CN110348289A - A kind of finger vein identification method based on binary map - Google Patents
A kind of finger vein identification method based on binary map Download PDFInfo
- Publication number
- CN110348289A CN110348289A CN201910446068.5A CN201910446068A CN110348289A CN 110348289 A CN110348289 A CN 110348289A CN 201910446068 A CN201910446068 A CN 201910446068A CN 110348289 A CN110348289 A CN 110348289A
- Authority
- CN
- China
- Prior art keywords
- finger
- binary map
- characteristic point
- image
- matching
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Collating Specific Patterns (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The present invention relates to a kind of finger vein identification methods based on binary map, comprising: the finger-image under acquisition near infrared light；The finger-image of acquisition is pre-processed, to obtain corresponding finger vena binary map；Feature point extraction is carried out to finger vena binary map using FAST algorithm；Vectorization description is carried out using characteristic point of the histogram of gradients to extraction；Based on the characteristic point of vectorization description, the matching distance in finger vena binary map and training library between all registered images is calculated；By comparing the matching distance of finger vena binary map and all registered images, recognition result is obtained.Object of the present invention using the vein texture binary map after Threshold segmentation as feature point extraction, the interference of non-vein factor can utmostly be eliminated, and algorithm design is optimized in itself, further increases the efficiency of finger vena identification and the accuracy rate of result identification.
Description
Technical field
The present invention relates to image procossings and technical field of biometric identification, quiet more particularly, to a kind of finger based on binary map
Arteries and veins recognition methods.
Background technique
Existing finger hand vein recognition algorithm is roughly divided into the method based on details and characteristic point, the side based on local mode
Method, and the method based on texture network.Wherein, the method based on details and characteristic point combines latter two to a certain extent
The advantage of method, development prospect are preferable.Method based on details and characteristic point, it is prominent as far as possible by handling image
Vein texture part out eliminates the influence of the factors such as non-vein structure, then extracts satisfactory feature in venous structures again
Point further eliminates the interference of non-vein structure.
However, inventor has found under study for action, the existing vein identification method based on details and characteristic point, mainly with
Comprising vein distribution grayscale image carry out object carry out algorithm design, due to the limitation of acquisition device, intensity of illumination not really
The presence of the factors such as qualitative and finger tissues surrounding vascular complexity, leads to ash obtained in subsequent image treatment process
Degree figure still remains irregular shade and non-vein feature, this meeting is not so that the vein pattern being extracted has good representative
Property and distinction, to reduce the accuracy of recognition result.In addition, algorithm design itself also it is in need improvement and optimization
Space
Summary of the invention
In view of this, it is necessary to for above-mentioned problem, a kind of finger vein identification method based on binary map is provided, with
Object of the vein texture binary map as feature point extraction after Threshold segmentation can utmostly eliminate the dry of non-vein factor
It disturbs, and algorithm design is optimized in itself, further increase the efficiency of finger vena identification and the accuracy rate of result identification.
A kind of finger vein identification method based on binary map, comprising:
Acquire the finger-image under near infrared light；
The finger-image of acquisition is pre-processed, to obtain corresponding finger vena binary map；
Feature point extraction is carried out to finger vena binary map using FAST algorithm；
Vectorization description is carried out using characteristic point of the histogram of gradients to extraction；
Based on the characteristic point of vectorization description, calculate in finger vena binary map and training library between all registered images
Matching distance；
By comparing the matching distance of finger vena binary map and all registered images, recognition result is obtained.
The finger-image of described pair of acquisition carries out pretreated step, comprising:
Picture size normalization, ROI extraction, image space domain and frequency domain enhancing and image threshold are carried out to finger-image
It is worth the processing of segmentation.
The ROI is extracted
Using Sobel operator longitudinal direction convolution coarse extraction finger edge, non-maxima suppression is carried out then along gradient direction,
Finally contour edge is fitted using RANSAC.
The step of described image spatial domain and frequency domain enhance, comprising:
Several sub-blocks are divided the image into using CLAHE, histogram equalization is carried out in each sub-block, if some sub-block
Histogram amplitude be greater than preset value, then it is cut and is evenly distributed in entire gray scale interval；And
Building withFor interval, from 0 toThe Gabor filter in totally 8 directions, is respectively filtered image,
Screening is compared to 8 obtained responses, retains the peak response of texture part in each result.
The step of described image Threshold segmentation, comprising:
Threshold segmentation is carried out using NiBlack algorithm, the pixel value that will be greater than threshold value is set as 255, less than the pixel of threshold value
Value is set as 0.
It is described that the step of vectorization describes is carried out using characteristic point of the histogram of gradients to extraction, comprising:
Using characteristic point as the center of circle, diameter is the neighbourhood circle of 7 pixels for building, calculates the gradient value of pixel and side in neighbourhood circle
To if the coordinate of pixel P is (x, y), the calculation formula of the mould m (x, y) and direction θ (x, y) of gradient are as follows:
A direction histogram is constructed, horizontal axis is the angular dimension of gradient direction, and gradient direction range is 0 to 360 degree,
Every 10 degree are chosen as a bins, is divided into 36 bins, the longitudinal axis is that gradient direction corresponds to the cumulative of gradient magnitude, direction Histogram
The peak value of figure represents the principal direction of characteristic point；
Centered on characteristic point, reference axis is nearby rotated into angle, θ in field, i.e., reference axis is rotated to be into characteristic point
Principal direction；
16 × 16 window is taken centered on characteristic point, and is divided into 16 4 × 4 fritters, and same building direction is straight
Fang Tu, but with every 45 degree for a bins, fritter each in this way has the gradient intensity information in 8 directions, thus each characteristic point
Available 128 dimensional feature description vectors.
The step of matching distance calculated between finger vena binary map and registered images, comprising:
Use dijIndicate special as j-th in ith feature point in the finger vena binary map of test image and registered images
Euclidean distance between sign point,
Wherein i ∈ [1, N1],j∈[1,N2]；
For the characteristic point i (x in test image1,y1), building is in training image with identical coordinate points (x1,y1) be
The center of circle, using r as the circle shaped neighborhood region O of radiusiIf calculating separately spy there are m characteristic point in the circle shaped neighborhood region in training image
The Euclidean distance in point i and training image between this m characteristic point is levied, if wherein minimum Eustachian distance is dip, then dipFor part
Optimum Matching；
To all non-circular neighborhood O in training libraryiInterior (N2- m) a characteristic point is calculated, if wherein minimum Euclidean
Distance is diq, then diqIt is matched for global optimum, if dip=diq, then it is assumed that in the characteristic point i and training image in test image
Characteristic point p is correct matching pair, if dip≠diq, then it is assumed that it is no opposite in training image for the characteristic point i of test image
The match point answered；
The matching distance between two images is defined,
Wherein, n indicates that correct matching logarithm, N indicate that the characteristic point sum of test image, d (i) indicate i-th pair correct
Euclidean distance between matching.
The matching distance by comparing finger vena binary map and all registered images, obtains the step of recognition result
Suddenly, comprising:
Matching distance between registered images all in finger vena binary map and database is ranked up, matching distance
Classification of the classification of the smallest registered images as recognition result identifies mistake if the minimal matching span is greater than preset threshold
It loses.
A kind of finger vein identification method based on binary map provided by the invention, with the vein texture two after Threshold segmentation
Object of the value figure as feature point extraction can utmostly eliminate the interference of non-vein factor, and to algorithm design itself
It optimizes, further increases the efficiency of finger vena identification and the accuracy rate of result identification.
Detailed description of the invention
Fig. 1 (a) is the original image schematic diagram of finger-image；
Fig. 1 (b) is the profile coarse extraction schematic diagram of finger-image；
Fig. 1 (c) is the ROI image schematic diagram of finger-image；
Fig. 1 (d) is that the CLAHE of finger-image enhances schematic diagram；
Fig. 1 (e) is that the Gabor of finger-image enhances schematic diagram；
Fig. 1 (f) is finger vena binary map schematic diagram；
Fig. 2 is the binary map schematic diagram by feature point extraction；
Fig. 3 is the Receiver operating curve in test scene of the present invention；
Fig. 4 is the matching distance distribution map in an application scenarios.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, carries out clearly and completely to the technical solution in example of the present invention
Description, described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.
The specific implementation steps are as follows for finger vein identification method of the present invention based on binary map:
Step S1: finger-image is acquired under the conditions of near-infrared.Schematic diagram such as Fig. 1 (a).
Step S2: using Sobel operator (the primary operator of rope) longitudinal convolution coarse extraction finger edge, then along gradient direction
Non-maxima suppression is carried out, is finally fitted contour edge using RANSAC (RANSAC algorithm).This step realizes ROI
(Region of Interest, area-of-interest) extracts.
Step S3: carrying out spatial domain and frequency domain respectively to finger vein image enhances, and steps are as follows:
Divide the image into several sub-blocks using CLAHE (limitation contrast histogram equalization method), in each sub-block into
Column hisgram equalization cuts it and is evenly distributed in if the histogram amplitude of some sub-block is greater than preset value
In entire gray scale interval.
Building withFor interval, from 0 toThe Gabor filter (sweet rich filter) in totally 8 directions, respectively to picture
It is filtered, screening is compared to 8 obtained responses, retains the peak response of texture part in each result.
Step S4: Threshold segmentation is carried out using NiBlack algorithm (Niblack algorithm), the pixel value that will be greater than threshold value is set
It is 255, the pixel value less than threshold value is set as 0.
Step S5: it is detected using FAST (extraction of quick segmentation test feature point) algorithm and is met the requirements on vein texture edge
Pixel as target feature point.Schematic diagram such as Fig. 2.
Step S6: vectorization description is carried out to the characteristic point extracted, steps are as follows:
Using characteristic point as the center of circle, diameter is the neighbourhood circle of 7 pixels for building, calculates the gradient value of pixel and side in neighbourhood circle
To.If the coordinate of pixel P is (x, y), the calculation formula of the mould m (x, y) and direction θ (x, y) of gradient are as follows:
Construct a direction histogram, horizontal axis be gradient direction angular dimension (gradient direction range is 0 to 360 degree,
Every 10 degree are chosen as a bins, is divided into 36 bins), the longitudinal axis is that gradient direction corresponds to the cumulative of gradient magnitude.Direction is straight
The peak value of square figure represents the principal direction of characteristic point.
In order to keep the rotational invariance of description vectors, nearby reference axis to be revolved in field centered on characteristic point
Turn θ (principal direction of characteristic point) angle, i.e., reference axis is rotated to be to the principal direction of characteristic point.
16 × 16 window is taken centered on characteristic point, and is divided into 16 4 × 4 fritters, and same building direction is straight
Fang Tu, but with every 45 degree for a bins, fritter each in this way has the gradient intensity information in 8 directions.Therefore final each spy
The feature description vectors of available 128 dimension of sign point.
Step S7: the matching distance between two images is calculated, d is usedijIndicate ith feature point and training in test image
Euclidean distance in image between j-th of characteristic point,
Wherein i ∈ [1, N1],j∈[1,N2]。
For the characteristic point i (x in test image1,y1), building is in training image with identical coordinate points (x1,y1) be
The center of circle, using r as the circle shaped neighborhood region of radius, if calculating separately feature there are m characteristic point in the circle shaped neighborhood region in training image
Euclidean distance in point i and training image between this m characteristic point, if wherein minimum Eustachian distance is dip, then dipMost for part
Excellent matching.
To all non-circular neighborhood O in training libraryiInterior (N2- m) a characteristic point is calculated, if wherein minimum Euclidean
Distance is diq, then diqFor global optimum's matching.If dip=diq, then it is assumed that in the characteristic point i and training image in test image
Characteristic point p is correct matching pair.If dip≠diq, then it is assumed that it is no opposite in training image for the characteristic point i of test image
The match point answered.
Matching distance between two images, is defined as follows:
Wherein, n indicates that correct matching logarithm, N indicate that the characteristic point sum of test image, d (i) indicate i-th pair correct
Euclidean distance between matching.
Step S8: the matching distance between registered images all in input picture and database is ranked up, matching away from
Classification from the smallest finger classification as input picture, if the matching distance is greater than preset threshold, recognition failures.
Referring to Fig. 3 and Fig. 4, in test scene, verified for finger vena recognizer of the invention, experiment is adopted
It is tested with Shandong University's machine learning and database disclosed in data mining laboratory.636 class hands are shared in the database
Refer to, each finger there are 6 width figures, shares 3816 width images, picture size 320x240.Experiment is divided into recognition mode and verifying mould
Formula, under recognition mode, every one kind finger randomly selects piece image as test image, the remaining 5 width figures of every one kind finger
It as composition template database image, tests ten times altogether, average recognition rate 99.3%, being averaged most when reaching 100% discrimination
Low sequence is 14.7.Under Validation Mode, using full matching in the matching of source, 500 class hands in database are chosen in homologous matching
The test of fingering row.Therefore on database, every piece image all carries out heterologous matching primitives with 6 width images of other 635 fingers,
The correspondingly available a heterologous matching distance in 14538960 (635 × 6 × 636 × 6).Simultaneously every piece image all with it is similar
Other 5 width image carries out homologous matching primitives, it is hereby achieved that a homologous matching distance of 15000 (500 × A_6^2).Experiment
The result shows that EER (waiting accidentally rate) is 0.0196.As can be seen that the index of correlation under recognition mode and Validation Mode all achieve compared with
It is good as a result, finger vein identification method practical value with higher provided by the invention.
The embodiments described above only express several embodiments of the present invention, and the description thereof is more specific and detailed, but simultaneously
Limitations on the scope of the patent of the present invention therefore cannot be interpreted as.It should be pointed out that for those of ordinary skill in the art
For, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to guarantor of the invention
Protect range.Therefore, the scope of protection of the patent of the invention shall be subject to the appended claims.
Claims (8)
1. a kind of finger vein identification method based on binary map characterized by comprising
Acquire the finger-image under near infrared light；
The finger-image of acquisition is pre-processed, to obtain corresponding finger vena binary map；
Feature point extraction is carried out to finger vena binary map using FAST algorithm；
Vectorization description is carried out using characteristic point of the histogram of gradients to extraction；
Based on the characteristic point of vectorization description, the matching in finger vena binary map and training library between all registered images is calculated
Distance；
By comparing the matching distance of finger vena binary map and all registered images, recognition result is obtained.
2. the finger vein identification method according to claim 1 based on binary map, which is characterized in that described pair acquisition
Finger-image carries out pretreated step, comprising:
Picture size normalization, ROI extraction, image space domain and frequency domain enhancing and image threshold point are carried out to finger-image
The processing cut.
3. the finger vein identification method according to claim 2 based on binary map, which is characterized in that the ROI is extracted
Include:
Using Sobel operator longitudinal direction convolution coarse extraction finger edge, non-maxima suppression is carried out then along gradient direction, finally
Contour edge is fitted using RANSAC.
4. the finger vein identification method according to claim 2 based on binary map, which is characterized in that described image space
The step of domain and frequency domain enhance, comprising:
Several sub-blocks are divided the image into using CLAHE, histogram equalization is carried out in each sub-block, if some sub-block is straight
Square map sheet degree is greater than preset value, then is cut and be evenly distributed in entire gray scale interval to it；And
Building withFor interval, from 0 toThe Gabor filter in totally 8 directions, is respectively filtered image, to
To 8 responses screening is compared, retain the peak response of texture part in each result.
5. the finger vein identification method according to claim 2 based on binary map, which is characterized in that described image threshold value
The step of segmentation, comprising:
Threshold segmentation is carried out using NiBlack algorithm, the pixel value that will be greater than threshold value is set as 255, and the pixel value less than threshold value is set
It is 0.
6. according to the described in any item finger vein identification methods based on binary map of claim 3 to 5, which is characterized in that institute
It states and the step of vectorization describes is carried out using characteristic point of the histogram of gradients to extraction, comprising:
Using characteristic point as the center of circle, diameter is the neighbourhood circle of 7 pixels for building, calculates the gradient value of pixel and direction in neighbourhood circle,
If the coordinate of pixel P is (x, y), the calculation formula of the mould m (x, y) and direction θ (x, y) of gradient are as follows:
A direction histogram is constructed, horizontal axis is the angular dimension of gradient direction, and gradient direction range is 0 to 360 degree, is chosen
Every 10 degree are a bins, are divided into 36 bins, and the longitudinal axis is that gradient direction corresponds to the cumulative of gradient magnitude, direction histogram
Peak value represents the principal direction of characteristic point；
Centered on characteristic point, reference axis is nearby rotated into angle, θ in field, i.e., reference axis is rotated to be to the main side of characteristic point
To；
16 × 16 window is taken centered on characteristic point, and is divided into 16 4 × 4 fritters, equally building direction histogram,
But with every 45 degree for a bins, fritter each in this way has the gradient intensity information in 8 directions, and thus each characteristic point can obtain
To 128 dimensional feature description vectors.
7. the finger vein identification method according to claim 6 based on binary map, which is characterized in that the calculating finger
The step of matching distance between vein binary map and registered images, comprising:
Use dijIt indicates as j-th of characteristic point in ith feature point in the finger vena binary map of test image and registered images
Between Euclidean distance,
Wherein i ∈ [1, N1],j∈[1,N2]；
For the characteristic point i (x in test image1,y1), building is in training image with identical coordinate points (x1,y1) it is the center of circle,
Using r as the circle shaped neighborhood region O of radiusiIf there are m characteristic points in the circle shaped neighborhood region in training image, characteristic point i is calculated separately
With the Euclidean distance in training image between this m characteristic point, if wherein minimum Eustachian distance be dip, then dipFor local optimum
Matching；
To all non-circular neighborhood O in training libraryiInterior (N2- m) a characteristic point calculated, if wherein minimum Eustachian distance
For diq, then diqIt is matched for global optimum, if dip=diq, then it is assumed that the characteristic point i in test image and the feature in training image
Point p is correct matching pair, if dip≠diq, then it is assumed that it is not corresponding in training image for the characteristic point i of test image
Match point；
The matching distance between two images is defined,
Wherein, n indicates that correct matching logarithm, N indicate that the characteristic point sum of test image, d (i) indicate that i-th pair correctly matches it
Between Euclidean distance.
8. the finger vein identification method according to claim 7 based on binary map, which is characterized in that it is described by comparing
The matching distance of finger vena binary map and all registered images, the step of obtaining recognition result, comprising:
Matching distance between registered images all in finger vena binary map and database is ranked up, matching distance is minimum
Registered images classification of the classification as recognition result, if the minimal matching span is greater than preset threshold, recognition failures.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910446068.5A CN110348289B (en) | 2019-05-27 | 2019-05-27 | Finger vein identification method based on binary image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910446068.5A CN110348289B (en) | 2019-05-27 | 2019-05-27 | Finger vein identification method based on binary image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110348289A true CN110348289A (en) | 2019-10-18 |
| CN110348289B CN110348289B (en) | 2023-04-07 |
Family
ID=68173969
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910446068.5A Active CN110348289B (en) | 2019-05-27 | 2019-05-27 | Finger vein identification method based on binary image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110348289B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111222456A (en) * | 2020-01-04 | 2020-06-02 | 圣点世纪科技股份有限公司 | High-speed retrieval algorithm under condition of great finger vein user quantity |
| CN111797828A (en) * | 2020-06-19 | 2020-10-20 | 智慧眼科技股份有限公司 | Method and device for acquiring ROI (region of interest) of finger vein image and related equipment |
| CN112069906A (en) * | 2020-08-10 | 2020-12-11 | 浙江工业大学 | Traffic light identification method based on OpenCV and comprehensive matching distance |
| CN113128378A (en) * | 2021-04-06 | 2021-07-16 | 浙江精宏智能科技有限公司 | Quick finger vein identification method |
| CN116778172A (en) * | 2023-08-18 | 2023-09-19 | 江苏圣点世纪科技有限公司 | Finger back vein image enhancement method |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104688184A (en) * | 2014-12-05 | 2015-06-10 | 南京航空航天大学 | Vein imaging method for visible-light skin images |
| CN105426821A (en) * | 2015-11-04 | 2016-03-23 | 浙江工业大学 | Palm vein feature extracting and matching method based on eight neighborhood and secondary matching |
| CN106096569A (en) * | 2016-06-20 | 2016-11-09 | 广州中国科学院先进技术研究所 | A kind of finger vein identification method |
| WO2017020741A1 (en) * | 2015-08-06 | 2017-02-09 | 阿里巴巴集团控股有限公司 | Method, apparatus and system for retrieving image, acquiring image information and identifying image |
| CN107145829A (en) * | 2017-04-07 | 2017-09-08 | 电子科技大学 | A kind of vena metacarpea recognition methods for merging textural characteristics and scale invariant feature |
| WO2018032861A1 (en) * | 2016-08-17 | 2018-02-22 | 广州广电运通金融电子股份有限公司 | Finger vein recognition method and device |
| CN108182399A (en) * | 2017-12-27 | 2018-06-19 | 北京智慧眼科技股份有限公司 | Refer to vein characteristic comparison method, apparatus, storage medium and processor |
| CN109101867A (en) * | 2018-06-11 | 2018-12-28 | 平安科技（深圳）有限公司 | A kind of image matching method, device, computer equipment and storage medium |
-
2019
- 2019-05-27 CN CN201910446068.5A patent/CN110348289B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104688184A (en) * | 2014-12-05 | 2015-06-10 | 南京航空航天大学 | Vein imaging method for visible-light skin images |
| WO2017020741A1 (en) * | 2015-08-06 | 2017-02-09 | 阿里巴巴集团控股有限公司 | Method, apparatus and system for retrieving image, acquiring image information and identifying image |
| CN105426821A (en) * | 2015-11-04 | 2016-03-23 | 浙江工业大学 | Palm vein feature extracting and matching method based on eight neighborhood and secondary matching |
| CN106096569A (en) * | 2016-06-20 | 2016-11-09 | 广州中国科学院先进技术研究所 | A kind of finger vein identification method |
| WO2018032861A1 (en) * | 2016-08-17 | 2018-02-22 | 广州广电运通金融电子股份有限公司 | Finger vein recognition method and device |
| CN107145829A (en) * | 2017-04-07 | 2017-09-08 | 电子科技大学 | A kind of vena metacarpea recognition methods for merging textural characteristics and scale invariant feature |
| CN108182399A (en) * | 2017-12-27 | 2018-06-19 | 北京智慧眼科技股份有限公司 | Refer to vein characteristic comparison method, apparatus, storage medium and processor |
| CN109101867A (en) * | 2018-06-11 | 2018-12-28 | 平安科技（深圳）有限公司 | A kind of image matching method, device, computer equipment and storage medium |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111222456A (en) * | 2020-01-04 | 2020-06-02 | 圣点世纪科技股份有限公司 | High-speed retrieval algorithm under condition of great finger vein user quantity |
| CN111797828A (en) * | 2020-06-19 | 2020-10-20 | 智慧眼科技股份有限公司 | Method and device for acquiring ROI (region of interest) of finger vein image and related equipment |
| CN112069906A (en) * | 2020-08-10 | 2020-12-11 | 浙江工业大学 | Traffic light identification method based on OpenCV and comprehensive matching distance |
| CN112069906B (en) * | 2020-08-10 | 2024-02-20 | 浙江工业大学 | Traffic light identification method based on OpenCV and comprehensive matching distance |
| CN113128378A (en) * | 2021-04-06 | 2021-07-16 | 浙江精宏智能科技有限公司 | Quick finger vein identification method |
| CN113128378B (en) * | 2021-04-06 | 2022-07-19 | 浙江精宏智能科技有限公司 | Finger vein rapid identification method |
| CN116778172A (en) * | 2023-08-18 | 2023-09-19 | 江苏圣点世纪科技有限公司 | Finger back vein image enhancement method |
| CN116778172B (en) * | 2023-08-18 | 2023-11-07 | 江苏圣点世纪科技有限公司 | Finger back vein image enhancement method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110348289B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10789465B2 (en) | Feature extraction and matching for biometric authentication | |
| CN110348289A (en) | A kind of finger vein identification method based on binary map | |
| CN103632136B (en) | Human-eye positioning method and device | |
| CN101317183B (en) | Method for localizing pixels representing an iris in an image acquired of an eye | |
| CN101339607B (en) | Human face recognition method and system, human face recognition model training method and system | |
| TWI434220B (en) | A method for recognizing the identity of user by palm vein biometric | |
| CN104123543B (en) | A kind of eye movement recognition methods based on recognition of face | |
| CN107729820B (en) | Finger vein identification method based on multi-scale HOG | |
| CN111401145B (en) | Visible light iris recognition method based on deep learning and DS evidence theory | |
| CN105956578A (en) | Face verification method based on identity document information | |
| CN101317184A (en) | Method for extracting features of an iris in images | |
| CN103914676A (en) | Method and apparatus for use in face recognition | |
| KR101315646B1 (en) | Method and apparatus for finger vein pattern extraction using guided gabor filter | |
| CN108564092A (en) | Sunflower disease recognition method based on SIFT feature extraction algorithm | |
| Choudhary et al. | A survey: Feature extraction methods for iris recognition | |
| CN113378790B (en) | Viewpoint positioning method, apparatus, electronic device, and computer-readable storage medium | |
| Tiwari et al. | A palmprint based recognition system for smartphone | |
| Ng et al. | An effective segmentation method for iris recognition system | |
| CN107145820B (en) | Binocular positioning method based on HOG characteristics and FAST algorithm | |
| KR101450247B1 (en) | Method for authenticating based on finger vein using SIFT keypoint | |
| Rossant et al. | A robust iris identification system based on wavelet packet decomposition and local comparisons of the extracted signatures | |
| Kovoor et al. | Iris biometric recognition system employing canny operator | |
| Kharat et al. | Iris recognition: a review | |
| Rajput et al. | Iris biometric technique for person authentication based on fusion of radon and 2D multi-wavelet transform | |
| Maske et al. | A Review on Detection and Recognition of Text |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |