CN112801034A - Finger vein recognition device - Google Patents
Finger vein recognition device Download PDFInfo
- Publication number
- CN112801034A CN112801034A CN202110203770.6A CN202110203770A CN112801034A CN 112801034 A CN112801034 A CN 112801034A CN 202110203770 A CN202110203770 A CN 202110203770A CN 112801034 A CN112801034 A CN 112801034A
- Authority
- CN
- China
- Prior art keywords
- hemoglobin
- image
- image data
- feature point
- finger vein
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06V10/30—Noise filtering
-
- 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
- G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- 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
Abstract
According to the finger vein recognition device, the image sensor with the near-infrared sensing capability is used for capturing the image data of hemoglobin radiography in a finger, the image acquisition capability of the hemoglobin radiography is higher than that of a common sensor, a coarse model is established according to the characteristics of hemoglobin for evaluating the image quality level, an image meeting the requirements can be screened out better, and the condition that the low-level image is attempted to be restored all the time is avoided. Through carrying out the preliminary treatment to the image that accords with the quality requirement, can remove the noise, highlight the hemoglobin radiography, then extract the hemoglobin radiography according to preset hemoglobin characteristic, compare according to with preset characteristic point set.
Description
Technical Field
The invention belongs to the technical field of biological feature recognition, and particularly relates to a finger vein recognition device.
Background
Along with the application and popularization of smart devices in modern society, authentication and verification methods are gradually changed from early card verification and password verification to human body biometric verification, such as fingerprint and iris verification. With the use of various biometric feature recognition techniques, related drawbacks are gradually emerging. As the most common fingerprint identification technology, fingerprints are easily worn and stolen, and the authentication device can be easily cheated by using one finger print. The security problem of biometric information has been a major obstacle to the spread of biometric identification devices.
The advent and emergence of finger vein recognition technology provides a new way for solving the technical problem of biometric information verification security. Finger vein recognition is fundamentally different from conventional fingerprint recognition, and has many advantages, such as: the finger vein recognition is realized according to a vein image formed by near infrared rays reflected by hemoglobin in a vein, belongs to a living body recognition technology, is not influenced by external environments such as epidermis, pollution, temperature and humidity due to the fact that the vein belongs to internal physiological characteristics, cannot be copied, and mainly depends on living body advantages brought by different structures in fingers due to different absorptivity of infrared light, advantages brought by non-contact verification and stability advantages that the recognition rate of the finger is not influenced by external damage of the finger, and the like. Since the concept of finger vein recognition was originally proposed by Joseph Rice, the technology has been used in the fields of civilian, finance and transportation after more than 20 years of exploration and development. However, due to the image sensor performance of different devices, device lighting conditions may cause the vein vessel image to be blurred or noise may mask the elongated vein vessel image to cause interruption. In the prior art, the ideal situation of finger vein identification comparison is based on the same or close imaging conditions, low-quality image comparison is less suitable, and the images of finger vein vessels in natural environments or other imaging devices have great differences, so that comparison with stored results in a database is difficult.
Disclosure of Invention
In order to solve the above problems, the present invention provides the following technical solutions:
a finger vein identification device, comprising: a housing, a capture window disposed within the housing, a finger capture area disposed within the housing above the capture window, an image sensor having near infrared sensing capability disposed behind the capture window, and a processor circuit in electrical communication with the image sensor, the processor circuit configured to perform the steps of:
s1, acquiring multi-frame image data which are used for hemoglobin shadow formed by absorption of near infrared light by hemoglobin in flowing blood in the finger to be acquired and captured by an image sensor with near infrared sensing capability;
s2: performing quality evaluation on the captured image data by using a hemoglobin shadow rough model, confirming the quality grade, and inputting the image data of the hemoglobin shadow characteristics meeting the quality requirement into S3;
s3, preprocessing the captured hemoglobin photographic image data of the finger vein to obtain data of the features to be extracted, wherein the preprocessing further comprises the following steps:
s31, enhancing image data;
s32, denoising the enhanced image data and highlighting a hemoglobin photographic image;
s33, carrying out target detection and extraction on the hemoglobin photographic image;
s34, collecting images containing a certain amount of hemoglobin by using an edge detection algorithm;
s4: extracting a second feature point image in the set containing the hemoglobin;
s5: and comparing the extracted feature points with a pre-stored first feature point image, comparing the second feature point set with the pre-stored first feature point set, judging whether the second feature point set is matched with the first feature point set within a threshold range, and if so, successfully identifying and comparing.
In step S31, the image data is copied to create a copy, the copy is image-inverted, and then image smoothing is performed after the image inversion, and then the image data is inverted and converted to obtain enhanced image data.
In step S32, the enhanced image data is processed by blocking, the image data is divided into a plurality of block images according to preset conditions, the block images are clustered, and then the solution is performed by using an augmented lagrange multiplier method, so as to obtain a de-noised highlighted hemoglobin photographic image.
In the step S33, a Region-normal (R-CNN) algorithm based on Region propofol is adopted to generate a hemoglobin target candidate box for the hemoglobin photographic image, and then the candidate box is classified and regressed, and most of hemoglobin is used as a target for detection and extraction.
Compared with the prior art, the invention has the beneficial effects that:
1. according to the finger vein recognition device, the image sensor with the near-infrared sensing capability is used for capturing the image data of hemoglobin radiography in a finger, the image acquisition capability of the hemoglobin radiography is higher than that of a common sensor, a coarse model is established according to the characteristics of hemoglobin for evaluating the image quality level, an image meeting the requirements can be screened out better, and the condition that the low-level image is attempted to be restored all the time is avoided. Through carrying out the preliminary treatment to the image that accords with the quality requirement, can remove the noise, highlight the hemoglobin radiography, then extract according to the hemoglobin radiography of predetermined hemoglobin characteristic team, compare according to with predetermined characteristic point set.
2. The finger vein recognition device provided by the invention is used for copying image data to create a copy, carrying out image inversion on the copy, carrying out image smoothing treatment after inversion, and then carrying out inversion conversion to obtain enhanced image data.
3. According to the finger vein recognition device, the enhanced image data are processed in a blocking mode, the image data are divided into the block images according to the preset conditions and processed, the overall processing speed can be improved, large operation resources do not need to be occupied, the hemoglobin can be highlighted and the edges are smooth through clustering operation and the Lagrange multiplier method, and the accuracy of subsequent comparison can be improved.
4. According to the finger vein recognition device, the hemoglobin photo can be quickly selected from the image frame by adopting an R-CNN fine algorithm, the size of the pattern is reduced, and the proportion of the hemoglobin photo in the image is increased.
Drawings
Fig. 1 is a schematic view of an identification process of a finger vein identification apparatus according to the present invention;
FIG. 2 is a schematic diagram of a finger vein recognition apparatus according to the present invention;
FIG. 3 is a diagram illustrating a finger vein de-noised finger vein recognition apparatus according to the present invention;
fig. 4 is a schematic structural diagram of a finger vein recognition apparatus according to the present invention.
Reference numerals: 1-a shell; 2-collecting window; 3-finger capture area.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 4, a finger vein recognition apparatus includes: a housing 1, a capture window 2 disposed within the housing, a finger capture area 3 disposed within the housing 1 above the capture window 2, an image sensor having near infrared sensing capability disposed behind the capture window 2, and a processor circuit in electrical communication with the image sensor, the processor circuit configured to capture an image by the image sensor under illumination light after a finger is inserted into the capture area 3 proximate to the capture window 2, the processor circuit performing the following steps, as shown in fig. 1:
s1, acquiring multi-frame image data which are used for the finger to be collected and contain hemoglobin shadow formed by the absorption of the hemoglobin in the flowing blood to the near infrared light by an image sensor with the near infrared sensing capability; as shown in fig. 2, fig. 2 is image data in an ideal case.
S2: and (3) performing quality evaluation on the captured image data by using a hemoglobin shadow rough model, confirming the quality grade, inputting the image data of the hemoglobin shadow characteristic meeting the quality requirement into S3, rejecting the image data which do not meet the quality requirement, and stopping if the image data do not meet the quality requirement, wherein the basis of the quality grade evaluation is hemoglobin, and whether the thin blood vessels can be partially restored by noise or not. And establishing a coarse model according to the characteristics of the hemoglobin, such as shape, chroma and the like, for evaluating the image quality grade, if the coarse models are not matched, the frame of picture cannot be subjected to subsequent identification, and the image meeting the requirements can be better screened out by setting the coarse model.
S3, preprocessing the captured hemoglobin photographic image data of the finger vein to obtain data of the features to be extracted, wherein the preprocessing further comprises the following steps:
and S21, enhancing the image data. The contrast of the overall image is enhanced.
And S22, denoising the enhanced image data, wherein salt and pepper noises and random noises are mainly removed, and the hemoglobin photographic image is highlighted.
And S23, carrying out target detection and extraction on the hemoglobin photographic image. Target detection is performed based on the characteristics of hemoglobin and a target that is considered likely to be a hemoglobin contrast image is extracted.
S24, collecting images containing a certain amount of hemoglobin by using an edge detection algorithm. The edge detection is to identify hemoglobin with obvious light and dark contrast in the picture. After a series of collection, extraction and reduction, the product is shown in figure 3.
S3: and extracting a second feature point image in the set containing the hemoglobin, as shown in FIG. 3.
S4: and comparing the extracted feature points with a pre-stored first feature point image, comparing the second feature point set with the pre-stored first feature point set, judging whether the second feature point set is matched with the first feature point set within a threshold range, and if so, successfully identifying and comparing.
In step S21, the hemoglobin photographic image data is copied to create a copy, the copy is subjected to image inversion, noise and hemoglobin are strongly contrasted, image smoothing is performed after inversion, and then inversion conversion is performed to obtain enhanced image data.
In step S22, the enhanced image data is processed by blocking, the image data is divided into a plurality of block images according to preset conditions, the block images are clustered, and then the solution is performed by using an augmented lagrange multiplier method, so as to obtain a de-noised highlighted hemoglobin photographic image.
In the step S23, a Region-normal (R-CNN) algorithm based on Region propofol is adopted to generate a hemoglobin target candidate box for the hemoglobin photographic image, and then the candidate box is classified and regressed, and most of hemoglobin is used as a target for detection and extraction.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Claims (4)
1. A finger vein recognition apparatus, comprising: a housing (1), a capture window (2) disposed within the housing, a finger capture area disposed within the housing (1) above the capture window (2), an image sensor having near infrared sensing capability disposed behind the capture window (2), and a processor circuit in electrical communication with the image sensor, the processor circuit configured to perform the steps of:
s1, acquiring multi-frame image data which are used for hemoglobin shadow formed by absorption of near infrared light by hemoglobin in flowing blood in the finger to be acquired and captured by an image sensor with near infrared sensing capability;
s2: performing quality evaluation on the captured image data by using a hemoglobin shadow rough model, confirming the quality grade, and inputting the image data of the hemoglobin shadow characteristics meeting the quality requirement into S3;
s3, preprocessing the captured hemoglobin photographic image data of the finger vein to obtain data of the features to be extracted, wherein the preprocessing further comprises the following steps:
s31, enhancing image data;
s32, denoising the enhanced image data and highlighting a hemoglobin photographic image;
s33, carrying out target detection and extraction on the hemoglobin photographic image;
s34, collecting images containing a certain amount of hemoglobin by using an edge detection algorithm;
s4: extracting a second feature point image in the set containing the hemoglobin;
s5: and comparing the extracted feature points with a pre-stored first feature point image, comparing the second feature point set with the pre-stored first feature point set, judging whether the second feature point set is matched with the first feature point set within a threshold range, and if so, successfully identifying and comparing.
2. The device of claim 1, wherein in step S31, the image data is copied to create a copy, the copy is image-inverted, and then image smoothing is performed after the image inversion, and then the image data is inverted and converted to obtain enhanced image data.
3. A finger vein recognition apparatus according to claim 2, wherein: in step S32, the enhanced image data is processed by blocking, the image data is divided into a plurality of block images according to preset conditions, the block images are clustered, and then the solution is performed by using an augmented lagrange multiplier method, so as to obtain a de-noised highlighted hemoglobin photographic image.
4. A finger vein recognition apparatus according to claim 3, wherein: in the step S33, a Region-normal (R-CNN) algorithm based on Region propofol is adopted to generate a hemoglobin target candidate box for the hemoglobin photographic image, and then the candidate box is classified and regressed, and most of hemoglobin is used as a target for detection and extraction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110203770.6A CN112801034A (en) | 2021-02-23 | 2021-02-23 | Finger vein recognition device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110203770.6A CN112801034A (en) | 2021-02-23 | 2021-02-23 | Finger vein recognition device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112801034A true CN112801034A (en) | 2021-05-14 |
Family
ID=75815427
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110203770.6A Pending CN112801034A (en) | 2021-02-23 | 2021-02-23 | Finger vein recognition device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112801034A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117333910A (en) * | 2023-09-20 | 2024-01-02 | 江苏芯灵智能科技有限公司 | Drunk detection method based on vein recognition |
-
2021
- 2021-02-23 CN CN202110203770.6A patent/CN112801034A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117333910A (en) * | 2023-09-20 | 2024-01-02 | 江苏芯灵智能科技有限公司 | Drunk detection method based on vein recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Qin et al. | Finger-vein verification based on the curvature in Radon space | |
| US9922238B2 (en) | Apparatuses, systems, and methods for confirming identity | |
| Syarif et al. | Enhanced maximum curvature descriptors for finger vein verification | |
| Han et al. | Palm vein recognition using adaptive Gabor filter | |
| CN108985134B (en) | Face living body detection and face brushing transaction method and system based on binocular camera | |
| TWI599964B (en) | Finger vein recognition system and method | |
| Alheeti | Biometric iris recognition based on hybrid technique | |
| Garland | Parallel computing with CUDA | |
| CN111462379A (en) | Access control management method, system and medium containing palm vein and face recognition | |
| CN102542281A (en) | Non-contact biometric feature identification method and system | |
| Vega et al. | Biometric personal identification system based on patterns created by finger veins | |
| Carney et al. | A multi-finger touchless fingerprinting system: Mobile fingerphoto and legacy database interoperability | |
| Tazim et al. | Biometric authentication using CNN features of dorsal vein pattern extracted from NIR image | |
| Estacio et al. | A rotation invariant algorithm for bimodal hand vein recognition system | |
| CN112801034A (en) | Finger vein recognition device | |
| Vyas et al. | Smartphone based iris recognition through optimized textural representation | |
| KR101767051B1 (en) | Method and apparatus for extracting finger vein image based on fuzzy inference | |
| Sisodia et al. | A conglomerate technique for finger print recognition using phone camera captured images | |
| Kubanek et al. | Feature extraction of palm vein patterns based on two-dimensional density function | |
| Amjed et al. | Noncircular iris segmentation based on weighted adaptive hough transform using smartphone database | |
| Safiullina et al. | Image processing for biometric scanning of the palm vein pattern | |
| Sharma et al. | Improved human identification using finger vein images | |
| Kaur et al. | Finger vein detection using repeated line tracking, even Gabor and multilinear discriminant analysis (mda) | |
| CN112861756A (en) | Finger vein identification method and system | |
| Rossant et al. | A robust iris identification system based on wavelet packet decomposition and local comparisons of the extracted signatures |
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 |