CN109145791A - One kind being based on the contactless fingers and palms recognition methods in mobile terminal and system - Google Patents
One kind being based on the contactless fingers and palms recognition methods in mobile terminal and system Download PDFInfo
- Publication number
- CN109145791A CN109145791A CN201810903071.0A CN201810903071A CN109145791A CN 109145791 A CN109145791 A CN 109145791A CN 201810903071 A CN201810903071 A CN 201810903071A CN 109145791 A CN109145791 A CN 109145791A
- Authority
- CN
- China
- Prior art keywords
- palm
- interest
- finger
- identification
- crease
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
- G06V40/13—Sensors therefor
- G06V40/1312—Sensors therefor direct reading, e.g. contactless acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
- G06V40/1347—Preprocessing; Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/12—Fingerprints or palmprints
- G06V40/1365—Matching; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/50—Maintenance of biometric data or enrolment thereof
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Collating Specific Patterns (AREA)
Abstract
The present invention is suitable for biometrics identification technology improvement areas, provides a kind of based on the contactless fingers and palms recognition methods in mobile terminal, comprising the following steps: S1, mobile terminal transfer camera and take pictures to pre-identification palm;S2, palm feature extraction is carried out to the pre-identification palm image of shooting;S3, the feature of the storage palm in the feature and database of the pre-identification palm of extraction is subjected to matching judgment;It is identified using Crease and palmmprint level, improves the stability and robustness of the identification of palmmprint single creature feature.Mobile terminal is used using contactless acquisition mode, and by palmmprint.
Description
Technical field
The invention belongs to biometrics identification technology improvement areas, more particularly to one kind to be based on the contactless fingers and palms in mobile terminal
Recognition methods and system.
Background technique
Information age today, human body biological characteristics identification technology is as a kind of strong protection personally identifiable information and object
The method of matter finance, is widely used in the every field such as security protection, education, medical treatment, financial payment and military affairs.So-called bio-identification skill
Art refers to the physiological property or behavioural characteristic of human body, such as: physiological property: fingerprint, palmmprint, face etc., behavioural characteristic: sound,
Gait, handwriting etc. identify a kind of technology of personal part as the individual character biological characteristic of people.Compared with traditional identity authenticating party
Method (such as key, identity card, password etc.), biological characteristic have that feature is stable, more difficult forge, is not easy to lose and using quick side
The characteristics such as just.
In numerous biological characteristics, fingerprint be considered as it is most popular and generally acknowledge most effective feature, it the shortcomings that be
Finger skin easily changes (such as dry and cracked and wet), and fingerprint is easy to obtain and copy.Recognition of face is current studies most
For burning hot feature, it the shortcomings that be easy to receive environment, illumination condition influences, and face characteristic is easily varied, and is not sufficiently stable
(such as makeup, lift face etc.).Vocal print speech recognition is highly prone to Environmental Noise Influence, and the sound of people also can by mood, age,
Physical condition is influenced.Iris recognition user's acceptance is lower, and equipment is more expensive.For the disadvantage more than solving, usually
Using the identification of the much information fusion including various biological characteristics.
Multi-modal biological characteristic identification technology is that a variety of biological characteristics of comprehensive utilization people carry out the emerging of identification
Biometrics identification technology.Identification system based on multi-modal biological characteristic identification technology is in noise immunity, universality, anti-vacation
It emits attack, inhibit to be superior to single mode living creature characteristic recognition system in many-sided performances such as big library decaying.Multi-modal feature group
The diversity and convergence strategy of conjunction and algorithm it is rich, some shortcomings of single mode are overcome, so as to realize more
The identification system of robust.
The research of multi-modal biological characteristic identification starts from nineteen ninety-five, Brunelli and Falavigna propose based on voice with
The bi-mode biology Feature Recognition System of face characteristic.Later, many new emerging systems are emerged, such as: fingerprint+vocal print,
Face+gait, vocal print+lip is dynamic, face+iris, the bi-mode biologies Fusion Features system such as palmmprint+hand-type.Multi-modal biology is special
Sign identification technology has become a popular research direction in living things feature recognition field.
The development of multi-modal biological characteristic identification technology based on hand-characteristic, hand images, which have, exists simultaneously a variety of lifes
Unique advantage, such as Kumar of object feature et al. propose fusion palmmprint and hand-shaped characteristic;Ribaric et al. verifies palmmprint and hand
The validity of shape, palmmprint and finger grain table;And fingerprint and hand back vein are merged;Hand shape, palmmprint and vena metacarpea
Feature carries out fusion recognition.Currently, only propose some new thinkings and convergence strategy, there are no formed system theory and at
Ripe method.
Summary of the invention
The purpose of the present invention is to provide one kind to be based on the contactless fingers and palms recognition methods in mobile terminal, it is intended to solve non-contact
Formula hand single mode recognition accuracy is low and hand identification cannot be the technical issues of mobile terminal be applied.
The invention is realized in this way it is a kind of based on the contactless fingers and palms recognition methods in mobile terminal, it is described to be based on mobile terminal
Contactless fingers and palms recognition methods the following steps are included:
S1, mobile terminal transfer camera and take pictures to pre-identification palm;
S2, palm feature extraction is carried out to the pre-identification palm image of shooting;
S3, the feature of the storage palm in the feature and database of the pre-identification palm of extraction is subjected to matching judgment, if it matches,
Then authentication success, if it does not match, authentication fails.
A further technical solution of the present invention is: further comprising the steps of in the step S2:
S21 removes ambient noise using Face Detection and largest connected domain to the pre-identification palm image of shooting, and utilizes shape
State pre-processes palm;
S22, to treated, pre-identification palm image carries out the feature extraction of palmmprint area-of-interest and Crease feature extraction.
A further technical solution of the present invention is: the step S22 is further comprising the steps of:
S221, binary conversion treatment and gaussian filtering process are carried out to the pre-identification palm image of removal background, and passes through Harris
Corner Detection determines the valley point B between index finger and valley point A, the third finger and the little finger of middle interphalangeal;
S222, connection valley point A and valley point B, the perpendicular bisector for making AB meets at point C, from A point along three minutes of perpendicular bisector direction line taking section AB
One of be a at distance, take from distance 128 from a point along the direction AB as b, take from distance 128 from b point along perpendicular bisector direction for c, from a
Point takes at distance 128 along perpendicular bisector direction as d, and point abcd is in turn connected to form palmmprint area-of-interest and to extract palmmprint sense emerging
Interesting provincial characteristics;
S223, whether continuous white area is more than half for the first time, is then partitioned into four hands in this way is judged to palm binary picture
Refer to region, and asks valley point coordinate to be partitioned into index finger, middle finger and the third finger grey level histogram;
S224, in segmentation finger-image, determine in grey level histogram that intermediate and continuous valley point is finger second knuckle folding line
Place, is partitioned into Crease area-of-interest and extracts Crease region of interest characteristic of field.
A further technical solution of the present invention is: further comprising the steps of in the step S3;
S31, by the Crease sense of palm in the Crease region of interest characteristic of field of the pre-identification palm of extraction and database
Whether the comparison of interest provincial characteristics matches, if so, then authentication success, if not, authentication fails and performs the next step;
S32, by the palmmprint area-of-interest of palm in the palmmprint region of interest characteristic of field of the pre-identification palm of extraction and database
Whether aspect ratio is to matching, if so, then authentication is successful, if not, authentication fails.
A further technical solution of the present invention is: further comprising the steps of before the step S1:
S01 takes pictures to palm by photographing device;
To palm photo, whether qualification judges S02, such as qualified, then photo is saved in database, extracts hand according to photo
The Crease region of interest characteristic of field and palmmprint region of interest characteristic of field of the palm simultaneously save in the database, on the other hand in database
Photo, Crease region of interest characteristic of field and the palmmprint region of interest characteristic of field of the palm are linked and are generated index number, such as
It is unqualified, then return step S01.
Another object of the present invention is to provide one kind to be based on the contactless fingers and palms identifying system in mobile terminal, described based on shifting
The contactless fingers and palms identifying system in moved end includes:
Image pickup module;Camera is transferred for mobile terminal to take pictures to pre-identification palm;
Characteristic extracting module, for carrying out palm feature extraction to the pre-identification palm image of shooting;
Authentication module is matched, the feature progress of the storage palm in the feature and database of the pre-identification palm for that will extract
With judgement, if it matches, then authentication success, if it does not match, authentication fails.
A further technical solution of the present invention is: in the characteristic extracting module further include:
Pretreatment denoising unit removes background using Face Detection and largest connected domain for the pre-identification palm image to shooting
Noise, and palm is pre-processed using morphology;
Feature extraction unit, for pre-identification palm image to carry out the feature extraction of palmmprint area-of-interest and finger joint to treated
Folding line feature extraction.
A further technical solution of the present invention is: the feature extraction unit is further comprising the steps of:
Valley point generates subelement, carries out at binary conversion treatment and gaussian filtering for the pre-identification palm image to removal background
Reason, and the valley point B between index finger and valley point A, the third finger and the little finger of middle interphalangeal is determined by Harris Corner Detection;
Palmprint feature extraction subelement, for linking valley point A and valley point B, the perpendicular bisector for making AB meets at point C, from A point along perpendicular bisector
It is a at the one third distance of direction line taking section AB, taking from distance 128 from a point along the direction AB is b, from b point along perpendicular bisector direction
It takes at distance 128 as c, takes from distance 128 from a point along perpendicular bisector direction as d, it is interested that point abcd is in turn connected to form palmmprint
Simultaneously extract palmmprint region of interest characteristic of field in region;
Region decision subelement, for judging whether continuous white area is more than half for the first time to palm binary picture, in this way
Four finger areas are then partitioned into, and ask valley point coordinate to be partitioned into index finger, middle finger and the third finger grey level histogram;
Crease feature extraction subelement, it is intermediate and continuous in grey level histogram for determining in segmentation finger-image
Valley point is to be partitioned into Crease area-of-interest at finger second knuckle folding line and extract Crease area-of-interest spy
Sign.
A further technical solution of the present invention is: the matching authentication module is further comprising the steps of;
In Crease authentication unit, the Crease region of interest characteristic of field of the pre-identification palm for that will extract and database
Whether the Crease area-of-interest aspect ratio of palm is to matching, if so, then authentication is successful, if not, authentication
Fail and executes palmprint authentication unit;
Palmprint authentication unit, the palm of the palmmprint region of interest characteristic of field and palm in database of the pre-identification palm for that will extract
Whether line area-of-interest aspect ratio is to matching, if so, then authentication is successful, if not, authentication fails.
A further technical solution of the present invention is: before described image acquisition module further include:
Photo module is sampled, for taking pictures by photographing device to palm;
Judgement saves index module, such as qualified for palm photo, whether qualification to judge, then photo is saved in data
Library according to the Crease region of interest characteristic of field of photo extraction palm and palmmprint region of interest characteristic of field and is stored in database
In, the photo of same palm, Crease region of interest characteristic of field and palmmprint region of interest characteristic of field are linked in database
And index number is generated, and it is such as unqualified, then return to sampling photo module.
The beneficial effects of the present invention are: being identified using Crease and palmmprint level, improves palmmprint single creature feature and know
Other stability and robustness.Mobile terminal is used using contactless acquisition mode, and by palmmprint.Contactless acquisition palm figure
Picture overcomes potential palmmprint leftover problem, overcome the problems, such as due to palm perspire, spot and can not be in the identification in contact system;
The utilization of traditional personal recognition is nearly all contact embedded system, and personal recognition is applied to mobile phone mobile terminal;It is slapping
It in the positioning of line area-of-interest, is positioned using Crease and key point, increases accuracy and the Shandong of palmmprint positioning
Stick increases stability for subsequent identification work.
Detailed description of the invention
Fig. 1 is system architecture diagram provided in an embodiment of the present invention.
Fig. 2 is method flow diagram provided in an embodiment of the present invention.
Fig. 3 is extraction palmmprint area-of-interest schematic diagram provided in an embodiment of the present invention.
Fig. 4 is extraction Crease area-of-interest schematic diagram provided in an embodiment of the present invention.
Specific embodiment
As shown in Figure 1, the flow chart provided by the invention based on the contactless fingers and palms recognition methods in mobile terminal, is described in detail such as
Under:
Step S01 takes pictures to palm by photographing device;It, be by taking pictures to pre-established hand identification authentication
And picture pick-up device, everyone palm take pictures acquiring image.
Step S02, to the palm image of acquisition, whether qualification judges, such as qualified, then photo is saved in database,
According to the Crease region of interest characteristic of field of photo extraction palm and palmmprint region of interest characteristic of field and save in the database,
The photo of same palm, Crease region of interest characteristic of field and palmmprint region of interest characteristic of field are linked and are given birth in database
It is such as unqualified at index number, then return step S01.
Step S1, mobile terminal transfer camera and take pictures to pre-identification palm;Meeting preview palm figure is completed in shooting every time
Picture, if fogging image, user may be selected to delete and re-shoot, until image meets.
Step S2 carries out palm feature extraction to the pre-identification palm image of shooting;Skin is used to the palm image of shooting
Color detection and largest connected domain remove ambient noise, are handled using morphology palm image, wherein Face Detection is not
Rgb value with the colour of skin under the conditions of illumination meets R > G > B, and the cluster structure of the colour of skin is simple and stablizes;Largest connected domain is in two-value
In image with same pixel value and position it is adjacent pixel composition maximum image region;Morphology: corrosion can melt
The boundary of object, expansion expand the boundary of object, opening operation be first corrode expand afterwards, closed operation is first to expand post-etching.It is right
The palm image for removing background carries out binaryzation, gaussian filtering, then determines index finger and middle interphalangeal by Harris Corner Detection
Valley point, the valley point between the third finger and little finger, establish coordinate system, wherein binary conversion treatment is by the pixel on image
Gray value is set as 0 or 255, and apparent black and white effect is presented;Gaussian filter is a kind of linear smoothing filter, filter
Template be obtain discrete to two-dimensional Gaussian function.Since the central value of Gaussian template is maximum, surrounding is gradually reduced, after filtering
Result it is more preferable for mean filter.The most important parameter of Gaussian filter is exactly the standard deviation sigma of Gaussian Profile, mark
The smoothing capability of quasi- difference and Gaussian filter has very big ability, and σ is bigger, and the frequency band of Gaussian filter is just wider, to image
Smoothness is better.By adjusting σ parameter, it can balance to the inhibition of the noise of image and image is obscured;Harris
Corner Detection is that the identification of human eye angle steel joint is usually to complete in the zonule of part or wicket.If in each side
The wicket of this feature is moved up, biggish variation has occurred in the gray scale of window inner region, it is judged that in window
Encounter angle point.If this specific window when being moved in image all directions, the gray scale of image in window there is no
Variation, then angle point is just not present in window;If window moves in a certain direction, the gray scale of image in window is had occurred
Biggish variation, and there is no variations on other directions, then, the image in window may be exactly the line of straight line
Section.Harris uses a more smooth Gauss window, insensitive to the variation of brightness and contrast, has rotation not
Denaturation does not have scale invariability.
Link valley point A and valley point B, the perpendicular bisector for making AB meets at point C, from A point along three minutes of perpendicular bisector direction line taking section AB
One of be a at distance, take from distance 128 from a point along the direction AB as b, take from distance 128 from b point along perpendicular bisector direction for c, from a
Point takes at distance 128 along perpendicular bisector direction as d, and point abcd is in turn connected to form palmmprint area-of-interest and to extract palmmprint sense emerging
Interesting provincial characteristics.
Whether continuous white area is more than half for the first time, is then partitioned into four fingers in this way is judged to palm binary picture
Region, and ask valley point coordinate to be partitioned into index finger, middle finger and the third finger grey level histogram.
In segmentation finger-image, determine that intermediate and continuous valley point is finger second knuckle folding line in grey level histogram
Place, is partitioned into Crease area-of-interest and extracts Crease region of interest characteristic of field.
The feature of the pre-identification palm of extraction match sentencing by step S3 with the feature of the storage palm in database
It is disconnected, if it matches, then authentication success, if it does not match, authentication fails.It is stored in when by Crease feature with registration
Crease feature in database carries out Hamming distance matching, determines Crease threshold value, if successful match, authentication
Pass through;If matching is unsuccessful, palm print characteristics are subjected to Hamming distance matching, determine palmmprint threshold value, if successful match, identity is recognized
Card passes through;If matching is unsuccessful, authentication failure;S31, by the Crease area-of-interest of the pre-identification palm of extraction
Whether feature and the Crease area-of-interest aspect ratio of palm in database are to matching, if so, then authentication success, such as
No, then authentication fails and performs the next step;S32, palmmprint region of interest characteristic of field and number by the pre-identification palm of extraction
Whether the palmmprint area-of-interest aspect ratio according to palm in library is to matching, if so, then authentication is successful, if not, identity is recognized
Card failure.
The present invention is based on the contactless fingers and palms recognition methods in mobile terminal, include the following steps:
This system is opened first to register and identify again if user is unregistered;If user is registered, Direct Recognition.
This system is used for the first time, is had System guides and is illustrated to show, then user fills in user name, password is registered,
Start mobile phone camera and shoots palm image.
Meeting preview palm image is completed in shooting every time, if fogging image, user may be selected to delete and re-shoot, and is stored in
In database.
Ambient noise is removed using Face Detection and largest connected domain to the palm image of shooting, using morphology to palm
Image is handled.
Binaryzation, gaussian filtering are carried out to the palm image of removal background, food is then determined by Harris Corner Detection
Refer to the valley point between the valley point of middle interphalangeal, the third finger and little finger, establishes coordinate system.
Fig. 3 is to extract palmmprint area-of-interest figure, links valley point A and valley point B, the perpendicular bisector for making AB meets at point C, from A point
It is a along the one third distance of perpendicular bisector direction line taking section AB, taking from a point along the direction AB is b from distance 128, from b point in
Vertical line direction takes for c at distance 128, and taking from a point along perpendicular bisector direction is d from distance 128, and rectangle abcd is palmmprint region of interest
Domain.
When continuous white area is more than the wide half of image for the first time to the judgement of palm binary map, it is partitioned into four
Then finger areas seeks valley point coordinate to grey level histogram, be partitioned into index finger, middle finger and the third finger.
By taking index finger as an example, in the index finger image of segmentation, determine that intermediate and continuous valley point is index finger in grey level histogram
At second knuckle folding line, it is then partitioned into Crease area-of-interest, extracts middle finger and nameless finger by the same way
Save folding line area-of-interest.
The area-of-interest of extraction is saved in the database in the form of picture.
The area-of-interest of extraction is extracted into Crease feature and palm print characteristics with Competition coding respectively.
When user identifies, start mobile phone camera, shoots palm image.
To the palm image zooming-out finger Crease of shooting and the area-of-interest of palmmprint, respective spy is then extracted
Sign.
The Crease feature saved in the database when first by Crease feature and registration carries out Hamming distance
Match, determines Crease threshold value, if successful match, authentication passes through;If matching is unsuccessful, palm print characteristics are subjected to Hamming
Distance matching, determines palmmprint threshold value, if successful match, authentication passes through;If matching is unsuccessful, authentication failure.
Another object of the present invention is to provide one kind to be based on the contactless fingers and palms identifying system in mobile terminal, described based on shifting
The contactless fingers and palms identifying system in moved end includes:
Image pickup module;Camera is transferred for mobile terminal to take pictures to pre-identification palm;
Characteristic extracting module, for carrying out palm feature extraction to the pre-identification palm image of shooting;
Authentication module is matched, the feature progress of the storage palm in the feature and database of the pre-identification palm for that will extract
With judgement, if it matches, then authentication success, if it does not match, authentication fails.
In the characteristic extracting module further include:
Pretreatment denoising unit removes background using Face Detection and largest connected domain for the pre-identification palm image to shooting
Noise, and palm is pre-processed using morphology;
Feature extraction unit, for pre-identification palm image to carry out the feature extraction of palmmprint area-of-interest and finger joint to treated
Folding line feature extraction.
The feature extraction unit is further comprising the steps of:
Valley point generates subelement, carries out at binary conversion treatment and gaussian filtering for the pre-identification palm image to removal background
Reason, and the valley point B between index finger and valley point A, the third finger and the little finger of middle interphalangeal is determined by Harris Corner Detection;
Palmprint feature extraction subelement, for linking valley point A and valley point B, the perpendicular bisector for making AB meets at point C, from A point along perpendicular bisector
It is a at the one third distance of direction line taking section AB, taking from distance 128 from a point along the direction AB is b, from b point along perpendicular bisector direction
It takes at distance 128 as c, takes from distance 128 from a point along perpendicular bisector direction as d, it is interested that point abcd is in turn connected to form palmmprint
Simultaneously extract palmmprint region of interest characteristic of field in region;
Region decision subelement, for judging whether continuous white area is more than half for the first time to palm binary picture, in this way
Four finger areas are then partitioned into, and ask valley point coordinate to be partitioned into index finger, middle finger and the third finger grey level histogram;
Crease feature extraction subelement, it is intermediate and continuous in grey level histogram for determining in segmentation finger-image
Valley point is to be partitioned into Crease area-of-interest at finger second knuckle folding line and extract Crease area-of-interest spy
Sign.
The matching authentication module is further comprising the steps of;
In Crease authentication unit, the Crease region of interest characteristic of field of the pre-identification palm for that will extract and database
Whether the Crease area-of-interest aspect ratio of palm is to matching, if so, then authentication is successful, if not, authentication
Fail and executes palmprint authentication unit;
Palmprint authentication unit, the palm of the palmmprint region of interest characteristic of field and palm in database of the pre-identification palm for that will extract
Whether line area-of-interest aspect ratio is to matching, if so, then authentication is successful, if not, authentication fails.
Before described image acquisition module further include:
Photo module is sampled, for taking pictures by photographing device to palm;
Judgement saves index module, such as qualified for palm photo, whether qualification to judge, then photo is saved in data
Library according to the Crease region of interest characteristic of field of photo extraction palm and palmmprint region of interest characteristic of field and is stored in database
In, the photo of same palm, Crease region of interest characteristic of field and palmmprint region of interest characteristic of field are linked in database
And index number is generated, and it is such as unqualified, then return to sampling photo module.
Fig. 1 is system architecture diagram of the invention, and the present invention is a kind of contactless fingers and palms identifying system in mobile terminal, including note
Volume module 1, database 2 and identification module 3;
Registration module 1 is for registering new user.The user used for the first time fills in username and password, then starting camera shooting
Head shooting palm image saves the palm image of complete display, image is not by user's interaction guidance user's palm placement position
When qualified, user can re-shoot palm image;
Database 2 is used to save the characteristic information of the information of user's registration, palm image and Crease and palmmprint.Setting is left
The right hand respectively saves 2, after saving palm image, to every image preprocessing and assessment, selects optimal image zooming-out Fingers
Then the area-of-interest for saving folding line and palmmprint extracts corresponding characteristic information, save in the database;
Identification module 3 is for verifying subscriber identity information, when user will verify identity, the automatic calling mobile phone camera of system into
Row is taken pictures, and feature extracting method when according to registration is to the palm image zooming-out Crease of shooting and the region of interest of palmmprint
Domain, then feature is extracted to each section, it is matched with the feature in database 2, the authentication if successful match, if matching is lost
It loses the optional input password of user and carries out verifying identity.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
1. one kind is based on the contactless fingers and palms recognition methods in mobile terminal, which is characterized in that described to be based on the contactless finger in mobile terminal
Slap recognition methods the following steps are included:
S1, mobile terminal transfer camera and take pictures to pre-identification palm;
S2, palm feature extraction is carried out to the pre-identification palm image of shooting;
S3, the feature of the storage palm in the feature and database of the pre-identification palm of extraction is subjected to matching judgment, if it matches,
Then authentication success, if it does not match, authentication fails.
2. according to claim 1 be based on the contactless fingers and palms recognition methods in mobile terminal, which is characterized in that the step S2
In it is further comprising the steps of:
S21 removes ambient noise using Face Detection and largest connected domain to the pre-identification palm image of shooting, and utilizes shape
State pre-processes palm;
S22, to treated, pre-identification palm image carries out the feature extraction of palmmprint area-of-interest and Crease feature extraction.
3. according to claim 2 be based on the contactless fingers and palms recognition methods in mobile terminal, which is characterized in that the step
S22 is further comprising the steps of:
S221, binary conversion treatment and gaussian filtering process are carried out to the pre-identification palm image of removal background, and passes through Harris
Corner Detection determines the valley point B between index finger and valley point A, the third finger and the little finger of middle interphalangeal;
S222, connection valley point A and valley point B, the perpendicular bisector for making AB meets at point C, from A point along three minutes of perpendicular bisector direction line taking section AB
One of be a at distance, take from distance 128 from a point along the direction AB as b, take from distance 128 from b point along perpendicular bisector direction for c, from a
Point takes at distance 128 along perpendicular bisector direction as d, and point abcd is in turn connected to form palmmprint area-of-interest and to extract palmmprint sense emerging
Interesting provincial characteristics;
S223, whether continuous white area is more than half for the first time, is then partitioned into four hands in this way is judged to palm binary picture
Refer to region, and asks valley point coordinate to be partitioned into index finger, middle finger and the third finger grey level histogram;
S224, in segmentation finger-image, determine in grey level histogram that intermediate and continuous valley point is finger second knuckle folding line
Place, is partitioned into Crease area-of-interest and extracts Crease region of interest characteristic of field.
4. according to claim 1-3 be based on the contactless fingers and palms recognition methods in mobile terminal, which is characterized in that institute
It states further comprising the steps of in step S3;
S31, by the Crease sense of palm in the Crease region of interest characteristic of field of the pre-identification palm of extraction and database
Whether the comparison of interest provincial characteristics matches, if so, then authentication success, if not, authentication fails and performs the next step;
S32, by the palmmprint area-of-interest of palm in the palmmprint region of interest characteristic of field of the pre-identification palm of extraction and database
Whether aspect ratio is to matching, if so, then authentication is successful, if not, authentication fails.
5. according to claim 4 be based on the contactless fingers and palms recognition methods in mobile terminal, which is characterized in that the step S1
It is before further comprising the steps of:
S01 takes pictures to palm by photographing device;
To palm photo, whether qualification judges S02, such as qualified, then photo is saved in database, extracts hand according to photo
The Crease region of interest characteristic of field and palmmprint region of interest characteristic of field of the palm simultaneously save in the database, on the other hand in database
Photo, Crease region of interest characteristic of field and the palmmprint region of interest characteristic of field of the palm are linked and are generated index number, such as
It is unqualified, then return step S01.
6. one kind is based on the contactless fingers and palms identifying system in mobile terminal, which is characterized in that described to be based on the contactless finger in mobile terminal
Slapping identifying system includes:
Image pickup module;Camera is transferred for mobile terminal to take pictures to pre-identification palm;
Characteristic extracting module, for carrying out palm feature extraction to the pre-identification palm image of shooting;
Authentication module is matched, the feature progress of the storage palm in the feature and database of the pre-identification palm for that will extract
With judgement, if it matches, then authentication success, if it does not match, authentication fails.
7. according to claim 6 be based on the contactless fingers and palms identifying system in mobile terminal, which is characterized in that the feature mentions
In modulus block further include:
Pretreatment denoising unit removes background using Face Detection and largest connected domain for the pre-identification palm image to shooting
Noise, and palm is pre-processed using morphology;
Feature extraction unit, for pre-identification palm image to carry out the feature extraction of palmmprint area-of-interest and finger joint to treated
Folding line feature extraction.
8. according to claim 7 be based on the contactless fingers and palms identifying system in mobile terminal, which is characterized in that the feature mentions
Take unit further comprising the steps of:
Valley point generates subelement, carries out at binary conversion treatment and gaussian filtering for the pre-identification palm image to removal background
Reason, and the valley point B between index finger and valley point A, the third finger and the little finger of middle interphalangeal is determined by Harris Corner Detection;
Palmprint feature extraction subelement, for linking valley point A and valley point B, the perpendicular bisector for making AB meets at point C, from A point along perpendicular bisector
It is a at the one third distance of direction line taking section AB, taking from distance 128 from a point along the direction AB is b, from b point along perpendicular bisector direction
It takes at distance 128 as c, takes from distance 128 from a point along perpendicular bisector direction as d, it is interested that point abcd is in turn connected to form palmmprint
Simultaneously extract palmmprint region of interest characteristic of field in region;
Region decision subelement, for judging whether continuous white area is more than half for the first time to palm binary picture, in this way
Four finger areas are then partitioned into, and ask valley point coordinate to be partitioned into index finger, middle finger and the third finger grey level histogram;
Crease feature extraction subelement, it is intermediate and continuous in grey level histogram for determining in segmentation finger-image
Valley point is to be partitioned into Crease area-of-interest at finger second knuckle folding line and extract Crease area-of-interest spy
Sign.
9. being based on the contactless fingers and palms identifying system in mobile terminal according to claim 6-8 is described in any item, which is characterized in that institute
It is further comprising the steps of to state matching authentication module;
In Crease authentication unit, the Crease region of interest characteristic of field of the pre-identification palm for that will extract and database
Whether the Crease area-of-interest aspect ratio of palm is to matching, if so, then authentication is successful, if not, authentication
Fail and executes palmprint authentication unit;
Palmprint authentication unit, the palm of the palmmprint region of interest characteristic of field and palm in database of the pre-identification palm for that will extract
Whether line area-of-interest aspect ratio is to matching, if so, then authentication is successful, if not, authentication fails.
10. according to claim 9 be based on the contactless fingers and palms identifying system in mobile terminal, which is characterized in that described image
Before acquisition module further include:
Photo module is sampled, for taking pictures by photographing device to palm;
Judgement saves index module, such as qualified for palm photo, whether qualification to judge, then photo is saved in data
Library according to the Crease region of interest characteristic of field of photo extraction palm and palmmprint region of interest characteristic of field and is stored in database
In, the photo of same palm, Crease region of interest characteristic of field and palmmprint region of interest characteristic of field are linked in database
And index number is generated, and it is such as unqualified, then return to sampling photo module.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810903071.0A CN109145791A (en) | 2018-08-09 | 2018-08-09 | One kind being based on the contactless fingers and palms recognition methods in mobile terminal and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810903071.0A CN109145791A (en) | 2018-08-09 | 2018-08-09 | One kind being based on the contactless fingers and palms recognition methods in mobile terminal and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109145791A true CN109145791A (en) | 2019-01-04 |
Family
ID=64792326
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810903071.0A Pending CN109145791A (en) | 2018-08-09 | 2018-08-09 | One kind being based on the contactless fingers and palms recognition methods in mobile terminal and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109145791A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109910818A (en) * | 2019-02-15 | 2019-06-21 | 东华大学 | A kind of VATS Vehicle Anti-Theft System based on human body multiple features fusion identification |
CN110012034A (en) * | 2019-05-10 | 2019-07-12 | 济南大学 | Biometric authentication method under cloud environment with secret protection |
CN110008824A (en) * | 2019-02-20 | 2019-07-12 | 平安科技(深圳)有限公司 | Palm grain identification method, device, computer equipment and storage medium |
CN110334598A (en) * | 2019-05-31 | 2019-10-15 | 平安科技(深圳)有限公司 | A kind of palm grain identification method and device |
CN110728232A (en) * | 2019-10-10 | 2020-01-24 | 清华大学深圳国际研究生院 | Hand region-of-interest acquisition method and hand pattern recognition method |
CN110879956A (en) * | 2019-07-12 | 2020-03-13 | 熵加网络科技(北京)有限公司 | Method for extracting palm print features |
CN114022912A (en) * | 2021-11-10 | 2022-02-08 | 厦门熵基科技有限公司 | Palm recognition model training method, device and equipment and readable storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102739856A (en) * | 2012-05-31 | 2012-10-17 | 西安电子科技大学 | Mobile phone unlocking system and method based on palm image information |
US20130027184A1 (en) * | 2011-07-29 | 2013-01-31 | Fujitsu Limited | Biometric authentication device and adjustment method for position of hand of user in biometric authentication device |
CN103268483A (en) * | 2013-05-31 | 2013-08-28 | 沈阳工业大学 | Method for recognizing palmprint acquired in non-contact mode in open environment |
CN107516088A (en) * | 2017-09-02 | 2017-12-26 | 宜宾学院 | A kind of more finger segments lines recognition methods |
-
2018
- 2018-08-09 CN CN201810903071.0A patent/CN109145791A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130027184A1 (en) * | 2011-07-29 | 2013-01-31 | Fujitsu Limited | Biometric authentication device and adjustment method for position of hand of user in biometric authentication device |
CN102739856A (en) * | 2012-05-31 | 2012-10-17 | 西安电子科技大学 | Mobile phone unlocking system and method based on palm image information |
CN103268483A (en) * | 2013-05-31 | 2013-08-28 | 沈阳工业大学 | Method for recognizing palmprint acquired in non-contact mode in open environment |
CN107516088A (en) * | 2017-09-02 | 2017-12-26 | 宜宾学院 | A kind of more finger segments lines recognition methods |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109910818A (en) * | 2019-02-15 | 2019-06-21 | 东华大学 | A kind of VATS Vehicle Anti-Theft System based on human body multiple features fusion identification |
CN109910818B (en) * | 2019-02-15 | 2021-10-08 | 东华大学 | Vehicle anti-theft system based on human body multi-feature fusion identity recognition |
CN110008824A (en) * | 2019-02-20 | 2019-07-12 | 平安科技(深圳)有限公司 | Palm grain identification method, device, computer equipment and storage medium |
CN110008824B (en) * | 2019-02-20 | 2023-09-22 | 平安科技(深圳)有限公司 | Palmprint recognition method, palmprint recognition device, palmprint recognition computer device and palmprint recognition storage medium |
CN110012034A (en) * | 2019-05-10 | 2019-07-12 | 济南大学 | Biometric authentication method under cloud environment with secret protection |
CN110334598A (en) * | 2019-05-31 | 2019-10-15 | 平安科技(深圳)有限公司 | A kind of palm grain identification method and device |
CN110879956A (en) * | 2019-07-12 | 2020-03-13 | 熵加网络科技(北京)有限公司 | Method for extracting palm print features |
CN110728232A (en) * | 2019-10-10 | 2020-01-24 | 清华大学深圳国际研究生院 | Hand region-of-interest acquisition method and hand pattern recognition method |
CN114022912A (en) * | 2021-11-10 | 2022-02-08 | 厦门熵基科技有限公司 | Palm recognition model training method, device and equipment and readable storage medium |
CN114022912B (en) * | 2021-11-10 | 2024-06-28 | 厦门熵基科技有限公司 | Palm recognition model training method, device, equipment and readable storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109145791A (en) | One kind being based on the contactless fingers and palms recognition methods in mobile terminal and system | |
Nigam et al. | Designing an accurate hand biometric based authentication system fusing finger knuckleprint and palmprint | |
CN105320946B (en) | A kind of fingerprint identification method based on MATLAB | |
JP4351982B2 (en) | Personal authentication method, apparatus and program | |
Parkavi et al. | Multimodal biometrics for user authentication | |
Bakshi et al. | A novel phase-intensive local pattern for periocular recognition under visible spectrum | |
CN111462379A (en) | Access control management method, system and medium containing palm vein and face recognition | |
CN110555380A (en) | Finger vein identification method based on Center Loss function | |
Prakash et al. | Ear biometrics in 2D and 3D: localization and recognition | |
CN109583279A (en) | A kind of fingerprint and refer to that vein combines recognizer | |
Chuang | Vein recognition based on minutiae features in the dorsal venous network of the hand | |
Patil et al. | A novel approach for fingerprint matching using minutiae | |
Yang et al. | A hybrid finger identification pattern using polarized depth-weighted binary direction coding | |
Burge et al. | Using ear biometrics for passive identification | |
Kassem et al. | An enhanced ATM security system using multimodal biometric strategy | |
Wirayuda | Palm vein recognition based-on minutiae feature and feature matching | |
Barham et al. | Fingerprint recognition using MATLAB | |
CN109697347A (en) | Based on the user characteristics authentication method and its device for referring to vein and finger feature | |
Sisodia et al. | A conglomerate technique for finger print recognition using phone camera captured images | |
Shukla et al. | A Hybrid Model of Multimodal Biometrics System using Fingerprint and Face as Traits | |
Rajbhoj et al. | An Improved binarization based algorithm using minutiae approach for Fingerprint Identification | |
KR100467392B1 (en) | Method for identifing biometric person using a finger crease pattern and apparatus thereof | |
Naveena et al. | Partial face recognition by template matching | |
Sharma et al. | Fingerprint matching Using Minutiae Extraction Techniques | |
Vasilopoulos et al. | A novel finger vein recognition system based on enhanced maximum curvature points |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190104 |
|
RJ01 | Rejection of invention patent application after publication |