CN106874824A - A kind of Frequency Band Selection method and apparatus for personal recognition - Google Patents

A kind of Frequency Band Selection method and apparatus for personal recognition Download PDF

Info

Publication number
CN106874824A
CN106874824A CN201510919101.3A CN201510919101A CN106874824A CN 106874824 A CN106874824 A CN 106874824A CN 201510919101 A CN201510919101 A CN 201510919101A CN 106874824 A CN106874824 A CN 106874824A
Authority
CN
China
Prior art keywords
frequency band
palm print
print image
function
sigma
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
Application number
CN201510919101.3A
Other languages
Chinese (zh)
Other versions
CN106874824B (en
Inventor
孙骏文
王伟明
王琼
王平安
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.)
Shenzhen Institute of Advanced Technology of CAS
Original Assignee
Shenzhen Institute of Advanced Technology of CAS
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 Shenzhen Institute of Advanced Technology of CAS filed Critical Shenzhen Institute of Advanced Technology of CAS
Priority to CN201510919101.3A priority Critical patent/CN106874824B/en
Publication of CN106874824A publication Critical patent/CN106874824A/en
Application granted granted Critical
Publication of CN106874824B publication Critical patent/CN106874824B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/12—Fingerprints or palmprints
    • G06V40/1347—Preprocessing; Feature extraction

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Collating Specific Patterns (AREA)

Abstract

The present invention provides a kind of Frequency Band Selection method and apparatus for personal recognition, with the operand reduced during personal recognition and improves the efficiency of personal recognition.Methods described includes：Correspond to the less frequency range of information content in rejecting palmprint image, obtain the first alternative frequency range；The relatively low frequency range of error rate such as select from the palmprint image to correspond to based on Gabor filter as the second alternative frequency range；According to k clustering algorithms, the good frequency range cluster of some clusters is calculated from lap of the described first alternative frequency range with the described second alternative frequency range, the corresponding frequency range in center of the good frequency range cluster of the cluster is used as the frequency range eventually for personal recognition.The technical scheme that the present invention is provided understands that the efficiency and accuracy rate of personal recognition can be significantly improved.

Description

Frequency band selection method and device for palm print identification
Technical Field
The invention belongs to the field of image processing, and particularly relates to a frequency band selection method and device for palm print identification.
Background
With the improvement of imaging technology and computational performance, palm print recognition gradually develops into one of the mainstream biological recognition technologies, and with the application of a depth camera and a multispectral/hyperspectral camera, recognition based on palm print veins, 3D information, multispectral and hyperspectral images becomes possible. Besides the complicated palm prints on the surface of the palm of a human hand, the palm of the human hand also has complicated arterial venous blood vessels and various connective tissues inside the palm of the human hand. Because different tissues have different optical characteristics under different spectrums, different images can be obtained under different spectrums, so that more effective information can be provided by combining palm images under various spectrums, and the accuracy of palm identification is improved.
In order to further improve the accuracy of palm print identification, the existing palm print identification method introduces a multispectral imaging technology. The multispectral imaging technology can obtain images of the palm epidermis, internal tissues and blood vessels under different spectrums. However, the number of pictures is too large, which causes a drastic increase in the amount of computation. Another existing method of palm print recognition is based on pixel-level multi-band image fusion, and applies wavelet and curvelet transforms. In this method, the selection of the frequency band is critical. In view of this, one method of selecting the frequency bands is to select the optimal combination of frequency bands by exhaustive selection.
However, the above-described palm print identification method by exhaustive selection of the optimal band combination is too inefficient for palm print identification of hyper-spectra, since the number of bands of the hyper-spectrum is typically on the order of 10^2 or more.
Disclosure of Invention
The invention aims to provide a frequency band selection method and a frequency band selection device for palm print recognition, so as to reduce the operation amount in the palm print recognition process and improve the palm print recognition efficiency.
The first aspect of the present invention provides a frequency band selection method for palm print identification, where the method includes:
eliminating frequency bands corresponding to a small amount of information in the palm print image to obtain a first alternative frequency band;
selecting a frequency band corresponding to a low equal error rate from the palm print image based on a Gabor filter as a second alternative frequency band;
and according to a k clustering algorithm, calculating a plurality of well-clustered frequency band clusters from the overlapped part of the first candidate frequency band and the second candidate frequency band, wherein the frequency band corresponding to the center of the well-clustered frequency band cluster is used as the frequency band finally used for palm print identification.
A second aspect of the present invention provides a frequency band selection apparatus for palm print recognition, the apparatus comprising:
the first selection module is used for eliminating frequency bands corresponding to a small amount of information in the palm print image to obtain a first alternative frequency band;
the second selection module is used for selecting a frequency band corresponding to a lower equal error rate from the palm print image as a second alternative frequency band based on a Gabor filter;
and the clustering module is used for calculating a plurality of well-clustered frequency band clusters from the overlapped part of the first candidate frequency band and the second candidate frequency band according to a k clustering algorithm, and the frequency band corresponding to the center of the well-clustered frequency band cluster is used as the frequency band finally used for palm print identification.
According to the technical scheme of the invention, on one hand, before the k clustering algorithm is carried out, the frequency bands with more information content and the frequency bands with lower error rate are screened out by processing the palm print images, so that in the palm print identification process, the operation amount is greatly reduced by only processing the palm print images corresponding to the overlapped part of the two frequency bands, and the palm print identification efficiency can be obviously improved; on the other hand, the k clustering algorithm is carried out on the palm print images corresponding to the frequency bands with more information content and the frequency bands with lower equal error rate, the frequency band clustering with good clustering is calculated, and the optimal frequency band combination is selected, so that the operation amount of palm print identification can be reduced, and the accuracy of palm print identification can be obviously improved.
Drawings
Fig. 1 is a schematic flow chart illustrating an implementation of a frequency band selection method for palm print identification according to an embodiment of the present invention;
fig. 2 is a schematic diagram of a correspondence relationship between an entropy value in an image and a frequency band according to a second embodiment of the present invention;
FIG. 3 is a schematic diagram illustrating a relationship between different frequency bands and equal error rates thereof in an image according to a third embodiment of the present invention;
fig. 4 is a schematic structural diagram of a frequency band selection apparatus for palm print recognition according to a fourth embodiment of the present invention;
fig. 5 is a schematic structural diagram of a frequency band selection apparatus for palm print identification according to a fifth embodiment of the present invention;
fig. 6 is a schematic structural diagram of a frequency band selection apparatus for palm print recognition according to a sixth embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more clearly apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The embodiment of the invention provides a frequency band selection method for palm print identification, which comprises the following steps: eliminating frequency bands corresponding to a small amount of information in the palm print image to obtain a first alternative frequency band; selecting a frequency band corresponding to a low equal error rate from the palm print image based on a Gabor filter as a second alternative frequency band; and according to a k clustering algorithm, calculating a plurality of well-clustered frequency band clusters from the overlapped part of the first candidate frequency band and the second candidate frequency band, wherein the frequency band corresponding to the center of the well-clustered frequency band cluster is used as the frequency band finally used for palm print identification. The embodiment of the invention also provides a corresponding frequency band selection device for palm print identification. The following are detailed below.
Referring to fig. 1, a schematic flow chart of a method for selecting a frequency band for palm print recognition according to an embodiment of the present invention is shown, which mainly includes the following steps S101 to S103:
s101, eliminating frequency bands corresponding to small information amount in the palm print image to obtain a first alternative frequency band.
It should be noted that, the images obtained in different frequency bands contain different amounts of information, the palm print image also conforms to the rule, and the palm print image with rich information content is easier to identify than the palm print image with less information content. Therefore, in the embodiment of the present invention, the frequency band corresponding to a small amount of information in the palm print image may be removed first to obtain the first candidate frequency band, which may be specifically implemented by the following steps S1011 and S1012:
s1011, a gray scale distribution value of the palm print image is calculated.
In the embodiment of the invention, the gray distribution value p of the palm print image is calculateddCan be obtained by the following formula (1):
..
Wherein, Ii,jRepresenting 8-bit coded palm prints, Ii,jSubscripts I and j of (a) indicate the position of the image, m and n are the size of the palm print image, and k is the palm print image Ii,jThe gray scale of (2).
S1012, calculating entropies corresponding to the frequency bands in the palm print image according to the gray scale distribution values calculated in step S1011, and using the frequency band with a larger entropy value as the first candidate frequency band.
In the embodiment of the present invention, according to the gray distribution value calculated in step S1011, the entropy E of the palm print image can be calculated by the following formula (2):
..
For an image, the larger the entropy value of the image is, the more information is contained, and the better the quality is; the palm print image also conforms to the rule. Therefore, in the embodiment of the present invention, in calculating the entropy corresponding to each frequency band in the palm print image, the frequency band with a larger entropy value may be used as the first candidate frequency band, so as to improve the recognition rate in the subsequent palm print recognition.
As shown in fig. 2, the correspondence between the entropy value and the frequency band in the image is shown. In the embodiment of the present invention, it is obvious that if the removed palm print image corresponds to a frequency band with a wavelength below 590nm and above 1020nm, and the remaining frequency band with a wavelength between 590nm and 1020nm has a relatively large entropy value, the information amount of the image in the frequency band range is relatively large, and therefore, the frequency band with a wavelength between 590nm and 1020nm can be used as the first candidate frequency band.
And S102, selecting a frequency band corresponding to a low equal error rate from the palm print image based on the Gabor filter as a second candidate frequency band.
In the field of image processing, Equal Error Rate (EER) is derived from two concepts of False Rejection Rate (FRR) and False Rejection Rate (FAR), wherein the False Rejection Rate, i.e. the probability of False Rejection, is formulated from intra-class matching, and if there are 10 samples of volunteers, 20 samples per volunteer, then with respect to intra-class testing, for example, for volunteer No. 1, the 20 images of the same class are matched with each other (assuming 1:1 matching), and they can be performed (20 × 19)/2 times without repeating each other; if 10 volunteers do the test, 10 × 19/2 times is the total number of matching times in the class, each matching will obtain a matching value TH according to the matching algorithm, the preset threshold value is TH, if TH>TH will be rejected erroneously; the false recognition rate, i.e. the probability of false acceptance, is to illustrate the problem from the matching between classes, i.e. the matching between different classes, and if the matching value TH obtained according to the matching algorithm is smaller than the preset threshold TH, it is considered that the same class belongs, which is the false acceptance. The FRR calculation formula isFAR is calculated as Wherein NGRA is the total number of intra-class testsThe number of times, NIRA is the total number of inter-class tests, NFR and NFA are the number of false rejects and false accepts. Since FRR and FAR contradict each other, the probability when FRR and FAR are equal is EER.
As an embodiment of the present invention, selecting a frequency band corresponding to a lower equal error rate from the palm print image based on a Gabor filter as the second candidate frequency band may be implemented by the following steps S1021 to S1022:
and S1021, filtering the palm print image based on the Gabor filter.
Specifically, the filtering of the palm print image based on the Gabor filter may be based on a Gabor filter expressed by a function G (x, y, θ, u, σ) that is filtered in six directions of the function G (x, y, θ, u, σ), where the function G (x, y, θ, u, σ) is defined as:
wherein,x and y represent the position coordinates of the Gabor filter, u is the frequency of the sine wave, θ is the direction of the function G (x, y, θ, u, σ), σ is the variance of the Gaussian function, and the six directions of the function G (x, y, θ, u, σ) are 0, and,And
in the embodiment of the invention, the two-dimensional Gabor filter can effectively extract the direction information in the palm print image, and different directions of the palm print image after filtering are represented as different gray scales.
And S1022, calculating the equal error rate of each frequency band of the filtered palm print image, and taking the corresponding frequency band with the equal error rate smaller than a preset threshold value as a second alternative frequency band.
After the palm print image passes through a Gabor Filter, a gray level image which is subjected to coding is obtained, each pixel in the image is coded into a 3-bit (bit) digital code, any two palm print images can be converted into a Gabor Filter Map firstly, and then the Hamming distance between the two palm print images, namely the matching value of intra-class or inter-class matching is calculated, so that the EER is obtained through calculation.
Since lower EER indicates higher accuracy of the identification system, after calculating the equal error rate of each frequency band, the corresponding frequency band with the equal error rate smaller than the preset threshold value may be used as the second candidate frequency band. As shown in fig. 3, the relationship between the different frequency bands and the equal error rates thereof can be seen from fig. 3 that the equal error rates are relatively low in the frequency band with the wavelength between 580nm and 1080nm, and therefore, as an embodiment of the present invention, the frequency band with the wavelength between 580nm and 1080nm can be used as the second candidate frequency band.
S103, according to a k clustering algorithm, calculating a plurality of well-clustered frequency band clusters from the overlapped part of the first candidate frequency band and the second candidate frequency band, wherein the frequency band corresponding to the center of the well-clustered frequency band clusters is used as the frequency band finally used for palm print identification.
The K-Means algorithm is an unsupervised learning method and can reveal the relation between each frequency band. The purpose of the embodiment of the present invention is to select a frequency band that contains more information and has a lower error rate for palm print recognition, so that the first candidate frequency band and the second candidate frequency band mentioned in the foregoing embodiment, for example, the frequency band with a wavelength between 590nm and 1020nm illustrated in fig. 2 and the frequency band with a wavelength between 580nm and 1080nm illustrated in fig. 3, and the overlapping portion, i.e., the frequency band with a wavelength between 580nm and 1020nm, can take into account the advantages of both containing more information and having a lower error rate, so that a plurality of well-clustered frequency bands can be calculated from the frequency band with a wavelength between 580nm and 1020nm according to a k-clustering algorithm.
As an embodiment of the invention, there are 45 frequency bands in the frequency band with wavelength between 580nm to 1020nm, according to K clustering (K-Means) algorithm, these 45 frequency bands are clustered into 4 kinds, the center is the frequency band corresponding to 600nm, 770nm, 850nm and 1000nm respectively; the calculation process and the result of the algorithm show that the 4 classes are well clustered frequency bands, and the frequency band corresponding to the center of the 4 classes is used as the frequency band finally used for palm print identification. The good clustering means that in the k clustering algorithm process, when the clustering number is set to be 4, four values are randomly selected from frequency bands with the wavelength of 580nm to 1020nm as initial clustering centers, and the sum S of Hamming distances between each frequency band and the class centers in the class is selected to measure the good clustering condition. With the different values of the centers of the four classes, the obtained S values are different, and when the S values are not changed, the stable centers of the k clusters, namely 4 frequency bands corresponding to 600nm, 770nm, 850nm and 1000nm, can be found. At this time, the combination of these 4 bands is optimal, which is the most appropriate band for palm print recognition.
As can be seen from the frequency band selection method for palm print recognition illustrated in fig. 1, on one hand, the frequency band with a large information amount and the frequency band with a low error rate of the palm print image are screened out by processing the palm print image, so that in the palm print recognition process, the operation amount is greatly reduced by only processing the palm print image corresponding to the overlapped part of the two frequency bands, and the palm print recognition efficiency can be significantly improved; on the other hand, the k clustering algorithm is carried out on the palm print images corresponding to the frequency bands with more information content and the frequency bands with lower equal error rate, the frequency band clustering with good clustering is calculated, and the optimal frequency band combination is selected, so that the operation amount of palm print identification can be reduced, and the accuracy of palm print identification can be obviously improved.
Fig. 4 is a schematic structural diagram of a frequency band selection device for palm print recognition according to a fourth embodiment of the present invention. For convenience of explanation, fig. 4 shows only portions related to the embodiment of the present invention. The frequency band selection apparatus for palm print recognition illustrated in fig. 4 may be an execution subject of the frequency band selection method for palm print recognition illustrated in fig. 1. The frequency band selection apparatus for palm print recognition illustrated in fig. 4 mainly includes a first selection module 401, a second selection module 402, and a clustering module 403, where:
the first selection module 401 is configured to remove a frequency band corresponding to a small amount of information in the palm print image, and obtain a first alternative frequency band;
a second selecting module 402, configured to select, based on a Gabor filter, a frequency band corresponding to a low equal error rate from the palm print image as a second candidate frequency band;
a clustering module 403, configured to calculate, according to a k-clustering algorithm, a plurality of well-clustered frequency band clusters from an overlapping portion of the first candidate frequency band and the second candidate frequency band, where a frequency band corresponding to a center of the well-clustered frequency band cluster is used as a frequency band finally used for palm print identification.
It should be noted that, in the embodiment of the frequency band selection apparatus for palm print recognition illustrated in fig. 4, the division of the functional modules is only an example, and in practical applications, the above function allocation may be performed by different functional modules according to needs, for example, configuration requirements of corresponding hardware or convenience of implementation of software, that is, the internal structure of the frequency band selection apparatus for palm print recognition is divided into different functional modules to perform all or part of the above described functions. Moreover, in practical applications, the corresponding functional modules in this embodiment may be implemented by corresponding hardware, or may be implemented by corresponding hardware executing corresponding software, for example, the second selecting module may be hardware that executes the Gabor-based filter to select a frequency band corresponding to a lower equal error rate from the palm print image as a second candidate frequency band, for example, a second selector, or a general processor or other hardware device that can execute a corresponding computer program to complete the foregoing functions; as another example, the clustering module may be hardware that performs clustering according to a k-clustering algorithm to calculate a plurality of well-clustered frequency bands from an overlapping portion of the first candidate frequency band and the second candidate frequency band, such as a clusterer, or may be a general processor or other hardware device that can execute a corresponding computer program to perform the foregoing functions (the above-described principles may be applied to each embodiment provided in this specification).
The first selection module 401 illustrated in fig. 4 may include a grayscale calculation unit 501 and an entropy calculation unit 502, and is a frequency band selection apparatus for palm print recognition provided in the fifth embodiment of the present invention as illustrated in fig. 5, where:
a gray level calculating unit 501, configured to calculate a gray level distribution value of the palm print image;
an entropy calculating unit 502, configured to calculate an entropy corresponding to each frequency band in the palm print image according to the gray distribution value calculated by the gray calculating unit 501, and use the frequency band with the larger entropy value as the first candidate frequency band.
The second selecting module 402 illustrated in fig. 4 may include a filtering unit 601 and an equal error rate calculating unit 602, which are shown in fig. 6, where the frequency band selecting apparatus for palm print recognition according to a sixth embodiment of the present invention, where:
a filtering unit 601, configured to filter the palm print image based on a Gabor filter;
and an equal error rate calculation unit 602, configured to calculate an equal error rate of each frequency band for the filtered palm print image, and use a corresponding frequency band with the equal error rate being smaller than a preset threshold as the second candidate frequency band.
In the frequency band selection apparatus for palm print recognition illustrated in fig. 6, the filtering unit 601 is specifically configured to filter the palm print image in six directions of the function G (x, y, θ, u, σ) based on a Gabor filter represented by the function G (x, y, θ, u, σ), where the function G (x, y, θ, u, σ) is defined as follows:
wherein,x and y represent the position coordinates of the Gabor filter, and u is a sine waveTheta is the direction of the function G (x, y, theta, u, sigma), sigma is the variance of the Gaussian function, and the six directions of the function G (x, y, theta, u, sigma) are 0, respectively, And
it should be noted that, because the contents of information interaction, execution process, and the like between the modules/units of the apparatus are based on the same concept as the method embodiment of the present invention, the technical effect brought by the contents is the same as the method embodiment of the present invention, and specific contents may refer to the description in the method embodiment of the present invention, and are not described herein again.
Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by associated hardware instructed by a program, which may be stored in a computer-readable storage medium, and the storage medium may include: read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disks, and the like.
The frequency band selection method and device for palm print recognition provided by the embodiment of the present invention are introduced in detail, and a specific example is applied in the text to explain the principle and the implementation of the present invention, and the description of the above embodiment is only used to help understanding the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.

Claims (10)

1. A frequency band selection method for palm print recognition is characterized by comprising the following steps:
eliminating frequency bands corresponding to a small amount of information in the palm print image to obtain a first alternative frequency band;
selecting a frequency band corresponding to a low equal error rate from the palm print image based on a Gabor filter as a second alternative frequency band;
and according to a k clustering algorithm, calculating a plurality of well-clustered frequency band clusters from the overlapped part of the first candidate frequency band and the second candidate frequency band, wherein the frequency band corresponding to the center of the well-clustered frequency band cluster is used as the frequency band finally used for palm print identification.
2. The method of claim 1, wherein the removing the frequency band corresponding to the smaller amount of information in the palm print image to be recognized to obtain a first candidate frequency band comprises:
calculating the gray distribution value of the palm print image;
and calculating entropies corresponding to all frequency bands in the palm print image according to the gray distribution values, and taking the frequency band with a larger entropy value as the first alternative frequency band.
3. The method of claim 1, wherein the selecting, from the palm print image based on the Gabor filter, a frequency band corresponding to a lower equal error rate as the second candidate frequency band comprises:
filtering the palm print image based on a Gabor filter;
and calculating the equal error rate of each frequency band of the filtered palm print image, and taking the corresponding frequency band with the equal error rate smaller than a preset threshold value as the second alternative frequency band.
4. The method of claim 3, wherein the filtering the palm print image based on a Gabor filter comprises: filtering the palm print image in six directions of the function G (x, y, theta, u, sigma) based on a Gabor filter represented by the function G (x, y, theta, u, sigma), the function G (x, y, theta, u, sigma) being defined as follows:
G ( x , y , θ , u , σ ) = 1 2 πσ 2 exp { - x 2 + y 2 2 σ 2 } * exp { 2 π i ( u x cos θ + u y sin θ ) } , the above-mentionedThe x and y represent the position coordinates of the Gabor filter, the u is the frequency of the sine wave, the θ is the direction of a function G (x, y, θ, u, σ), and the σ is the variance of a gaussian function.
5. The method of claim 4, wherein the function G (x, y, θ, u, σ) has six directions of 0, and,And
6. a frequency band selection apparatus for palm print recognition, the apparatus comprising:
the first selection module is used for eliminating frequency bands corresponding to a small amount of information in the palm print image to obtain a first alternative frequency band;
the second selection module is used for selecting a frequency band corresponding to a lower equal error rate from the palm print image as a second alternative frequency band based on a Gabor filter;
and the clustering module is used for calculating a plurality of well-clustered frequency band clusters from the overlapped part of the first candidate frequency band and the second candidate frequency band according to a k clustering algorithm, and the frequency band corresponding to the center of the well-clustered frequency band cluster is used as the frequency band finally used for palm print identification.
7. The apparatus of claim 6, wherein the first selection module comprises:
the gray level calculating unit is used for calculating a gray level distribution value of the palm print image;
and the entropy calculation unit is used for calculating the entropy corresponding to each frequency band in the palm print image according to the gray distribution value, and taking the frequency band with a larger entropy value as the first alternative frequency band.
8. The apparatus of claim 6, wherein the second selection module comprises:
the filtering unit is used for filtering the palm print image based on a Gabor filter;
and the equal error rate calculation unit is used for calculating the equal error rate of each frequency band of the filtered palm print image, and taking the corresponding frequency band with the equal error rate smaller than a preset threshold value as the second alternative frequency band.
9. The apparatus according to claim 8, wherein said filtering unit is in particular adapted to filter said palm print image in six directions of a function G (x, y, θ, u, σ) based on a Gabor filter represented by said function G (x, y, θ, u, σ), said function G (x, y, θ, u, σ) being defined as follows:
G ( x , y , θ , u , σ ) = 1 2 πσ 2 exp { - x 2 + y 2 2 σ 2 } * exp { 2 π i ( u x cos θ + u y sin θ ) } , the above-mentionedX and y represent the position coordinates of the Gabor filter, and u is a sine waveIs the direction of the function G (x, y, theta, u, sigma), and sigma is the variance of the gaussian function.
10. The apparatus of claim 9, wherein the function G (x, y, θ, u, σ) has six directions, 0, G,And
CN201510919101.3A 2015-12-11 2015-12-11 Frequency band selection method and device for palm print identification Active CN106874824B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510919101.3A CN106874824B (en) 2015-12-11 2015-12-11 Frequency band selection method and device for palm print identification

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510919101.3A CN106874824B (en) 2015-12-11 2015-12-11 Frequency band selection method and device for palm print identification

Publications (2)

Publication Number Publication Date
CN106874824A true CN106874824A (en) 2017-06-20
CN106874824B CN106874824B (en) 2020-09-25

Family

ID=59177952

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510919101.3A Active CN106874824B (en) 2015-12-11 2015-12-11 Frequency band selection method and device for palm print identification

Country Status (1)

Country Link
CN (1) CN106874824B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107292273A (en) * 2017-06-28 2017-10-24 西安电子科技大学 Based on the special double Gabor palmmprint ROI matching process of extension eight neighborhood

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101859383A (en) * 2010-06-08 2010-10-13 河海大学 Hyperspectral remote sensing image band selection method based on time sequence important point analysis
CN102521605A (en) * 2011-11-25 2012-06-27 河海大学 Wave band selection method for hyperspectral remote-sensing image

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101859383A (en) * 2010-06-08 2010-10-13 河海大学 Hyperspectral remote sensing image band selection method based on time sequence important point analysis
CN102521605A (en) * 2011-11-25 2012-06-27 河海大学 Wave band selection method for hyperspectral remote-sensing image

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
LINLIN SHEN: "Band selection for Gabor feature based hyperspectral palmprint recognition", 《2015 INTERNATIONAL CONFERENCE ON BIOMETRICS (ICB)》 *
粘永健: "基于聚类的高光谱图像无损压缩", 《电子与信息学报》 *
葛亮: "基于波段聚类的高光谱图像波段选择", 《计算机辅助设计与图形学学报》 *
马文英: "多光谱掌纹识别波段选择方法研究", 《中国优秀硕士学位论文全文数据库》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107292273A (en) * 2017-06-28 2017-10-24 西安电子科技大学 Based on the special double Gabor palmmprint ROI matching process of extension eight neighborhood
CN107292273B (en) * 2017-06-28 2021-03-23 西安电子科技大学 Eight-neighborhood double Gabor palm print ROI matching method based on specific expansion

Also Published As

Publication number Publication date
CN106874824B (en) 2020-09-25

Similar Documents

Publication Publication Date Title
Emeršič et al. Evaluation and analysis of ear recognition models: performance, complexity and resource requirements
Jia et al. Gabor cube selection based multitask joint sparse representation for hyperspectral image classification
Mehri et al. Texture feature benchmarking and evaluation for historical document image analysis
US9070002B2 (en) Method of performing fingerprint matching
Barpanda et al. Iris recognition with tunable filter bank based feature
CN102332084B (en) Identity identification method based on palm print and human face feature extraction
Davarzani et al. Scale-and rotation-invariant texture description with improved local binary pattern features
CN111814574A (en) Face living body detection system, terminal and storage medium applying double-branch three-dimensional convolution model
CN111160273A (en) Hyperspectral image space spectrum combined classification method and device
Marcos et al. Texture classification using discrete Tchebichef moments
CN107408195B (en) Iris identification method and device
CN107092874A (en) Personal identification method, apparatus and system based on electrocardio and fingerprint fusion feature
JP6945253B2 (en) Classification device, classification method, program, and information recording medium
CN110751024A (en) User identity identification method and device based on handwritten signature and terminal equipment
CN103839042A (en) Human face recognition method and human face recognition system
Pamplona Segundo et al. Pore-based ridge reconstruction for fingerprint recognition
Yang et al. Finger-vein network enhancement and segmentation
Muhammad et al. Copy move image forgery detection method using steerable pyramid transform and texture descriptor
Xia et al. Rotation-invariant Weber pattern and Gabor feature for fingerprint liveness detection
Houcine et al. Ear recognition based on multi-bags-of-features histogram
Oktiana et al. Cross-spectral iris recognition using phase-based matching and homomorphic filtering
Kumar et al. Non-overlapped blockwise interpolated local binary pattern as periocular feature
Khalaf et al. Robust partitioning and indexing for iris biometric database based on local features
Huynh-Kha et al. A robust algorithm of forgery detection in copy-move and spliced images
Okawa Vector of locally aggregated descriptors with KAZE features for offline signature verification

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant