WO2008099041A1 - Procédé et système de reconnaissance de personnes par analyse biométrique de mains - Google Patents

Procédé et système de reconnaissance de personnes par analyse biométrique de mains Download PDF

Info

Publication number
WO2008099041A1
WO2008099041A1 PCT/ES2008/000080 ES2008000080W WO2008099041A1 WO 2008099041 A1 WO2008099041 A1 WO 2008099041A1 ES 2008000080 W ES2008000080 W ES 2008000080W WO 2008099041 A1 WO2008099041 A1 WO 2008099041A1
Authority
WO
WIPO (PCT)
Prior art keywords
hands
user
hand
database
points
Prior art date
Application number
PCT/ES2008/000080
Other languages
English (en)
Spanish (es)
Inventor
Antonio Adan Oliver
Miguel Adan Oliver
Original Assignee
Universidad De Castilla-La Mancha
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Universidad De Castilla-La Mancha filed Critical Universidad De Castilla-La Mancha
Publication of WO2008099041A1 publication Critical patent/WO2008099041A1/fr

Links

Classifications

    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
    • G07C9/00—Individual registration on entry or exit
    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
    • G07C9/00—Individual registration on entry or exit
    • G07C9/30—Individual registration on entry or exit not involving the use of a pass
    • G07C9/38—Individual registration on entry or exit not involving the use of a pass with central registration
    • G—PHYSICS
    • G07—CHECKING-DEVICES
    • G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
    • G07C9/00—Individual registration on entry or exit
    • G07C9/30—Individual registration on entry or exit not involving the use of a pass
    • G07C9/32—Individual registration on entry or exit not involving the use of a pass in combination with an identity check
    • G07C9/37—Individual registration on entry or exit not involving the use of a pass in combination with an identity check using biometric data, e.g. fingerprints, iris scans or voice recognition

Definitions

  • the present invention falls within the scope of computer vision and specifically in the application of biometric personal recognition systems.
  • the system and method described here is applicable in the control of access of people to public buildings (ministries, airports, etc.), to private homes, companies or organizations, to restricted areas or to reserved services because of their danger, security or Privacy.
  • the invention applies to cases of control of assistance to the workplace in companies or public bodies, where the verification or identification of certain people who are authorized to access these spaces for various reasons (residents, employees, customers) , maintenance personnel, police, firefighters, specialized personnel, etc.) or who have an obligation to access them (employees, officials, students, etc.).
  • this invention can be used to directly verify and ensure with precision the identity of a certain person, whose identifying characteristics are stored on a card with a user key for access to electronic banking services. In this case, a comparison is not made with the rest of the individuals in the database (user identification), but with the stored data that correspond exclusively to the individual whose identity is claimed (user verification). Also, in the sector of computer science and electronic communications,
  • the validation through the hands can be easily applied to provide security with access control to information or use of telematic resources and services.
  • Biometrics is the discipline that is dedicated to the identification of individuals based on anatomical characteristics or behavioral traits.
  • An anatomical characteristic has the quality of being relatively stable over time, while a behavior trait is less stable, since it depends on the psychology of the person.
  • Static biometrics uses techniques based on the anatomy of the person
  • dynamic biometrics uses techniques based on the behavior of the person.
  • a biometric system is an automated system that performs biometrics, that is, the system bases its recognition decisions through a personal characteristic that can be identified or verified in an automated way.
  • biometric technologies based on different characteristics: voice, iris, fingerprint, hand, facial morphology, handwritten signatures, etc.
  • biometric system technologies of user recognition based on their hands are achieving a special interest in recent years in various fields of application such as access to buildings, airports, restricted areas, nuclear plants, stadiums, etc. It is known that one of the advantages of these systems compared to alternative systems - recognition by iris, fingerprints, faces, etc. - is that they are able to maintain an online operation due to the relatively low computational cost of the process. In addition, they are usually systems well accepted by the user and easy to use, implementation and maintenance. Within the process of user recognition, it is necessary to distinguish if it is of interest, depending on the application of the biometric system, perform a user identification procedure or a verification of the same or both processes.
  • the identification of the user who performs a biometric system tries to answer the question: who is the person X ?; that is, obtaining information from a person whose identity is unknown.
  • This type of comparison is called one to many (1: N), since in order to proceed with the identification of a person it is necessary to compare the biometric characteristics of the same with those of all the people stored in a database.
  • the biometric system tries to answer the question: is this person the person X ?, that is, a person claims to have a certain identity and the system must verify that it is true.
  • This type of comparison is called one to one (1: 1), since only the biometric characteristics of a person are compared with those of the person who claims to be.
  • a single hand is always analyzed, indistinctly right or left, using the image of the hand to define a set of characteristics, which may be based on its geometry, on the geometry of the fingers, in the grooves of the palm of the hand, in the contour of the hand, in the knuckles of the fingers, in the three-dimensional geometry, etc.
  • the process of registering the images of a hand is usually performed using devices that have a set of small pivots that control the position of the hand. Once the hand is properly placed, images can be taken of the back (palm), anterior (opposite the palm), or even the edge of the hand, and then perform a feature extraction process.
  • the biometric systems that perform identification of existing hands on the market today are designed to work applied only for one hand, extracting the biometric characteristics directly on the image of the hand, from which a very precise placement is required by user part.
  • this type of systems that carry out the biometric analysis of a single hand (right or left) or of fingers or parts of fingers of the hand to make the recognition (verification or identification) of the user, several unresolved problems can be found until the date: - They do not allow a natural placement of the hand, use predefined positions of the fingers of the hand or do not take into account the way in which the hand is open.
  • the invention described here comes to solve the problem described above, in each and every one of the aspects presented, conceiving a method and system of user recognition that can perform the functions of identification and personal verification in an optimal way, reducing the probabilities of error to the minimum when using images of the two hands of the user, right and left and captured in a way that is easy for the user, to perform a quick biometric analysis while being effective as described below.
  • the method of recognition of people by biometric analysis of hands that constitutes an aspect of the present invention comprises the following five stages: Stage 1): Capture of at least one image of the left hand and the right hand, which collects the entire contour of both hands of a person (which here we can call person entering the system and whose recognition of their identity is required : identification or verification).
  • One or more images can be registered by one or several cameras, usually two -one focused on each hand-, so that both contours and the upper part (opposite the palms) of the two hands are obtained.
  • Stage 2) Digital processing of the images, which in the previous stage can be captured by digital or analog cameras, to extract a one-dimensional function that represents at least a part of the contour of each of the two hands and from which a set of control points that will serve to define a series of biometric characteristics in the next stage.
  • Stage 3) Extraction of at least one set of biometric characteristics defined on the function and that make up a pattern of each hand, represented in a coordinate or image reference system that is typical of the image of the hand and, therefore, These characteristics are independent of the position of the hands when placed naturally when capturing images.
  • Stage 4 Comparison between pairs of hands, of the person of entry and some user people (those who have been previously recognized by the system and of which they have the patterns of their hands stored in a database or knowledge base of the system).
  • the user recognition is carried out by comparing the set of biometric characteristics of each of the two hands of the input person and with the patterns of the database of both hands of a user (verification) or of all the users (identification) of the system.
  • Stage 5 Generation of a series of output signals that is determined by the result of user recognition. Next, the steps involved in the development of each of these stages are explained in more detail.
  • Image capture or capture stage The acquisition of the images corresponding to each of the hands (left and right) of the user is performed in a capture system that preferably comprises two black and white cameras that focus from below a area in which the person must place their hands so that the anterior (upper) part of them is photographed.
  • a capture system that preferably comprises two black and white cameras that focus from below a area in which the person must place their hands so that the anterior (upper) part of them is photographed.
  • the result of this stage is to extract the necessary information from the images of the hands taken in the image taking stage, based on it to be able to define a set of biometric characteristics in The next stage.
  • the images from stage 1) captured by the cameras are converted to a digital format to be processed on a computer.
  • a binarization of the image is performed through thresholding of the histogram converting the intensity images (in shades of gray) to binary images.
  • related regions are identified in the binary image and the image is filtered taking the region of greater area. After these phases, the image obtained corresponds to the image of the filtered noise hand.
  • the Natural Reference System is defined as a coordinate system with respect to two perpendicular axes that intersect at an origin point, which are constructed from the positions of the middle and thumb fingers of a fully extended hand.
  • the Y axis or axis Vertical Natural Reference System is defined through the skeleton of the middle finger.
  • the X or horizontal axis is defined as the axis perpendicular to the Y axis passing through the point of the thumb furthest from said Y axis.
  • the center or origin of the Natural Reference System is defined as the cut of X and Y.
  • the I chain is resampled to an established number of equidistant pixels, obtaining a normalized polar representation of the contour of the hand, invariant before transformations (translation and rotation), since it is defined on an intrinsic reference system to the user, which is their own Natural Reference System of each of their hands.
  • the polar representation of the normalized contour provides functions of modules f and arguments g, from which you can define a set of characteristics that have no meaning extrapolated in the image although some may correspond approximately to classical geometric measurements (height or width of the fingers) extracted on the image itself.
  • Such control points are transferred from the domain of the modules to that of the image of the hand.
  • Stage of extraction of biometric characteristics On the control points transferred to the image, at least one set or vector of biometric characteristics is defined, established as a result of a correlation study on a large number of hand characteristics in the that correlated characteristics have been ruled out. In this way, the number of components is minimized by eliminating redundant information.
  • the feature vector can be divided into four groups, each with its own meaning within the type of measure that is established and are the following: - Set defined by the respective distances between some Points
  • nir a set of descriptors that constitute patterns that characterize each hand, avoiding segmentation processes in the image, it is possible to obtain invariant information of the hand with hardly any computational cost. This is achieved through the polar representation of the contour of the hand in the reference system proper to said hand and means moving away from the expensive techniques based on the extraction of features through image processing.
  • User recognition stage verification and / or identification: The set of characteristics obtained in the previous stage is used to perform the recognition operations required by the specific application of this system.
  • a distance is obtained between the data obtained for the user who claims an identity and a part or all of the data collected in a knowledge bank or database within the system, producing as output an distance vector of dimension one for the case of verification or a vector of distance of dimension equal to the number of individuals in the database for the case of identification.
  • This database can be seen essentially as a collection of patterns (that is, sets of descriptor vectors corresponding to the individuals to be recognized) normally arranged in two code books: one for the parameters of the right hand and the other of the hand left. Several patterns corresponding to a person are stored in said database. The comparison between hands is carried out through the similarity measure based on a set of distances between both types of hands (right and left).
  • m 14 1 Hot distances between features of the hands are taken.
  • a one-to-one comparison is made between the input sample and the prototype stored in the database.
  • the verification processes compare the input (or test) hand with the hand prototypes through feature vectors or templates stored in a database.
  • Conventional verification systems use one-hand geometric parameters. Obviously, if both hands are considered, as the present invention does, there is more data of the same individual and, therefore, the possibility of increasing the effectiveness of the verification system. That is, it is possible to verify that some R / L hand or both hands belong to a specific individual. With this, obviously, the skill in the verification task increases.
  • G nk is the sum of normalized values between [0,1]
  • a iqk min ⁇ d ⁇ (M q , M J k) ⁇ is the minimum ith distance for the samples
  • the decision in the verification process depends on whether the Standardized Similarity Measure 9 ' k ' exceeds or does not exceed an empirically threshold prefixed. Therefore, the threshold value is dependent on the actual updated database.
  • ⁇ R M L is t ⁇ ⁇ is to biece a decision verification test: acceptable or rejectable.
  • a final result of the positive user verification is determined if an acceptable result is obtained for at least a certain number of comparisons between G (M q , M k ) and the comparison threshold; for example, it is enough that one of these comparisons is acceptable.
  • the proposed verification method applies a fusion algorithm combining information from left and right hands.
  • a fusion algorithm combining information from left and right hands.
  • the characteristics of the right and left hands are concatenated to calculate the expression of Normalized Similarity Measurement.
  • merger is applied at the decision level since the final verification decision is based on the decision on couples ⁇ R, R ⁇ , ⁇ R, L ⁇ , ⁇ L, R ⁇ and ⁇ L, L ⁇ , by that four decision votes are combined, taking into account that for Me ⁇ R, L ⁇ the Normalized Similarity Measures G (Rq 1 Lk) and G (Lq 1 Rk) are calculated for individuals q and k-th according to the cited expression previously.
  • G (Rq 1 Lk) and G (Lq 1 Rk) are calculated for individuals q and k-th according to the cited expression previously.
  • a single pattern or prototype is not used here, but multiple patterns are stored for each user: the feature vectors corresponding to the last correctly verified samples of the user. To positively verify a new case, it is required that there be at least one verification success with any of these prototypes.
  • FRR False Reject Rate
  • the decision threshold is set in advance to modify the behavior of the system, according to the desired criteria, being able to apply two criteria:
  • the objective is to find a decision threshold to achieve the optimal point of the biometric system.
  • EER Equal Error Rate
  • the EER point is usually applied in biometric systems to predetermine such threshold used in user verification.
  • the system must discover the identity of the user that, in principle, is unknown.
  • a comparison of the user characteristics vector is made against all the users stored in the database.
  • an optimal candidate can be determined, that is to say, which is the user among all that best fits the entry person through the Normalized Similarity Measure G (M q , M k ) for each association R / R, R / L, L / L, L / R.
  • the optimal candidate can be considered as the one that minimizes the average of all associations.
  • the final result of the identification is positive if at least an average is obtained between the four results of all the comparisons G (R q , R k ),
  • Output stage A sequence of output signals is generated from the verification and identification results, which can be:
  • a visual and / or acoustic output signal that alerts the entry person and access control personnel about the result of the verification or identification.
  • Said result (positive or negative) can be displayed in a window on a display screen.
  • At least one internal signal that transmits information related to the recognition result together with the time of the registration, in order to check the time at which a certain person arrives or leaves a certain area.
  • This information is generally stored in an additional module of the system for subsequent security checks.
  • At least one physical signal of action whose purpose is to be sent to one or more drive elements to activate certain devices external to the system but associated with it (for example, the lock that controls the opening of a door, any member that allows the activation or use of a certain device or service, such as, for example, a terminal of a data processing system, etc.) according to the result of the recognition performed.
  • Another aspect of the invention relates to a system of recognition of people who performs the method described for the biometric analysis of both hands, comprising the following elements: - At least one camera for image acquisition according to stage 1) of the method defined above, in whose visual field the person of entry or user must place both hands as explained in said stage 1).
  • Diffuse lighting means to illuminate the visual field of said cameras.
  • Electrical means to power the cameras and associated diffused lighting means.
  • An enclosure or anti-reflective interior box that surrounds the cameras, equipped with a face that has a transparent area, for example where a glass platform or similar is located where the hands are to be placed for the taking of images, in correspondence with the visual field of the cameras.
  • - Complementing the anti-reflective wrap it has a panel that also prevents reflections and provides a background contrasted to the images of the hands.
  • At least one computer or computer processing means that collect the images delivered by the cameras through an interface that digitizes the information. Such computer processing means execute steps 2 to 4 of the method of the invention.
  • At least one output module controlled by said at least one computer and in charge of generating the various output signals described in step 5).
  • the invention has the following advantages as compared to other verification and identification systems:
  • the presented invention has industrial application in the control of assistance and access control to any enclosure, area, building or space through which a set of users must access. It is especially suitable when the following restrictions are imposed: it must be easy to use, it must be comfortable and immediate in use, it is not necessary to get rid of clothing or accessories (watches, rings, bracelets, etc.) or go through a training phase in the system.
  • the system is also applicable in areas with smaller groups of users, such as access to homes with intercom and restricted areas in buildings or factories. In these cases, the False Acceptance Rate must be reduced to the maximum at the expense of raising the Rate of
  • Figure 1. Shows a schematic representation of the system object of the invention.
  • Figure 3. Shows a set of descriptors that determine a pattern corresponding to a polar representation of the contour of a hand in the Natural Reference System defined in the previous figure.
  • Figure 4. Shows a graphic example of a one-dimensional function that corresponds to the polar representation of part of the contour of a hand and a set of control points that characterize
  • Figure 5. Shows the set of control points, belonging to the function drawn in the previous figure, transferred to the image of the hand.
  • Figure 6. Shows in the Natural Reference System, a first sequence of characteristics of the image of the hand determined by a subset of the set of control points represented in Figure 5.
  • Figure 7. Shows in the Natural Reference System, a second sequence of characteristics of the image of the hand determined by another subset of the set of control points represented in Figure 5.
  • Figure 8 shows in the Natural Reference System, a third sequence of characteristics of the image of the hand determined by another possible subset of the set of control points represented in Figure 5.
  • Figure 9 shows in the Natural Reference System, a fourth sequence of characteristics of the image of the hand determined by another possible subset of the set of control points represented in Figure 5.
  • a practical embodiment of the invention can be described as a system for personal recognition by biometric analysis of both hands, comprising the elements represented in Figure 1:
  • the cameras can be analog or digital, preferably CCD cameras, for example, two black and white analog cameras with a focal length of 16 millimeters.
  • Both chambers (1) are surrounded by an anti-reflective interior envelope (5), for example, a black box with rectangular faces with dimensions 37x90x66 centimeters that houses all the lighting tubes and CCD cameras.
  • the two chambers (1) are oriented upwards and in the upper face a transparent platform (3) is fitted, for example of glass, coupled in the upper part of the box but protruding from the outside , intended for the user to place their hands on it and extend them to the greatest tension.
  • the envelope (5) has all its internal black faces to prevent the cameras (1) from capturing lighting reflections from the inside with the glass where the hands are placed or from the outside.
  • a rotating or folding cover of black color and dimensions 40x66 cm, which acts as an anti-reflective panel (4) providing a dark background to the images of the hands
  • the platform (3) of transparent rectangular glass next to the rotating lid are located in the upper part of the box, inside and immediately below the diffused lighting tubes.
  • the chambers (1) are located in the lower part oriented towards the glass, each one registering approximately 60% of the area of the transparent glass.
  • the user To take an image, the user must place their hands apart facing the thumbs - without touching - so that the entire surface of the hand is within the area of the glass.
  • each camera fully captures the image of one hand (right and left) and, perhaps, a part of the other. In the stage of image capture performed by this system, the user must follow a very simple protocol that has these three steps:
  • Image registration is done at two different times. In the first place, it is carried out for new users that are not registered in the database. In this case, 6 to 8 shots are taken per hand and they select the five best ones to subsequently generate their characteristics and store them in a database (B) or knowledge bank. Secondly, a registration is made when a user requires the system to carry out the verification or identification process. In the case of verification, the user must enter his / her personal code, which can be an alias or "login" identifier or their national identity document number. In case of identification, the user must not enter any information to the system, since the system itself is the one that intends to find out its identification.
  • the system has at least one computer (8) with the associated database (B), connected to the cameras (1) through an interface (7) that digitizes the images. If the cameras (1) are analog, you can use external analog / digital converters or digitizing cards inserted in the computer itself (8).
  • the images captured by the system, images of the right and left hands, are digitally processed to finally extract a vector of biometric characteristics that represents the pattern of each hand and has to be compared with the patterns stored in the database (B) .
  • manipulation, improvement and filtering of the original image are carried out, performing the following phases: - Binarization of the image by thresholding the histogram converting the intensity image, into gray levels, to a binary image. I) Labeling of related regions in the binary image. II) Obtaining the binary image of the filtered noise hand through the region labeled with greater area.
  • SRN Natural Reference System
  • Each control point has two coordinates (a, f (a)) in the module domain.
  • the first coordinate (a) corresponds to the index in the normalized chain (I n ) and the second coordinate f (a) corresponds to the value of the module.
  • the first coordinates a of control points in the module domain are recovered in the image of the hand. For simplicity the same nomenclature of the control points in the domains of modules and image is maintained.
  • Figure 5 marks the location of these control points (P 1 , ... P 13 ) on the contour represented by the normalized chain (I n ).
  • Figure 7 Set defined by distances between some Midpoints and distance between some Relative Minimum Points
  • Figure 8 Set defined by distances between some Relative Maximum Points with non-contiguous Relative Minimum Points
  • the verification is carried out by calculating a Standardized Similarity Measure for each and every one of the associations between right and left hands of the input person Rq / Lq, the q-th individual to verify, compared with the hands of each user person Rk / Lk or k-th individual, from the base of
  • G q, k is
  • a itq ⁇ k a ⁇ n ⁇ d i (M q , M J k) ⁇ normalized values between [0,1]
  • j is the minimum ith distance for the five samples of the k-th individual
  • the data collection was carried out during intervals of three months over a period of two years, where the records for each subject were taken over two or three sessions. More than 6000 samples were made by acquiring 6 shots of both hands per user. Among them, a base of prototypes B of 5 samples was formed and with the remaining one a test base for verification test was formed. In total, 2820 verification tests were performed. The overall results show average rates of 1.3% for the value of EER. If low values of
  • the system comprises at least one output module (9) controlled by the computer (8) and configured to generate signals in the last stage of the proposed method.
  • the output module (9) can provide a visual and acoustic alert signal that indicates the result of the verification or identification, showing in a window of a monitor associated with the computer (8) if the verification or identification is correct.
  • said output module (9) additionally provides an ordered list of the users most similar to the input user.
  • an information signal is sent to update the database (B), with the set of patterns formed by five feature vectors ordered by date and structured as a FIFO stack, where the last pattern corresponds to the last case of success in verification or identification.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Collating Specific Patterns (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

L'invention permet la reconnaissance automatique de personnes moyennant des comparaisons croisées de l'image des mains droite et gauche d'un individu selon un processus en cinq étapes : prise d'images des mains séparées, ouvertes en tension, sur une plate-forme adéquate dotée de deux caméras; traitement des images et définition d'un système de référence propre à chaque main; établissement d'une série de descripteurs robustes déterminés par un jeu de caractéristiques non variables impliquant les deux mains; vérification et identification de l'individu au moyen d'une banque de données comprenant de multiples patrons des deux mains de chaque utilisateur; envoi des signaux et de l'information de sortie. Application industrielle directe dans le domaine de la surveillance d'assistance et de l'accès à des zones nécessitant une vérification ou une identification personnelle dans des conditions de brefs temps d'attente.
PCT/ES2008/000080 2007-02-15 2008-02-14 Procédé et système de reconnaissance de personnes par analyse biométrique de mains WO2008099041A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
ESP200700405 2007-02-15
ES200700405A ES2315155B1 (es) 2007-02-15 2007-02-15 Metodo y sistema para reconocimiento de personas por analisis biometrico de manos.

Publications (1)

Publication Number Publication Date
WO2008099041A1 true WO2008099041A1 (fr) 2008-08-21

Family

ID=39689691

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/ES2008/000080 WO2008099041A1 (fr) 2007-02-15 2008-02-14 Procédé et système de reconnaissance de personnes par analyse biométrique de mains

Country Status (2)

Country Link
ES (1) ES2315155B1 (fr)
WO (1) WO2008099041A1 (fr)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040057604A1 (en) * 2002-09-25 2004-03-25 The Hong Kong Polytechnic University Method of palmprint identification

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040057604A1 (en) * 2002-09-25 2004-03-25 The Hong Kong Polytechnic University Method of palmprint identification

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
JAIN A.K. AND ROSS A. ET AL.: "A Prototype Hand Geometry-based Verification System", APPEARED IN PROC. OF 2ND INT'L CONFERENCE ON AUDIO- AND VIDEO-BASED BIOMETRIC: PERSON AUTHENTICATION (AVBPA), WASHINGTON D.C., 22 March 1999 (1999-03-22) - 29 March 1999 (1999-03-29), pages 166 - 171 *
WEI XIONG ET AL.: "Model-guided deformable hand shape recognition without positioning aids", PATTERN RECOGNITION, vol. 38, no. 10, 2005, pages 1651 - 1664, XP004988746 *

Also Published As

Publication number Publication date
ES2315155A1 (es) 2009-03-16
ES2315155B1 (es) 2010-01-12

Similar Documents

Publication Publication Date Title
CN101669824B (zh) 基于生物特征识别的人与身份证同一性检验装置
CN108985134B (zh) 基于双目摄像机的人脸活体检测及刷脸交易方法及系统
US9064145B2 (en) Identity recognition based on multiple feature fusion for an eye image
Schouten et al. Biometrics and their use in e-passports
JP4246154B2 (ja) バイオメトリックな認証方法
Burge et al. Ear biometrics for machine vision
CN104143086A (zh) 人像比对在移动终端操作系统上的应用技术
CN201867835U (zh) 联合指纹识别与人脸近红外识别的系统
Ibrahim et al. Performance analysis of biometric recognition modalities
Bong et al. Palm print verification system
Yadav et al. Fingerprint recognition based on minutiae information
WO2017060805A1 (fr) Procédé et dispositif d'identification automatique de personnes et à mesurer la teneur en alcool dans l'air expiré
Gururaj et al. Threats, consequences and issues of various attacks on online social networks
ES2315155B1 (es) Metodo y sistema para reconocimiento de personas por analisis biometrico de manos.
Singh et al. Attendance monitoring and management using QR code based sensing with cloud based Processing
El-Bakry et al. Personal identification through biometric technology
Sharma et al. Biometric Security: A Review to Future
Sanchez-Reillo et al. Improving access control security using iris identification
Patil et al. Iris recognition using fuzzy system
JP2010009377A (ja) 照合システム、照合方法、プログラム及び記録媒体
Lodrová Security of biometric systems
KR20170067107A (ko) 복수 인원 홍채 인식 방법
Barra et al. MOHAB: Mobile hand-based biometric recognition
Rao et al. Fraudulent Detection in Examination Department
Chiesa Revisiting face processing with light field images

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 08736684

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 08736684

Country of ref document: EP

Kind code of ref document: A1