CN101425127A - Electronic encode for individual one of the twelve animals and identity confirming technology thereof - Google Patents
Electronic encode for individual one of the twelve animals and identity confirming technology thereof Download PDFInfo
- Publication number
- CN101425127A CN101425127A CNA2007100189901A CN200710018990A CN101425127A CN 101425127 A CN101425127 A CN 101425127A CN A2007100189901 A CNA2007100189901 A CN A2007100189901A CN 200710018990 A CN200710018990 A CN 200710018990A CN 101425127 A CN101425127 A CN 101425127A
- Authority
- CN
- China
- Prior art keywords
- looks
- appearance
- person
- data
- identity
- 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.)
- Pending
Links
Images
Landscapes
- Collating Specific Patterns (AREA)
Abstract
The invention relates to a technique which can determine the identity of a person according to appearance characteristic of the person orderly by means of encoding and decoding through an electronic data measure. The appearance is the innate characteristic of a person and changes little along with the change of age and life experience. The appearance which can be used as an identity characteristic key of a person comprises the face, fingerprints, retinas and the like of the person. The technique comprises an appearance sensor, a data collecting sensor, a data encoder, an appearance database, a data analytic algorithm. Characteristics collection of a contact type and a non-contact type are adopted according to the difference of the appearances. The sensor for collecting the appearance can adopt a digital camera (a CCD or a CMOS) or a special living body sensor. According to the collected appearance data, calculation (the invention provides three algorithms) with data pre-registered in the appearance database is performed so that whether the identity of the person is legal can be determined. The appearance encoding is innate, without human encoding or memorization, and codes exist together with the person, can be used at any time, more importantly is unique and cannot be decoded or falsified. The invention can be used in any places and occasions needing identity identification, such as banks, stock exchanges, working attendance check, entrance guard, census register, admission cards for entrance examination, purchase cards, and the like. The product of the invention can effectively enhance the uniqueness of the identity identification of the person and can assist relative sections to uncover illegal identity cases.
Description
One, technical field
The invention belongs to complex art fields such as information coding theory, electronic technology, (image) information Recognition.Can be applicable to that securities are cashed or the identity validation during depositing and drawing in bank, be applied to that come in and go out in important place, the pass purposes when classified papers are had access to.The present invention is mainly by looks sensor, data acquisition unit, compositions such as information Recognition software, appearance database.After adopting the looks electronic code and affirmation technology thereof of this personal identification, the deficiency of existing less figure place numerical ciphers will be remedied, strengthen the legal identity discrimination greatly, prevent that effectively the lawless person from assuming another's name to convert certificate, withdraw the money or passing through, and strengthened individual and the property of country and the security of confidential information.
Two, background technology
The place that often accesses to your password in the daily life mainly is a bank, in addition also has security trading place, important place, secret strategic point or the like.The purpose that password is set is in order to protect individual and national wealth not to suffer a loss, to protect valuable equipment, state secret or technology secret not to be leaked.
The password of existing bank card is 6 bit digital, and this has considered roughly to distinguish the user and has remembered easily again and be provided with.This sign indicating number since figure place cause its maximum weakness than short-range missile and decoded easily, thereby with it during as identity validation the degree of reliability not high, very easily cause user's property loss.Increasing Password Length has increased the difficulty of memory again.
There are the problems referred to above equally in other occasion that needs identification cipher to confirm.
In order to address the above problem, must seek a kind of length long enough, easy coding, easy memory, easy-to-use coded system.For this reason, the present invention proposes to utilize the scheme of individual looks feature as coding.
Individual's looks is meant the inherent exclusive physiological characteristic of people, and is very little with the variation of age and experience of life.The looks that can be used as personal identification feature password comprises individual face phase, fingerprint and eyes retina or the like.This looks coding has day natural disposition, do not need artificial coding, need not remember, sign indicating number is with People's Bank of China, have uniqueness with getting with usefulness, the more important thing is, is difficult to decode and forge.
Utilize the coding of looks as the personal identification feature, can not direct input as existing numerical ciphers, must adopt modern electronic technology that individual looks data are gathered and stored.Thereby this scheme relates to modern electronic technology.
Whether identical for the looks data of confirming just to have gathered with the looks data picture of storage in advance, can not as 6 existing bit digital passwords, carry out equating one by one coupling, must adopt Technology of Data Fitting or similarity technology.Thereby this scheme relates to information discriminating technology.
Three, summary of the invention
1. the information source code that can not decode and can not forge
Investigation and statistics show common people one people one phase! Promptly almost there are not mutually identical two people of face.This is because face depends on gene and posteriori nutrition, activity and living environment mutually.Only with regard to gene, there are not identical two people of gene.Even father and son, father and daughter, mothers and sons, mother and daughter, or multiparity siblings, or throwback direct line (collateral line) relatives, also just picture or the spitting image of, but never be theoretic striking resemblances.As for the people who does not have sibship, alike degree is just lower.
Secondly the sign that has uniqueness to everybody is exactly the retina of eye, and situation is mutually suitable with face.
Another sign that has uniqueness to everybody is exactly a fingerprint.Research indicates that any two people's fingerprint difference degree is also very big.If the difference of any two people's same finger print is identical with the difference degree of face phase, then the difference that combines of ten fingers just has been tending towards infinitely.Change an angle, if the close degree of any two people's same fingerprint is a, then 10 common close degree in fingerprint combination back are a just
10Go bail for and keep data a=10
-8Total then the degree of closeness of 10 fingerprints is exactly 10
-80In fact, as long as everyone is got 1 fingerprint just can tell 100,000,000 people, get 2 fingerprints and just can tell 10
16The people, and be unique.
The property decoded of this sign indicating number is based upon on the forgery property.This sign indicating number can not to guess forgery property be conspicuous.Make silicon rubber mould then if taking pains takes pictures to someone looks,, perhaps can decode its coding, but difficulty is very big from the looks that the input of looks sensor is forged.Existing a kind of now live body sensor comes out, and will penetrate the true man of the mould or the live body of dead volume to a great extent.Thereby the looks of live body will be forged, thereby can't decode based on the coding of looks source code.
2. looks information electronic code/data acquisition
Looks information is nature, have the three-dimensional colour object of countless fine features, the object that can roughly discern of consciousness only, rather than calculates the information of function Direct Recognition.Looks information must be converted into the information of calculating function identification for this reason.
The most direct way of gathering looks information is the plane figure that needed looks is converted into digital format, and this is most basic coding.This perspective view can have three kinds of patterns: cromogram, artwork master, binary map.Coding and identification are simple also to keep source information as far as possible in order to make, and the present invention advises adopting artwork master or binary map.
(1) gray level image and coding thereof: suppose that a width of cloth looks figure is made up of m * n pixel, each pixel has k level gray scale.Then this width of cloth image can accurately be distinguished A * (m * n) at least
kObject.If these pixels are arranged according to natural order, A=1 is arranged then.If the method that adopts other is to these pixel codings, then A depends on the kind tree of coding.In the present invention, take from right sequential encoding (A=1), m * n=160 * 120, k=256 (i.e. the gray scale of a byte) is enough to satisfy the requirement of distinguishing personal identification, and the number that will hold on this coding theory this moment is (256)
160 * 120≈ 10
46238Individual.
In order to take into account the relation of memory space, operand and difference rate, can suitably increase and decrease number of picture elements or gray shade scale.
(2) bianry image and coding thereof: the simplest a kind of picture format is a bianry image, and promptly the gray shade scale of image has only secondary, and is bright with 1 representative, and dark with 0 representative, the pixel of piece image is exactly a binary sequence like this.If the looks image of being gathered has m * n pixel, then the binary sequence of this pixel can distinguish 2
M * nIndividual's looks information.If now get m * n=100 * 100, such binary code sequence can hold 2 in theory
100 * 100≈ 10
3010The people.
(3) optimize coding: looks information coding is in order to distinguish personal identification, needn't to pursue the true to nature of image.Thereby be enough to reach this target from above-mentioned analysis bianry image.Above-mentioned data are theoretic value, and in theory, distinguishing global people (10,000,000,000) also only needs 10
10The two-value pixel.But the same looks of considering any two people all has a lot of similaritys, still needs abundant pixel just can distinguish it, and this need optimize between theoretical value and actual value a little.For the redundant data of the minimizing looks information of maximum possible, can in the large image element image, choose several view of cells pictures and be encoded in addition, reduce number of picture elements on the one hand, not reduce the differentiation rate on the other hand with feature.
(4) preferred looks/data acquisition modes: in above-mentioned several looks of mentioning, fingerprint and retina are very little with the variation in years, and the face phase change is bigger, and gather at every turn face phase time eyes open with close, face Zhang Yuhe or the like may be different with the status data of gathering in advance, the influence decoding.Thereby preferably fingerprint and retina as the looks object.Image and data collector has two kinds of working methods: contact and contactless.The appropriate to the occasion corresponding fingerprint of contact, contactless suitable corresponding retina.Also can adopt contactless sampling to fingerprint.Contactlessly can not leave a trace, thereby can not stay the leeway of stealing the looks vestige the lawless person.Thereby preferred contactless sampling.
3. looks information Recognition/identity validation
Utilizing looks information to carry out identity validation at first needs to need the people's of identification looks information to gather and register to those, promptly forms an appearance database.Carry out the place of identity validation or constantly at needs, gather this person's looks information and compare with the corresponding informance of database registration, whether legal with decision this person's identity.
The method of identification following (is example with the bi-level image coding):
(1) equate judgement method (also can be called template matching method) fully: whether the data that promptly contrast collection in worksite by turn equate with the corresponding data of database, if have at least one unequal, then identity can not obtain confirming, has only bits per inch according to all equating, could pass through identity validation.
(2) approximation ratio operation method: equate that fully the judgement method is too harsh, because the ductility of looks object, equal not to the utmost fully to the each data of gathering of same individual, this will take the computing of degree of approximation.Information encoded binary sequence is thought equal interval sampling to a two-dimensional curve, can calculate the degree of approximation degree of these two curves according to certain rule.Obtain a result according to predetermined threshold value then.If calculate degree of approximation greater than this threshold value, then by authentication, otherwise can not pass through.
(3) regional area fault tolerant approach: the two-dimensional data array that m * n pixel collecting and the corresponding data in the database are formed m * n respectively, each corresponding element in two arrays is carried out the logical difference exclusive disjunction one time, form a new m * n data array with its result, obviously being not equal to 0 element in this array is exactly the difference images of this two width of cloth binary picture.If p for the piece number that is not equal to 0 data in this data array, q in all pieces not being the maximum number of 0 data.For carrying out identity validation, be necessary for p, q setting threshold p
0, q
0If p and q have one at least greater than corresponding threshold value, then can not be by the identity identification.
The application system principle of composition of the present invention as shown in drawings.
Four, description of drawings
This figure is encoded to example explanation the present invention with binary image.
The looks sensor is the parts that looks carried out basic coding, can adopt electronics (CCD or CMOS) camera, if The print looks is optional capacitive fingerprint sensing device then, generally gets number of picture elements and is 100 * 100 and be advisable.
Data acquisition unit be basic coding is become computer (computer can be embedded or any SCM system) herein can The parts of the data of identification can be selected general electronic device.
The binaryzation encoder is with Binary Sketch of Grey Scale Image, and the simplest directly is the natural order coding, also can add other volume The code rule. If the bianry image to 100 * 100 pixels is encoded by natural order, then can distinguish 2100×100≈10
3010Individual user.
Looks registration/appearance database is the looks of service-user to be encoded be registered in the database, in order to later on its identity is carried out Confirm.
Data-analyzing machine utilize recognizer (equate fully one of judgement method, approximation ratio operation method, regional area fault tolerant approach or Other algorithm) user's to be identified data is adjudicated, determine whether the user has legal identity. If user identity is legal, Then can continue to handle follow-up business.
Five, embodiment
The following print phase object of making a living is the specific embodiments that example illustrates this invention with the identity validation of bank-user.
(1) fingerprint register: when the user opens an account with the bank, the fingerprint capturer of bank, is associated this coding through inner binaryzation and suitable coding to user's fingerprint sampling with user's card number, 6 bit digital passwords, deposit database in, finish the user fingerprints identification number register.
When (2) user handled the deposit or the withdrawal in bank, after having imported 6 correct bit digital passwords, finger-print cipher also will be gathered to the user by bank.
(3) coding be scheduled to of the finger print data that will just gather of banking computer, then with database in this user's finger-print codes carry out computing, if operation result meets by condition, then user identity obtains confirming, can continue to handle the deposit or the withdrawal.Otherwise stop to handle.
When (4) this invention is used for ATM, need the networking of ATM and bank to carry out reading of fingerprint database.
Other need the industry of identity validation or password, unit can be with reference to the corresponding fingerprint recognition system of the present embodiment manufacturing, as electronic attendance, gate inhibition, door lock, household register, admission card for entrance examination, purchase card etc.
Claims (3)
1. an electronic code that utilizes individual looks carries out the technology of identity validation, and looks refers to individual face phase, fingerprint, and retina etc., this technical characterictic partly comprise looks sensor, coding rule, data analysis algorithm etc.
The looks sensor can adopt electronics (CCD or CMOS) camera, or the capacitance type fingerprint sensor.
2. individual looks electronic code identity validation technology according to claim 1, comprise a kind of coding rule for the looks data, it is characterized in that: this rule adopts the binaryzation two-dimensional image data of looks as basic coding, the number of picture elements of image is advisable with 100 * 100, and the natural order coding of this image is adopted in in-line coding.
3. individual looks electronic code identity validation technology according to claim 1 comprises looks code identification algorithm, and this invention proposes three kinds of algorithms, it is characterized in that: equate judgement method, approximation ratio operation method, regional area fault tolerant approach fully.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007100189901A CN101425127A (en) | 2007-11-02 | 2007-11-02 | Electronic encode for individual one of the twelve animals and identity confirming technology thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007100189901A CN101425127A (en) | 2007-11-02 | 2007-11-02 | Electronic encode for individual one of the twelve animals and identity confirming technology thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101425127A true CN101425127A (en) | 2009-05-06 |
Family
ID=40615736
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2007100189901A Pending CN101425127A (en) | 2007-11-02 | 2007-11-02 | Electronic encode for individual one of the twelve animals and identity confirming technology thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101425127A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105389764A (en) * | 2015-11-25 | 2016-03-09 | 安徽瑞信软件有限公司 | Household registration identification management system |
| CN109255537A (en) * | 2018-09-10 | 2019-01-22 | 珠海格力电器股份有限公司 | A kind of labour discipline attendance management method and system |
| CN111401315A (en) * | 2020-04-10 | 2020-07-10 | 浙江大华技术股份有限公司 | Face recognition method, recognition device and storage device based on video |
-
2007
- 2007-11-02 CN CNA2007100189901A patent/CN101425127A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105389764A (en) * | 2015-11-25 | 2016-03-09 | 安徽瑞信软件有限公司 | Household registration identification management system |
| CN109255537A (en) * | 2018-09-10 | 2019-01-22 | 珠海格力电器股份有限公司 | A kind of labour discipline attendance management method and system |
| CN111401315A (en) * | 2020-04-10 | 2020-07-10 | 浙江大华技术股份有限公司 | Face recognition method, recognition device and storage device based on video |
| CN111401315B (en) * | 2020-04-10 | 2023-08-22 | 浙江大华技术股份有限公司 | Face recognition method based on video, recognition device and storage device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7172115B2 (en) | Biometric identification system | |
| Bolle et al. | Biometric perils and patches | |
| US6719200B1 (en) | Checking of right to access | |
| CN202257700U (en) | Multi-authentication mode access control system | |
| Elhoseny et al. | Multimodal biometric personal identification and verification | |
| CN1164712A (en) | Biometric identification process and system | |
| CN101145261A (en) | ATM system automatic recognition device | |
| CN1168051C (en) | Checking of right to access | |
| US20070075130A1 (en) | Mid-Level Local Biometric Identification Credit Card Security System | |
| Ambadiyil et al. | Biometric based unique ID generation and one to one verification for security documents | |
| EP2186040A1 (en) | Method for transforming a feature vector | |
| Kaur | A study of biometric identification and verification system | |
| Ayoup et al. | Cancellable Multi-Biometric Template Generation Based on Arnold Cat Map and Aliasing. | |
| CN101425127A (en) | Electronic encode for individual one of the twelve animals and identity confirming technology thereof | |
| Sharma et al. | Encryption of text using fingerprints as input to various algorithms | |
| Mishra | AAdhar based smartcard system for security management in South Asia | |
| Patil et al. | ATM Transaction Using Biometric Fingerprint Technology | |
| JPS61199162A (en) | Personal identification system | |
| Divan et al. | A fingerprint matching technique using minutiae based algorithm for voting system: A survey | |
| Praseetha et al. | Novel Web Service Based Fingerprint Identification Using Steganography and Xml Mining | |
| KORICHI | Biometrics and Information Security for a Secure Person Identi cation | |
| Kumari et al. | A novel approach for secure multimodal biometric system using multiple biometric traits | |
| Sahana et al. | Multi Biometric Recognition System | |
| Hassan et al. | Smart Lock using Image Recognition | |
| Mishra | Importance of AADHAR-Based Smartcard System’s Implementation in Developing Countries |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20090506 |