CN108564031A - Single width near-infrared palm image-recognizing method based on multi-modal fusion - Google Patents
Single width near-infrared palm image-recognizing method based on multi-modal fusion Download PDFInfo
- Publication number
- CN108564031A CN108564031A CN201810324828.0A CN201810324828A CN108564031A CN 108564031 A CN108564031 A CN 108564031A CN 201810324828 A CN201810324828 A CN 201810324828A CN 108564031 A CN108564031 A CN 108564031A
- Authority
- CN
- China
- Prior art keywords
- image
- vena metacarpea
- palmmprint
- infrared
- enhanced
- 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
- G06V40/1347—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
-
- 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
- G06V40/70—Multimodal biometrics, e.g. combining information from different biometric modalities
-
- 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
- G06V40/14—Vascular patterns
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)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of single width near-infrared palm image-recognizing method based on multi-modal fusion.First, the vena metacarpea in image is removed using sectional pattern and obtains palmmprint structure, palmmprint structure is blurred by self-defined membership function, then carry out unsharp masking enhancing, prominent palmmprint structural information；Then, vena metacarpea structure is enhanced using edge detection weighting guiding filtering, prominent vena metacarpea structure；Finally, palmmprint and vena metacarpea image are adaptively merged.In near-infrared palm image recognition, contrast experiment is carried out using the near-infrared palmprint image database that The Hong Kong Polytechnic University provides, experimental result is shown, compares other congenic methods, and the present processes have higher discrimination, reached 99.81%.
Description
Technical field
The invention belongs to biological characteristics identity recognizing technology field, more particularly to a kind of single width based on multi-modal fusion is close
Infrared palm image-recognizing method.
Background technology
With social economy's fast development, requirement of the people to identification technology is higher and higher.Due to magnetic card, the tradition such as password
There is the risk lost and stolen in identification authentication mode, cannot meet the needs of today's society, at the same time, biological identification technology
Appearance, be greatly improved in safety and convenience, therefore, biological identification technology is widely used.It is raw
Object identification technology has single creature feature identification technique and multiple features fusion identification technology；Human body different parts living things feature recognition
There is also very big differences for technology.Such as：Face recognition technology, precision are relatively low；Iris recognition technology is relatively safer can
By bigger but acquire expense.Compared to these biometrics identification technologies, the information based on hand biological identification technology is more
It is easy acquisition and acquisition cost is relatively low.Single creature identification technology based on hand includes：Fingerprint recognition, finger vena are known
Not, personal recognition, Palm-dorsal vein recognition, palm vein identification etc..Multiple features fusion technology includes：Based on discrete cosine transform
Face, palm vein and palmprint image Feature-level fusion, this method uses partial statistics method, with pre-defined DCT coefficient
Block calculates standard deviation, and is stored as feature vector.Utilize the progress of the distance between test vector and training dataset
Match；Hand, palmmprint and vena metacarpea fusion recognition, this method acquire vena metacarpea and palmprint image under same equipment difference light source.
First, it is matched for the first time using the relative length of finger as hand-shaped characteristic.Then, it is slapped using piecemeal texture primitive model
The feature extraction of vein and palmmprint blending image.Finally, Secondary Match is carried out and as final recognition result；Palmmprint is quiet with the back of the hand
Arteries and veins merges, and this method shoots palmmprint and vein image by low resolution digital scanner and infrared camera, is merged using mixing
Rule carries out Fusion Features；Multiple features fusion identification technology needs to acquire more information, and increasing for information content increases information
Acquisition, information fusion, the difficulty identified etc., these problems also bring new challenge to multi-biometric feature recognition technical research.
Currently, most researchers in multiple features fusion, are multiple characteristic images under the different illumination of acquisition, then into
Row fusion, gatherer process are complex；Li Junlin etc. believes comprising palmmprint and vena metacarpea structure simultaneously according to near-infrared palm image
Breath is attempted to obtain palmmprint, vena metacarpea structure from piece image, then be merged, and can thus improve image recognition
Reduce Image Acquisition while rate and reduces the difficulty of system globe area.But the fusion recognition algorithm fails the prominent palm well
Line structure and vena metacarpea structure.
Invention content
The purpose of the present invention is in view of the drawbacks of the prior art, provide a kind of single width near-infrared hand based on multi-modal fusion
Slap image-recognizing method.
To achieve the goals above, the present invention uses following technical scheme：A kind of single width based on multi-modal fusion is closely red
Outer palm image-recognizing method, includes the following steps：
Step 1：It inputs original near-infrared palm image and picture size is normalized；
Step 2：To treated, image removes vena metacarpea simultaneously using the texture structure of piecemeal enhancing model extraction palmmprint
Information；
Step 3：Using the membership function of definition, the palmmprint structure that step 2 is extracted is carried out obscuring unsharp enhance
To the palmprint image of enhancing；
Step 4：It is then right to step 1 treated near-infrared palm image using bootstrap filtering removal palm print information
Vena metacarpea information carries out the vena metacarpea image that adaptive-filtering is enhanced；
Step 5：Palmprint image and vena metacarpea image that step 3,4 obtain are carried out feature adaptively to merge, obtained similar
Coefficient；
Step 6：The near-infrared palm image for choosing database is trained according to step 1 to 5, similar after being weighted
Coefficient, and be recognition threshold by the minimum value of similarity factor is weighted, to the weighting similarity factor and training sample of the sample of identification
Threshold value be compared, if the threshold value of weighting similarity factor >=training sample of the sample of identification, identification are correct.
Further, step 2 specifically uses following steps：
Step 2.1：Near-infrared palm image I after normalized is divided into the fritter of R*R, calculates each fritter picture
The mean value of plain gray value obtains the background matrix I that dimension becomes smallerback；
Step 2.2：To the background matrix I obtained after piecemeal processingbackBicubic interpolation is carried out, the dimension of background matrix is made
Identical with the dimension of near-infrared palm image array I, background matrix is denoted as I' at this timeback；
Step 2.3：Near-infrared palm image array I is subtracted into bicubic interpolation treated background matrix I'back, to I-
I'backCarry out histogram equalization, the palmmprint structural images enhanced.
Further, step 3 specifically uses following steps：
Step 3.1:The membership function of definition is as follows：
Parameter after optimization is as follows：
Wherein, δ is the standard deviation of domain X, and m is the mean value of domain X.C1, c2, c3, c4 are four parameters, this four values
It is to be determined by statistical value；
Step 3.1：Using step 2 treated palmprint image as input picture I0；
Step 3.2：To input picture I0Carry out Ienhance=I0-μIfuzzyProcessing, obtains enhanced palmmprint structural images；
Wherein, IfuzzyFor I0Image array after membership function is blurred, μ are details enhancing coefficients.
Further, step 4 specifically uses following steps：
Step 4.1：Using bootstrap filtering the palm image I of input is filtered, take filter radius r=2 and
Regularization factor lambda=0.01 obtains filtered result q1；
Step 4.2：By q1As the navigational figure and input picture of guiding filtering, r=16 and λ=0.01 are chosen, is carried out
After bootstrap is filtered, the image q after smooth palmmprint is obtained2；
Step 4.3：By linearly enhancing model I'=(q1-q2)·t+q2Vena metacarpea structure is enhanced, is enhanced
Vena metacarpea image I' afterwards, the linear coefficient t=5 enhanced in model；
Step 4.4：Step 4.1 is repeated to 4.3, further enhances vena metacarpea structure, output image is vena metacarpea enhancing
Image afterwards.
Further, step 5 specifically uses following steps：
Step 1：It inputs original near-infrared palm image and picture size is normalized；
Step 2：To treated, image removes vena metacarpea simultaneously using the texture structure of piecemeal enhancing model extraction palmmprint
Information；
Step 3：Using the membership function of definition, the palmmprint structure that step 2 is extracted is carried out obscuring unsharp enhance
To the palmprint image of enhancing；
Step 4：It is then right to step 1 treated near-infrared palm image using bootstrap filtering removal palm print information
Vena metacarpea information carries out the vena metacarpea image that adaptive-filtering is enhanced；
Step 5：Palmprint image and vena metacarpea image that step 3,4 obtain are carried out feature adaptively to merge, obtained similar
Coefficient；
Step 6：The near-infrared palm image for choosing database is trained according to step 1 to 5, similar after being weighted
Coefficient, and be recognition threshold by the minimum value of similarity factor is weighted, to the weighting similarity factor and training sample of the sample of identification
Threshold value be compared, if the threshold value of weighting similarity factor >=training sample of the sample of identification, identification are correct.
Further, the similarity factor in step 6 is obtained using formula 4：
R′s(i)=w1(i)Rvein(i)+(1-w1(i))Rprint(i) (4)
Wherein Rvein(i) it is the minimum value of the similarity factor of vena metacarpea in training sample；Rprint(i) it is in training sample,
The minimum value of the similarity factor of palmmprint；w1It is the weights of vena metacarpea similarity factor, is obtained using formula 5：
w1=1/3*LV1/(LV1+LP1)+1/3*LV2/(LV2+LP2)+1/3*LV3/(LV3+LP3) (5)
G is database sample, i ∈ (1,2,3 ..., G)；LV1,LV2,LV3, it is weigh vena metacarpea picture quality three
Index；LP1,LP2,LP3It is three indexs for weighing palmprint image quality.
No matter the present invention has best performance from discrimination or stability.Since the present invention enhances in palmmprint
When, using customized membership function, high gray areas and low gray level areas are separated, when by obscuring unsharp enhancing,
It can be very good prominent palmmprint structure so that near-infrared palm image Palm-print Recognizing Rate higher.When vena metacarpea enhances, because of side
Edge weight detection guiding filtering has carried out self-adaptive processing to image venous structures regional peace skating area domain, is filtered using the guiding
Wave enhances vena metacarpea structure, preferably prominent vena metacarpea structural information so that near-infrared palm image vena metacarpea identifies
Rate higher.Therefore, the discrimination after fusion also improves；The present invention is applied in near-infrared palm image palmmprint, the quiet fusion of the palm
In identification, preferable effect is achieved.
The application is by the palmmprint vena metacarpea fusion identification method application near-infrared palm image proposed, in Hong Kong science and engineering
Contrast experiment is carried out in the near-infrared palmprint image database that university provides, experimental result is shown, compares other congenic methods, this
Invention has higher discrimination, has reached 99.81%.Experiment has absolutely proved the palmmprint vena metacarpea fusion recognition side of the present invention
The validity and universality of method.
Description of the drawings
Fig. 1 is the flow chart of embodiment near-infrared palm image palmmprint Enhancement Method.
Fig. 2 is embodiment near-infrared palm image vena metacarpea Enhancement Method flow chart.
Fig. 3 is embodiment near-infrared palm image palmmprint vena metacarpea fusion method flow chart.
Fig. 4 is embodiment (a) original image；(b) palmmprint enhances image；(c) vena metacarpea enhances image.
Specific implementation mode
The present invention is described further in the following with reference to the drawings and specific embodiments.
Picture library used in the present invention is the near-infrared palmprint image database that The Hong Kong Polytechnic University provides, and database has 500
A sample, each sample have 12 width images, database to have passed through area-of-interest (ROI) extraction, and image size is 128*128.
Near-infrared palm image palmmprint Enhancement Method flow is shown in Fig. 1, comprises the steps of：
Step 1：Picture I to be reinforced is divided into the fritter of R*R, calculates the mean value of each fritter grey scale pixel value, is tieed up
Spend the background matrix I to become smallerback。
Step 2：To the background matrix I obtained after piecemeal processingbackCarry out bicubic interpolation, make background matrix dimension and
The dimension of picture matrix I to be reinforced is identical, and background matrix is denoted as I' at this timeback；
Step 3：I pictures to be reinforced are subtracted into bicubic interpolation treated background matrix I'back, to I-I'backIt carries out straight
Side's figure equalization, the palmmprint structure I enhanced0。
Step 4：The palmprint image that step 3 is obtained input picture the most is blurred it using membership function, obtains
Obscure the image I after spendingfuzzy；
Step 5：Using fuzzy unsharp algorithm to Ienhance=I0-μIfuzzyProcessing；Obtain enhanced palmmprint structure.
Near-infrared palm image vena metacarpea Enhancement Method flow is shown in Fig. 2, comprises the steps of：
Step 1：The palm image of input is filtered using bootstrap filtering, takes the right r of filter radius and regular
Change factor lambda=0.01, obtains filtered result q1；
Step 2：By q1As the navigational figure and input picture of guiding filtering, r=16 and λ=0.01 are chosen, is carried out certainly
After guiding filtering processing, the image q after smooth palmmprint is obtained2；
Step 3：By linearly enhancing model I'=(q1-q2)·t+q2Vena metacarpea structure is enhanced, after obtaining enhancing
Vena metacarpea image I'；
Step 4：It repeats step 1 and step 2 step 3, further enhances vena metacarpea structure.At this point, output image is to slap
The enhanced image of vein.
The entire process of step (1) (2) (3) is known as one-step boot image filtering details enhancing process, wherein r takes in (2)
Be worth it is larger be in order to filter away more details, i.e. vein grain details, then again again multiply amplify these details be added to guiding filtering
As a result q2On do the enhancing of finger vena structure.The present invention has carried out the details enhancing process of navigational figure filtering twice, by Fig. 4
As can be seen that enhanced image keeps and highlights the vein grain details of artwork twice.
Near-infrared palm image palmmprint vena metacarpea fusion method flow is shown in Fig. 3, comprises the steps of：
Step 1：The near-infrared palm image of input is passed through into Fig. 1, Fig. 2 flow operations obtain enhanced palmmprint structure
IprintWith vena metacarpea structure Ivein；
Step 2：The average image quality for calculating image, obtains the identification weights of palmmprint, vena metacarpea；
Step 3：The two level wavelet character of palmmprint, vena metacarpea is extracted respectively；
Step 4：The weights that the similarity factor and step 2 obtained by training obtains respectively obtain palmmprint, vena metacarpea weighting
Similarity factor afterwards；
Step 5：There is step 4 to obtain the weighting similarity factor R after the fusion of palmmprint vena metacarpeafusion；
Training stage is to be trained according to above-mentioned steps 1 to 5 to the 6 width images of 500 samples of database, takes similar
The minimum value of coefficient is recognition threshold, i.e. RthresholD=min (R's(i)), (i=1,2,3...).The 6 of other 500 samples
Width image is as identification image, when identification, according to the weighting similarity factor R of the sample acquisition of identifications'≥RthresholdIt is determined as
Correct identification.
Fig. 4 respectively shows the enhanced palmmprint structure obtained by above-mentioned flow chart 1,2 and vena metacarpea structure.It can be with
See high-visible from the palmmprint of one infrared palm image zooming-out and the Edge texture feature of vena metacarpea image.
In order to verify improved single width near-infrared palmmprint vena metacarpea fusion identification method proposed by the invention compared to existing
The superiority of some single width palmmprint vena metacarpea fusion identification methods, in the near-infrared palmprint image data that The Hong Kong Polytechnic University provides
It is tested on library.Discrimination after being enhanced respectively after Palm-print Recognizing Rate comparison, the comparison of vena metacarpea discrimination and fusion
Comparison.Experimental result is as shown in Tables 1 and 2.
The Palm-print Recognizing Rate of table 1 different piecemeal radiuses and enhancing coefficient
The vena metacarpea discrimination of table 2 different filter windows and enhancing coefficient
3 palmmprint vena metacarpea fusion recognition rate of table
Known by table 1：In R=9, μ=50, the discrimination of palmmprint reaches peak, value 95.68%, than existing
Single width palmmprint vena metacarpea fusion recognition algorithm in Palm-print Recognizing Rate 94.00% be higher by 1.68%.By 2 data of table it is found that
The vena metacarpea of the present invention enhances algorithm in filter window (r1=2, r2=16) optimal identification rate 99.28% is obtained, than existing list
Palm-print Recognizing Rate 99.23% in width palmmprint vena metacarpea fusion recognition algorithm；Since parameter is in R=9, μ=50 and (r1=2, r2
=16, t=5) under the conditions of, palmmprint, vena metacarpea discrimination reach highest, so palmmprint vena metacarpea fusion is carried out with this condition,
Discrimination after fusion is 99.81%.It is higher than existing single width palmmprint vena metacarpea fusion recognition rate 99.69%.This illustrates this hair
It is bright to the original improved validity of single width palmmprint vena metacarpea recognizer.
In addition, in order to evaluate the time performance of fusion recognition algorithm of the present invention, experimental record chooses 2000 from sample
Width image is identified and calculates average recognition time.Present invention experiment carries out on Matlab R2014a, allocation of computer
For Core i52.50GHz CPU/4GB RAM.By table 3, it can be seen that, fusion recognition algorithm time performance of the invention is more good
It is good, there is actual application value.
The 4 palmmprint vena metacarpea adaptive weight fusion recognition time of table
The above is only the preferred embodiments of the present invention, not does any type of limitation to the present invention.It is every according to
According to the technology and methods essence of the present invention to any simple modification, equivalent change and modification made by above example, still fall within
In the range of the technology and methods scheme of the present invention.
Claims (6)
1. a kind of single width near-infrared palm image-recognizing method based on multi-modal fusion, it is characterised in that include the following steps：
Step 1：It inputs original near-infrared palm image and picture size is normalized；
Step 2：To treated, image removes vena metacarpea information simultaneously using the texture structure of piecemeal enhancing model extraction palmmprint；
Step 3：Using the membership function of definition, the palmmprint structure that step 2 is extracted is carried out obscuring unsharp increased
Strong palmprint image；
Step 4：It is then quiet to slapping to step 1 treated near-infrared palm image using bootstrap filtering removal palm print information
Arteries and veins information carries out the vena metacarpea image that adaptive-filtering is enhanced；
Step 5：Palmprint image and vena metacarpea image that step 3,4 obtain are carried out feature adaptively to merge, obtain similarity factor；
Step 6：The near-infrared palm image for choosing database is trained according to step 1 to 5, the similar system after being weighted
Number, and be recognition threshold by the minimum value for weighting similarity factor, the weighting similarity factor and training sample to the sample of identification
Threshold value is compared, if the threshold value of weighting similarity factor >=training sample of the sample of identification, identification are correct.
2. single width near-infrared palm image-recognizing method according to claim 1, it is characterised in that：The step 2 is specific
Using following steps：
Step 2.1：Near-infrared palm image I after normalized is divided into the fritter of R*R, calculates each fritter pixel ash
The mean value of angle value obtains the background matrix I that dimension becomes smallerback；
Step 2.2：To the background matrix I obtained after piecemeal processingbackBicubic interpolation is carried out, the dimension of background matrix and close is made
The dimension of infrared palm image array I is identical, and background matrix is denoted as I' at this timeback；
Step 2.3：Near-infrared palm image array I is subtracted into bicubic interpolation treated background matrix I'back, to I-I'back
Carry out histogram equalization, the palmmprint structural images enhanced.
3. single width near-infrared palm image-recognizing method according to claim 1, it is characterised in that：The step 3 is specific
Using following steps：
Step 3.1:The membership function of definition is as follows：
Parameter after optimization is as follows：
Wherein, δ is the standard deviation of domain X, and m is the mean value of domain X.C1, c2, c3, c4 are four parameters, this four values be by
What statistical value determined；
Step 3.1：Using step 2 treated palmprint image as input picture I0；
Step 3.2：To input picture I0Carry out Ienhance=I0-μIfuzzyProcessing, obtains enhanced palmmprint structural images；
Wherein, IfuzzyFor I0Image array after membership function is blurred, μ are details enhancing coefficients.
4. single width near-infrared palm image-recognizing method according to claim 1, it is characterised in that：The step 4 is specific
Using following steps：
Step 4.1：The palm image I of input is filtered using bootstrap filtering, takes filter radius r=2 and regular
Change factor lambda=0.01, obtains filtered result q1；
Step 4.2：By q1As the navigational figure and input picture of guiding filtering, r=16 and λ=0.01 are chosen, carries out bootstrap
After being filtered, the image q after smooth palmmprint is obtained2；
Step 4.3：By linearly enhancing model I'=(q1-q2)·t+q2Vena metacarpea structure is enhanced, is obtained enhanced
Vena metacarpea image I', the linear coefficient t=5 enhanced in model；
Step 4.4：Step 4.1 is repeated to 4.3, further enhances vena metacarpea structure, output image is that vena metacarpea is enhanced
Image.
5. single width near-infrared palm image-recognizing method according to claim 1, it is characterised in that：The step 5 is specific
Using following steps：
Step 5.1：Extract the two level 2-d wavelet feature of palmmprint enhancing image；
Step 5.2：Extract the two level 2-d wavelet feature of vena metacarpea enhancing image；
Step 5.3：Enhanced palmmprint and vena metacarpea feature are adaptively merged, similarity factor is obtained using formula 3：
Wherein A, B are two matrixes that size is M*N.
6. single width near-infrared palm image-recognizing method according to claim 1, it is characterised in that：Phase in the step 6
It is obtained using formula 4 like coefficient：
Rs' (i)=w1(i)Rvein(i)+(1-w1(i))Rprint(i) (4)
Wherein Rvein(i) it is the minimum value of the similarity factor of vena metacarpea in training sample；Rprint(i) it is palmmprint in training sample
Similarity factor minimum value；w1It is the weights of vena metacarpea similarity factor, is obtained using formula 5：
w1=1/3*LV1/(LV1+LP1)+1/3*LV2/(LV2+LP2)+1/3*LV3/(LV3+LP3) (5)
G is database sample, i ∈ (1,2,3 ..., G)；LV1,LV2,LV3, it is three indexs for weighing vena metacarpea picture quality；
LP1,LP2,LP3It is three indexs for weighing palmprint image quality.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810324828.0A CN108564031A (en) | 2018-04-12 | 2018-04-12 | Single width near-infrared palm image-recognizing method based on multi-modal fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810324828.0A CN108564031A (en) | 2018-04-12 | 2018-04-12 | Single width near-infrared palm image-recognizing method based on multi-modal fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108564031A true CN108564031A (en) | 2018-09-21 |
Family
ID=63534726
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810324828.0A Pending CN108564031A (en) | 2018-04-12 | 2018-04-12 | Single width near-infrared palm image-recognizing method based on multi-modal fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108564031A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110163123A (en) * | 2019-04-30 | 2019-08-23 | 杭州电子科技大学 | One kind referring to vein fusion identification method based on single width near-infrared finger-image fingerprint |
| CN112241721A (en) * | 2020-10-27 | 2021-01-19 | 上海掌腾信息科技有限公司 | Method for realizing complete palm vein image acquisition based on light irradiation of different wavelengths |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120194662A1 (en) * | 2011-01-28 | 2012-08-02 | The Hong Kong Polytechnic University | Method and system for multispectral palmprint verification |
| CN104318213A (en) * | 2014-10-21 | 2015-01-28 | 沈阳大学 | Method for using human body palm biology information to identify identities |
| CN104951774A (en) * | 2015-07-10 | 2015-09-30 | 浙江工业大学 | Palm vein feature extracting and matching method based on integration of two sub-spaces |
| CN106022218A (en) * | 2016-05-06 | 2016-10-12 | 浙江工业大学 | Palm print palm vein image layer fusion method based on wavelet transformation and Gabor filter |
| CN107195124A (en) * | 2017-07-20 | 2017-09-22 | 长江大学 | The self-service book borrowing method in library and system based on palmmprint and vena metacarpea |
-
2018
- 2018-04-12 CN CN201810324828.0A patent/CN108564031A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120194662A1 (en) * | 2011-01-28 | 2012-08-02 | The Hong Kong Polytechnic University | Method and system for multispectral palmprint verification |
| CN104318213A (en) * | 2014-10-21 | 2015-01-28 | 沈阳大学 | Method for using human body palm biology information to identify identities |
| CN104951774A (en) * | 2015-07-10 | 2015-09-30 | 浙江工业大学 | Palm vein feature extracting and matching method based on integration of two sub-spaces |
| CN106022218A (en) * | 2016-05-06 | 2016-10-12 | 浙江工业大学 | Palm print palm vein image layer fusion method based on wavelet transformation and Gabor filter |
| CN106022218B (en) * | 2016-05-06 | 2019-07-05 | 浙江工业大学 | A kind of palmmprint vena metacarpea image layer fusion method based on wavelet transformation and Gabor filter |
| CN107195124A (en) * | 2017-07-20 | 2017-09-22 | 长江大学 | The self-service book borrowing method in library and system based on palmmprint and vena metacarpea |
Non-Patent Citations (2)
| Title |
|---|
| 中国优秀硕士学位论文全文数据库 信息科技辑: "基于单幅近红外手掌图像的掌静脉和掌纹融合识别方法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 王艳霞 等: "一种新的掌纹图像预处理方法", 《中国图像图形学报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110163123A (en) * | 2019-04-30 | 2019-08-23 | 杭州电子科技大学 | One kind referring to vein fusion identification method based on single width near-infrared finger-image fingerprint |
| CN110163123B (en) * | 2019-04-30 | 2021-02-26 | 杭州电子科技大学 | Fingerprint finger vein fusion identification method based on single near-infrared finger image |
| CN112241721A (en) * | 2020-10-27 | 2021-01-19 | 上海掌腾信息科技有限公司 | Method for realizing complete palm vein image acquisition based on light irradiation of different wavelengths |
| CN112241721B (en) * | 2020-10-27 | 2022-07-05 | 上海掌腾信息科技有限公司 | Method for realizing complete palm vein image acquisition based on light irradiation of different wavelengths |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107729820B (en) | Finger vein identification method based on multi-scale HOG | |
| Yang et al. | Finger-vein segmentation based on multi-channel even-symmetric Gabor filters | |
| CN105426843B (en) | The single-lens lower vena metacarpea of one kind and palmprint image collecting device and image enhancement and dividing method | |
| Banerjee et al. | ARTeM: A new system for human authentication using finger vein images | |
| CN106548176B (en) | Finger vein image enhancement method based on self-adaptive guide filtering | |
| Trabelsi et al. | A new multimodal biometric system based on finger vein and hand vein recognition | |
| CN112507804A (en) | Identity recognition method based on hand image | |
| WO2021027155A1 (en) | Verification method and apparatus based on finger vein image, and storage medium and computer device | |
| CN108564031A (en) | Single width near-infrared palm image-recognizing method based on multi-modal fusion | |
| Albadarneh et al. | Iris recognition system for secure authentication based on texture and shape features | |
| Lei et al. | A sparse representation denoising algorithm for finger-vein image based on dictionary learning | |
| Achban et al. | Wrist hand vein recognition using local line binary pattern (LLBP) | |
| Bharath et al. | A novel multimodal hand database for biometric authentication | |
| Wulandari et al. | On the performance of pretrained CNN aimed at palm vein recognition application | |
| Chen et al. | Adapted deep convnets technology for robust iris recognition | |
| Kumar et al. | Finger Vein based Human Identification and Recognition using Gabor Filter | |
| Bashar et al. | Texture based vein biometrics for human identification: A comparative study | |
| Kosmala et al. | Human identification by vascular patterns | |
| Thenmozhi et al. | Comparative analysis of finger vein pattern feature extraction techniques: An overview | |
| Babatunde et al. | A multi-level model for fingerprint image enhancement | |
| Hariprasath et al. | Bimodal biometric pattern recognition system based on fusion of iris and palmprint using multi-resolution approach | |
| Pooja | Biometric Security: Palm Vein Recognition Using Lbp and Sift | |
| CN116778538B (en) | Vein image recognition method and system based on wavelet decomposition | |
| Alhanjouri | Improved HMM by deep learning for ear classification | |
| Massihi et al. | Extracting features from wrist vein images using fractional fourier transform for person verification |
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: 20180921 |
|
| RJ01 | Rejection of invention patent application after publication |