value, though can take security into account, easily allow and allow the user who passes through to have the higher chance that cannot pass through checking, i.e. higher false rejection rate (FRR), cause on the contrary user's inconvenience, hence one can see that, and prior art is difficult to take into account security and convenience.
Yet, according to the authentication means based on biological characteristic proposed by the invention and method thereof, can adjust threshold value by the drawn comparison result of pattern Comparison Method, the comparison result drawn when pattern Comparison Method is high, reduces the threshold value of unique point Comparison Method; Otherwise, improve the threshold value of unique point Comparison Method, and rule of thumb rule is dynamically adjusted this threshold value, make to take into account false rejection rate (FRR) and wrong acceptance rate (FAR).Thus, authentication means and method thereof based on biological characteristic of the present invention not only can improve the security of authentication, more can take into account user's ease of use, overcome prior art and only can take false rejection rate (FRR) or wrong acceptance rate (FAR) into account, and then reached the object of simultaneously improving false rejection rate (FRR) and wrong acceptance rate (FAR).
The foregoing is only illustrative, but not be restricted person.Anyly do not depart from spirit of the present invention and category, and the equivalent modifications that it is carried out or change all should be contained in claim scope.
Claims (6)
1. the auth method based on biological characteristic, described method comprises the following step:
At storage unit stores the first biometric image and corresponding First Characteristic dot information;
Utilize Extraction of Image unit to extract the second biometric image to target to be verified;
A plurality of unique points the correspondence of by biological characteristic extraction unit, analyzing this second biometric image provide Second Characteristic dot information;
By processing unit, compare this First Characteristic dot information and this Second Characteristic dot information, to draw the first comparison result and contraposition information, wherein this processing unit is to draw this first comparison result according to the similarity of this Second Characteristic dot information and this First Characteristic dot information;
The contraposition of carrying out this first biometric image and this second biometric image according to this contraposition information at this processing unit is carried out the similarity of both patterns again and is compared, to draw the second comparison result, this processing unit calculates threshold value according to this second comparison result by expression formula, wherein, the calculating of this expression formula refers to this second comparison result by computing to draw threshold value, wherein, this expression formula makes the similarity corresponding with this second comparison result and this threshold value be negative correlation; And
Utilize this processing unit to judge that according to this first comparison result and this threshold value whether this target to be verified is by authentication.
2. the auth method based on biological characteristic as claimed in claim 1, wherein, carries out, in the step of pattern comparison, also comprising at this processing unit:
Utilize this processing unit according to this this first biometric image of contraposition information contraposition and this second biometric image;
In this feature extraction unit, according to this Second Characteristic dot information, identify the singular point of this second biometric image; And
Utilize this processing unit to surrounding, to launch specific region scope centered by this singular point, and compare this first biometric image and this second biometric image within the scope of this specific region in the mode of pattern comparison, and obtain this second comparison result.
3. the auth method based on biological characteristic as described in claim 1, wherein, carries out, in the step of pattern comparison, also comprising at this processing unit:
Utilize this processing unit according to this this first biometric image of contraposition information contraposition and this second biometric image; And
The pattern that carries out this first biometric image and this second biometric image in the mode of contract drawing by this processing unit is compared, and obtains this second comparison result.
4. the authentication means based on biological characteristic, comprises:
Storage element is to store the first biometric image and corresponding First Characteristic dot information;
Extraction of Image unit, extracts the second biometric image to target to be verified;
Biological characteristic extraction unit, a plurality of unique points the correspondence of analyzing this second biometric image provide Second Characteristic dot information; And
Processing unit, connect this storage element, this Extraction of Image unit and this biological characteristic extraction unit, compare this First Characteristic dot information and this Second Characteristic dot information, to draw the first comparison result and contraposition information, wherein this processing unit is to draw this first comparison result according to the similarity of this Second Characteristic dot information and this First Characteristic dot information, the contraposition that this processing unit carries out this first biometric image and this second biometric image according to this contraposition information is carried out the similarity of both patterns again and is compared, to draw the second comparison result, and by expression formula, calculate threshold value according to this second comparison result, wherein the calculating of this expression formula be by this second comparison result by computing to draw this threshold value, and this expression formula makes the similarity corresponding with this second comparison result and this threshold value be negative correlation, this processing unit judges that according to this first comparison result and this threshold value whether this target to be verified is by authentication.
5. the authentication means based on biological characteristic as claimed in claim 4, wherein, this processing unit is according to this this first biometric image of contraposition information contraposition and this second biometric image, this biological characteristic extraction unit more identifies the singular point of this second biometric image according to this Second Characteristic dot information, this processing unit launches specific region scope to surrounding centered by this singular point, and compare this first biometric image and this second biometric image within the scope of this specific region in the mode of pattern comparison, and obtain this second comparison result.
6. the authentication means based on biological characteristic as claimed in claim 4, wherein, this processing unit is according to this this first biometric image of contraposition information contraposition and this second biometric image, and compare with the pattern that the mode of contract drawing is carried out this first biometric image and this second biometric image, and obtain this second comparison result.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| TW099141103 | 2010-11-26 | ||
| TW99141103A TWI437501B (en) | 2010-11-26 | 2010-11-26 | Identity verification apparatus and method thereof based on biometric features |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102479328A CN102479328A (en) | 2012-05-30 |
| CN102479328B true CN102479328B (en) | 2014-11-05 |
Family
ID=46091965
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110141655.7A Active CN102479328B (en) | 2010-11-26 | 2011-05-19 | Identity verification device and method based on biological characteristics |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN102479328B (en) |
| TW (1) | TWI437501B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105205439B (en) * | 2015-02-13 | 2017-05-03 | 比亚迪股份有限公司 | Method for calculating area of fingerprint overlapping region and electronic device |
| CN106156784B (en) * | 2015-03-30 | 2019-10-29 | 联想(北京)有限公司 | A kind of characteristic recognition method and electronic equipment |
| CN106156234A (en) * | 2015-04-24 | 2016-11-23 | 上海箩箕技术有限公司 | Biological information recognition method, identification device and smart lock |
| CN105069343A (en) * | 2015-08-26 | 2015-11-18 | 宇龙计算机通信科技(深圳)有限公司 | Fingerprint verification method, fingerprint verification apparatus and terminal |
| CN108491707B (en) * | 2016-05-30 | 2020-01-14 | Oppo广东移动通信有限公司 | Unlocking control method, terminal equipment and medium product |
| CN106250740A (en) * | 2016-07-19 | 2016-12-21 | 青岛海信移动通信技术股份有限公司 | A kind of fingerprint identification method and device |
| CN107609461A (en) * | 2017-07-19 | 2018-01-19 | 阿里巴巴集团控股有限公司 | The training method of model, the determination method, apparatus of data similarity and equipment |
| CN108140120A (en) * | 2017-12-08 | 2018-06-08 | 深圳市汇顶科技股份有限公司 | Biological characteristic detection method and identification system |
| CN110362977B (en) * | 2018-04-10 | 2022-03-15 | 义隆电子股份有限公司 | Biological feature identification method and electronic device with biological feature identification function |
| CN110276190B (en) * | 2019-06-28 | 2021-06-15 | Oppo广东移动通信有限公司 | Fingerprint identification method and related product |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101329727A (en) * | 2008-06-27 | 2008-12-24 | 哈尔滨工业大学 | Fingerprint identification method combining point with line |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100347719C (en) * | 2004-07-15 | 2007-11-07 | 清华大学 | Fingerprint identification method based on density chart model |
-
2010
- 2010-11-26 TW TW99141103A patent/TWI437501B/en not_active IP Right Cessation
-
2011
- 2011-05-19 CN CN201110141655.7A patent/CN102479328B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101329727A (en) * | 2008-06-27 | 2008-12-24 | 哈尔滨工业大学 | Fingerprint identification method combining point with line |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102479328A (en) | 2012-05-30 |
| TW201222433A (en) | 2012-06-01 |
| TWI437501B (en) | 2014-05-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102479328B (en) | Identity verification device and method based on biological characteristics | |
| Weaver | Biometric authentication | |
| CN105981045B (en) | Use the finger print identifying of touch sensor data | |
| JP3356144B2 (en) | User authentication device using biometrics and user authentication method used therefor | |
| TWI599964B (en) | Finger vein recognition system and method | |
| CN101901336A (en) | Fingerprint and finger vein bimodal recognition decision level fusion method | |
| EP1804193A3 (en) | Biometric information verifying apparatus | |
| Shoewu et al. | Biometric-based attendance system: LASU Epe campus as case study | |
| US10115249B2 (en) | Card-compatible biometric access control system | |
| KR20200137450A (en) | Artificial fingerprint detection system and artificial fingerprint detection method using the same | |
| CN105138959A (en) | Image processing based fingerprint matching and control method | |
| Sharma et al. | Role of biometric technology over advanced security and protection in auto teller machine transaction | |
| CN105069428B (en) | A kind of multi-template iris identification method and device based on similarity principle | |
| CN100365646C (en) | Differential operation based high-precision palm print recognition method | |
| Manivannan et al. | Fingerprint biometric for identity management | |
| Patel et al. | Employee Attendance Management System Using Fingerprint Recognition | |
| Barham et al. | Fingerprint recognition using MATLAB | |
| Sharma et al. | Fingerprint matching Using Minutiae Extraction Techniques | |
| CN108416882A (en) | Portable fingerprint identification device | |
| Pankanti et al. | Biometrics: Promising frontiers for emerging identification market | |
| Tiwari et al. | An efficient fingerprint matching using continuous minutiae template learning | |
| KR20050014406A (en) | System And Method For Identifying Fingerprint In Mobile Communication Terminal | |
| KR20040105065A (en) | Strong correlated or weakly correlated grouping fingerprint matching method | |
| Yao | Digital Fingerprint Quality Assessment | |
| OLUSEGUN | BIOMETRICS IDENTIFICATION SYSTEM FOR STAKEHOLDERS (A CASE STUDY OF MOUNTAIN TOP UNIVERSITY) |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |