CN111611890B - Fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic - Google Patents

Fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic Download PDF

Info

Publication number
CN111611890B
CN111611890B CN202010406720.3A CN202010406720A CN111611890B CN 111611890 B CN111611890 B CN 111611890B CN 202010406720 A CN202010406720 A CN 202010406720A CN 111611890 B CN111611890 B CN 111611890B
Authority
CN
China
Prior art keywords
fingerprint
finger vein
matching
matching scores
cluster
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.)
Active
Application number
CN202010406720.3A
Other languages
Chinese (zh)
Other versions
CN111611890A (en
Inventor
彭剑
戴建新
吴婧漪
陈芸
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Posts and Telecommunications
Original Assignee
Nanjing University of Posts and Telecommunications
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nanjing University of Posts and Telecommunications filed Critical Nanjing University of Posts and Telecommunications
Priority to CN202010406720.3A priority Critical patent/CN111611890B/en
Publication of CN111611890A publication Critical patent/CN111611890A/en
Application granted granted Critical
Publication of CN111611890B publication Critical patent/CN111611890B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G06V40/12—Fingerprints or palmprints
    • G06V40/1347—Preprocessing; Feature extraction
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G06F18/253—Fusion techniques of extracted features
    • 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)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Human Computer Interaction (AREA)
  • Multimedia (AREA)
  • Probability & Statistics with Applications (AREA)
  • Collating Specific Patterns (AREA)
  • Image Analysis (AREA)

Abstract

The invention provides a fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic, which comprises the following steps: step 1) extracting fingerprint and finger vein images, carrying out image preprocessing, extracting characteristic points, matching the characteristic points, and calculating matching scores; step 2) using a k-medoid clustering algorithm to respectively perform clustering analysis on the fingerprint matching scores and the finger vein matching scores, and dividing the matching scores of each mode into three types: the same type, uncertain type and different type are used, and a set formed by the matching scores of the three types is used as a fuzzy subset; step 3) selecting a proper membership function, fuzzifying the variable, and establishing a fuzzy logic relationship; step 4), obtaining a fuzzy inference conclusion by using a mamdani fuzzy inference method; and 5) defuzzification is carried out by using a weighted average judgment method to obtain a final fusion result, and a decision is made. The method can improve the identification rate of the biological characteristic identification system, thereby improving the working efficiency of the system.

Description

Fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic
Technical Field
The invention relates to a fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic, which can be used for biological feature identification and belongs to the technical field.
Background
Identity authentication is a closely related link in daily life of people, and is as small as home management and fund management and as large as national management and financial management, and identity authentication is carried out all the time. The identity recognition technology based on the biological characteristics is favored by users by means of higher safety and convenience, and has stronger advantages. The single-mode biometric recognition often has a high false recognition rate due to noise of a sensor and defects of feature extraction and matching, or has a safety problem due to easy loss of features, so that the application environment is limited too much, and each biometric feature cannot be universal in a true sense. The existence of such inherent problems is difficult to solve by simply improving a matching method, and the problems can be overcome by using a multi-modal biometric recognition system. The fingerprint features have the advantages of permanence and universality of application, the finger veins have high anti-counterfeiting performance, the two modes are fused, the advantage complementation of accuracy and safety can be realized, the identification performance and the use value of the system are greatly improved, and the two features of the fingerprint and the finger veins are simultaneously acquired at the same part of a finger, so that the hardware integration is facilitated, the cost is reduced, and the safety of the system is greatly improved. The bimodal fusion recognition based on the fingerprint and the finger vein features has strong advantages in the field of biological feature recognition.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic, so as to improve the identification rate of biological feature matching, and improve the robustness, operation efficiency and safety of a system.
In order to achieve the purpose, the invention adopts the following technical scheme:
the fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic comprises the following steps:
step 1: extracting a fingerprint image and a finger vein image, respectively carrying out image preprocessing, feature point extraction and feature point matching, calculating a matching score and carrying out normalization processing;
step 2: respectively carrying out cluster analysis on the normalized fingerprint matching scores and the normalized finger vein matching scores by using a k-medoid clustering algorithm, dividing the matching scores of each mode into a plurality of classes, and taking a set formed by the matching scores of the classes as a fuzzy subset;
and step 3: selecting a membership function, fuzzifying an input variable and an output variable respectively, and establishing a fuzzy logic relationship between the input variable and the output variable;
and 4, step 4: obtaining a fuzzy inference conclusion by using a fuzzy inference method;
and 5: and (4) performing defuzzification by using a weighted average judgment method to obtain a final fusion result and making a decision.
Further, in step 1, the image preprocessing includes image scale and gray scale normalization, image noise removal, image hole and burr removal and image thinning.
Further, in step 1, a Poincare Index fingerprint singular point detection algorithm is adopted for extracting the feature points of the fingerprint image.
Further, in the step 1, the characteristic points are matched, and for fingerprint characteristics, the Euclidean distance between an original image and an image to be matched is calculated for matching; and for the finger vein features, matching is carried out by calculating the Hamming distance between the original image and the feature code of the image to be matched.
Further, in the step 2, a k-medoid clustering algorithm is used for respectively carrying out clustering analysis on the fingerprint matching score and the finger vein matching score, and the specific steps are as follows:
step 2.1: inputting a single-mode matching score set;
step 2.2: randomly selecting a plurality of representative objects as initial central points;
step 2.3: assigning each remaining object to the cluster represented by the center point closest thereto;
step 2.4: calculating the distances from all the other points to all the central points, and taking the cluster with the shortest distance from each point to each central point as the cluster to which the cluster belongs;
step 2.5: sequentially selecting points in each cluster according to the sequence, calculating the sum of distances from the point to all points in the current cluster, and regarding the point with the minimum sum of distances as a new central point;
step 2.6: repeat step 2.4, step 2.5 until the center point of each cluster does not change.
Further, the matching scores of each modality are classified into three categories in step 2: the classification principle of the same type, uncertain type and different type is as follows: according to the size of the normalized matching score corresponding to each cluster center, the matching scores are divided into the same type, the uncertain type and the different type when the matching scores are smaller or larger.
Further, in step 3, a triangular membership function is adopted for the uncertain matching scores in the input variables, and a trapezoidal membership function is adopted for the same and different matching scores in the input variables.
Further, fuzzification processing is carried out on the output variable in the step 3, and a triangular membership function is adopted; naming the output variable as a recognition result, and defining the range of the recognition result as [0,1 ]; there are five fuzzy subsets of this output variable, named: good, medium, poor, very poor.
Further, the fuzzy logic relationship between the input variable and the output variable in step 3 includes the following rules:
rule 1: if the fingerprint matching scores belong to the same class and the finger vein matching scores belong to the same class, the recognition result is good;
rule 2: if the fingerprint matching scores belong to the same class and the finger vein matching scores belong to the uncertain class, the identification result is good;
rule 3: if the fingerprint matching scores belong to the same class and the finger vein matching scores belong to different classes, the identification result is medium;
rule 4: if the fingerprint matching score belongs to the uncertain class and the finger vein matching score belongs to the same class, the identification result is good;
rule 5: if the fingerprint matching score belongs to the uncertain class and the finger vein matching score belongs to the uncertain class, the identification result is medium;
rule 6: if the fingerprint matching score belongs to the uncertain class and the finger vein matching score belongs to the different classes, the identification result is poor;
rule 7: if the fingerprint matching scores belong to different classes and the finger vein matching scores belong to the same class, the identification result is medium;
rule 8: if the fingerprint matching score belongs to different classes and the finger vein matching score belongs to an uncertain class, the identification result is poor;
rule 9: if the fingerprint matching score belongs to a different class and the finger vein matching score belongs to a different class, the recognition result is poor.
Further, the fuzzy inference method in step 4 is a mamdani fuzzy inference method.
Has the advantages that: the method can improve the identification rate of the biological characteristic identification system, thereby improving the working efficiency of the system. The fingerprint characteristics have persistence and universality, the finger veins have high anti-counterfeiting performance, and the two modes are fused, so that the advantages of accuracy and safety can be complemented, and the system robustness is greatly improved.
Drawings
FIG. 1 is a flow chart of the fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic.
Detailed Description
The technical scheme of the invention is further explained in detail by combining the attached drawings:
the embodiment provides a fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic, the specific flow is as shown in the attached figure 1, and the method comprises the following steps:
1. extracting fingerprint and finger vein images, performing image preprocessing, extracting characteristic points, matching the characteristic points, calculating matching scores and performing normalization processing;
the image preprocessing process mainly comprises the steps of image scale and gray scale normalization, image noise removal, image hole and burr removal and image thinning.
And aiming at the characteristic point extraction stage of fingerprint identification, adopting a Poincarendex fingerprint singular point detection algorithm. And judging whether a certain pixel point is a core point or not according to the index value of the pixel point in the calculation process, and extracting the coordinate and direction length information of the point if the certain pixel point is the core point.
And a characteristic point matching stage, namely calculating the Euclidean distance of the characteristic points for the fingerprint characteristics:
Figure BDA0002491575090000041
wherein x is 0 Representing the characteristic points of the original fingerprint image, and x representing the characteristic points of the image to be matched.
For finger vein features, matching is carried out by calculating the Hamming distance between an original image and the feature codes of the image to be matched:
Figure BDA0002491575090000042
wherein, the finger vein images processed by M and N, codeM and codeN are feature codes obtained from the two vein images, and mask M and mask N are mask scores introduced by the two images due to light shielding respectively.
2. Using a k-medoid clustering algorithm to perform clustering analysis on the fingerprint matching scores and the finger vein matching scores respectively, and dividing the matching scores of each mode into three types: the same type, uncertain type and different type are used, and a set formed by the matching scores of the three types is used as a fuzzy subset;
for an input fingerprint matching score set, namely a vein matching score set, a k-medoid clustering algorithm is used for respectively carrying out clustering analysis on the fingerprint matching score and the vein matching score, and the method specifically comprises the following steps:
(1) inputting a single-mode matching score set;
(2) randomly selecting 3 representative objects as initial central points;
(3) assigning each remaining object to the cluster represented by the center point closest thereto;
(4) calculating the distances from all the other points to 3 central points, and taking the cluster with the shortest distance from each point to 3 central points as the cluster to which the cluster belongs;
(5) sequentially selecting points in each cluster according to the sequence, calculating the sum of distances from the point to all points in the current cluster, and regarding the point with the minimum sum of distances as a new central point;
(6) repeating (4) and (5) until the center point of each cluster is not changed.
After the cluster analysis is finished, distinguishing each cluster, wherein the specific classification principle is as follows: according to the size of the normalized matching score corresponding to each cluster center, the matching scores are divided into the same type, the uncertain type and the different type when the matching scores are smaller or larger. And the boundary values for each cluster are recorded for subsequent calculations.
3. Selecting a proper membership function, fuzzifying the variable, and establishing a fuzzy logic relationship;
and selecting a triangular membership function and a trapezoidal membership function for input variables, namely the fingerprint matching fraction set and the finger vein matching fraction set.
The triangle membership function calculation formula is as follows:
Figure BDA0002491575090000051
wherein a, b and c are determined by the boundary value of each cluster in step 2.
The trapezoidal membership function has the calculation formula:
Figure BDA0002491575090000052
wherein a, b, c and d are determined by the boundary value of each cluster in the step 2.
For the output variable, it is named "recognition result", and its range is defined as [0,1 ]. There are five fuzzy subsets of this output variable, named: good, medium, poor, very poor. And the output variable is fuzzified by using a triangular membership function.
For the relationship between input variables and output variables, the following rules are formulated:
rule 1: if the fingerprint matching scores belong to the same class and the finger vein matching scores belong to the same class, the recognition result is good;
rule 2: if the fingerprint matching score belongs to the same class and the finger vein matching score belongs to the uncertain class, the recognition result is good.
Rule 3: if the fingerprint match scores belong to the same class and the finger vein match scores belong to different classes, the recognition result is medium.
Rule 4: if the fingerprint matching score belongs to the uncertain class and the finger vein matching score belongs to the same class, the recognition result is good.
Rule 5: if the fingerprint matching score belongs to the uncertain class and the finger vein matching score belongs to the uncertain class, the recognition result is medium.
Rule 6: if the fingerprint matching score belongs to the uncertain class and the finger vein matching score belongs to the different class, the recognition result is poor.
Rule 7: if the fingerprint matching scores belong to different categories and the finger vein matching scores belong to the same category, the recognition result is medium.
Rule 8: if the fingerprint matching score belongs to a different class and the finger vein matching score belongs to an uncertain class, the recognition result is poor.
Rule 9: if the fingerprint matching score belongs to a different class and the finger vein matching score belongs to a different class, the recognition result is poor.
4. Obtaining a fuzzy inference conclusion by using a Mamdani fuzzy inference method;
the formula of the Mamdani fuzzy inference method is as follows:
u(s)＝max n [min(u fing (fing)，u vein (vein))，u s (s)] (5)
wherein n is a regular number in the range of 1 to 9 positive integers, and u is fing (fing) and u vein (vein) is the degree of membership, u, calculated from the fingerprint and finger vein match scores s (s) is a membership function under the action of rule n.
5. And (4) performing defuzzification by using a weighted average judgment method to obtain a final fusion result and making a decision.
The weighted average calculation formula is as follows:
Figure BDA0002491575090000061
where u(s) is the maximum value obtained by the Mamdani fuzzy inference method, and s is the weight, usually taking the median of each set.
The final decision should be determined according to the relationship between the threshold and the weighted average, and the optimal threshold should be obtained by multiple experiments and comparing the experimental results.
The above description is only an embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can understand that the modifications or substitutions within the technical scope of the present invention are included in the scope of the present invention, and therefore, the scope of the present invention should be subject to the protection scope of the claims.

Claims (10)

1. The fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic is characterized by comprising the following steps:
step 1: extracting a fingerprint image and a finger vein image, respectively carrying out image preprocessing, feature point extraction and feature point matching, calculating a matching score and carrying out normalization processing;
step 2: respectively carrying out cluster analysis on the normalized fingerprint matching scores and the normalized finger vein matching scores by using a k-medoid clustering algorithm, dividing the matching scores of each mode into a plurality of classes, and taking a set formed by the matching scores of the classes as a fuzzy subset;
and step 3: selecting a membership function, fuzzifying an input variable and an output variable respectively, and establishing a fuzzy logic relationship between the input variable and the output variable; inputting a variable fingerprint matching score set and a finger vein matching score set; there are five fuzzy subsets of output variables, named: good, medium, poor, very poor;
and 4, step 4: obtaining a fuzzy inference conclusion by using a fuzzy inference method;
and 5: and (4) performing defuzzification by using a weighted average judgment method to obtain a final fusion result and making a decision.
2. The fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic according to claim 1, wherein in step 1, the image preprocessing comprises image scale and gray level normalization, image noise removal, image hole and burr removal and image refinement.
3. The fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic as claimed in claim 1, wherein in step 1, the feature point extraction of the fingerprint image adopts Poincare Index fingerprint singular point detection algorithm.
4. The fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic according to claim 1, characterized in that, the feature points in step 1 are matched, and for the fingerprint features, the matching is performed by calculating the Euclidean distance between the original image and the image to be matched; and for the finger vein features, matching is carried out by calculating the Hamming distance between the original image and the feature code of the image to be matched.
5. The fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic as claimed in claim 1, wherein said step 2 uses k-medoid clustering algorithm to perform cluster analysis on fingerprint matching score and finger vein matching score respectively, and its specific steps are as follows:
step 2.1: inputting a single-mode matching score set;
step 2.2: randomly selecting a plurality of representative objects as initial central points;
step 2.3: assigning each remaining object to the cluster represented by the center point closest thereto;
step 2.4: calculating the distances from all the other points to all the central points, and taking the cluster with the shortest distance from each point to each central point as the cluster to which the cluster belongs;
step 2.5: sequentially selecting points in each cluster according to the sequence, calculating the sum of distances from the point to all points in the current cluster, and regarding the point with the minimum sum of distances as a new central point;
step 2.6: repeat step 2.4, step 2.5 until the center point of each cluster does not change.
6. The fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic according to claim 5, wherein the matching scores of each modality are classified into three categories in step 2: the classification principle of the same type, uncertain type and different type is as follows: according to the size of the normalized matching score corresponding to each cluster center, the matching scores are divided into the same type, the uncertain type and the different type when the matching scores are smaller or larger.
7. The fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic as claimed in claim 6, wherein step 3 employs triangle membership functions for uncertain class matching scores in input variables and trapezoidal membership functions for same and different class matching scores in input variables.
8. The fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic as claimed in claim 6, wherein step 3 fuzzifies the output variable using triangle membership function; naming the output variable as a recognition result, and defining the range of the recognition result as [0,1 ]; there are five fuzzy subsets of this output variable, named: good, medium, bad, very bad.
9. The fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic according to claim 8, wherein the fuzzy logic relationship between the input variables and the output variables in step 3 comprises the following rules:
rule 1: if the fingerprint matching scores belong to the same class and the finger vein matching scores belong to the same class, the recognition result is good;
rule 2: if the fingerprint matching scores belong to the same class and the finger vein matching scores belong to the uncertain class, the identification result is good;
rule 3: if the fingerprint matching scores belong to the same class and the finger vein matching scores belong to different classes, the identification result is medium;
rule 4: if the fingerprint matching score belongs to the uncertain class and the finger vein matching score belongs to the same class, the identification result is good;
rule 5: if the fingerprint matching score belongs to the uncertain class and the finger vein matching score belongs to the uncertain class, the identification result is medium;
rule 6: if the fingerprint matching score belongs to the uncertain class and the finger vein matching score belongs to the different classes, the identification result is poor;
rule 7: if the fingerprint matching scores belong to different classes and the finger vein matching scores belong to the same class, the identification result is medium;
rule 8: if the fingerprint matching score belongs to different classes and the finger vein matching score belongs to an uncertain class, the identification result is poor;
rule 9: if the fingerprint matching score belongs to a different class and the finger vein matching score belongs to a different class, the recognition result is poor.
10. The method for fusing fingerprint and finger vein features based on cluster analysis and fuzzy logic according to claim 8, wherein the fuzzy inference method in step 4 is mamdani fuzzy inference method.
CN202010406720.3A 2020-05-14 2020-05-14 Fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic Active CN111611890B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010406720.3A CN111611890B (en) 2020-05-14 2020-05-14 Fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010406720.3A CN111611890B (en) 2020-05-14 2020-05-14 Fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic

Publications (2)

Publication Number Publication Date
CN111611890A CN111611890A (en) 2020-09-01
CN111611890B true CN111611890B (en) 2022-08-02

Family

ID=72194660

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010406720.3A Active CN111611890B (en) 2020-05-14 2020-05-14 Fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic

Country Status (1)

Country Link
CN (1) CN111611890B (en)

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101847208B (en) * 2010-06-11 2012-06-06 哈尔滨工程大学 Secondary classification fusion identification method for fingerprint and finger vein bimodal identification
CN101901336B (en) * 2010-06-11 2012-03-14 哈尔滨工程大学 Fingerprint and finger vein bimodal recognition decision level fusion method

Also Published As

Publication number Publication date
CN111611890A (en) 2020-09-01

Similar Documents

Publication Publication Date Title
Lim et al. Efficient iris recognition through improvement of feature vector and classifier
CN111507206B (en) Finger vein identification method based on multi-scale local feature fusion
Vishi et al. An evaluation of score level fusion approaches for fingerprint and finger-vein biometrics
Aizi et al. Score level fusion in multi-biometric identification based on zones of interest
Latha et al. Efficient approach to normalization of multimodal biometric scores
Rajasekar et al. Efficient multimodal biometric recognition for secure authentication based on deep learning approach
CN109508689B (en) Face recognition method for strengthening confrontation
Obaidullah et al. Comparison of different classifiers for script identification from handwritten document
Li et al. A fingerprint indexing scheme with robustness against sample translation and rotation
CN111611890B (en) Fingerprint and finger vein feature fusion method based on cluster analysis and fuzzy logic
Pawar et al. An optimize multimodal biometric authentication system for low classification error rates using face and fingerprint
CN111898400A (en) Fingerprint activity detection method based on multi-modal feature fusion
Akulwar et al. Secured multi modal biometric system: a review
CN111428064A (en) Small-area fingerprint image fast indexing method, device, equipment and storage medium
CN111209850A (en) Method for generating applicable multi-device identification finger vein image based on improved cGAN network
Verma et al. Iris recognition based on robust iris segmentation and image enhancement
CN111931749B (en) Face clustering method and device based on neural network
Hameed et al. Fuzzy logic decision fusion in a fingerprints based multimodal biometric system
CN114973307A (en) Finger vein identification method and system for generating countermeasure and cosine ternary loss function
Prakash et al. Fusion of multimodal biometrics using feature and score level fusion
Sharma et al. Multimodal biometric system fusion using fingerprint and face with fuzzy logic
Kundu et al. An efficient integrator based on template matching technique for person authentication using different biometrics
Miron et al. Fuzzy logic method for partial fingerprint recognition
Samai et al. Oriented Local Binary Pattern (LBP θ): A new scheme for an efficient feature extraction technique
Sheena et al. Fingerprint Classification with reduced penetration rate: Using Convolutional Neural Network and DeepLearning

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