CN112580506A - Iris characteristic point comparison method based on bidirectional minimum Hamming distance - Google Patents
Iris characteristic point comparison method based on bidirectional minimum Hamming distance Download PDFInfo
- Publication number
- CN112580506A CN112580506A CN202011504929.XA CN202011504929A CN112580506A CN 112580506 A CN112580506 A CN 112580506A CN 202011504929 A CN202011504929 A CN 202011504929A CN 112580506 A CN112580506 A CN 112580506A
- Authority
- CN
- China
- Prior art keywords
- iris
- characteristic
- feature
- row
- characteristic point
- 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
- 238000000034 method Methods 0.000 title claims abstract description 16
- 230000002457 bidirectional effect Effects 0.000 title claims abstract description 12
- 210000000554 iris Anatomy 0.000 claims abstract description 106
- 239000013598 vector Substances 0.000 claims abstract description 45
- 238000012795 verification Methods 0.000 claims abstract description 11
- 239000011159 matrix material Substances 0.000 claims description 50
- 238000001514 detection method Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 229910002056 binary alloy Inorganic materials 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
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
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/193—Preprocessing; Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
-
- 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
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/197—Matching; Classification
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Ophthalmology & Optometry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses an iris feature point comparison method based on a bidirectional minimum Hamming distance. The invention determines the comparison strategy of the iris characteristic points according to the binarization result of the characteristic point descriptors to the iris characteristic point codes. Firstly, respectively constructing vector matrixes by binary characteristic vectors of all characteristic points in the iris to be verified and the reference iris, wherein the number of lines represents the number of the characteristic points detected in the two irises, and the number of columns is the length of the characteristic vectors; then, cross-computing Hamming distances among the feature vectors, if the Hamming distances are the minimum Hamming distances, keeping, and if not, not keeping; and finally, calculating the comparison rate of the iris characteristic points to complete iris verification. Compared with the existing iris feature point comparison method, the iris feature point comparison method improves the verification speed of the iris and reduces the feature point comparison rate of non-homologous irises.
Description
Technical Field
The invention belongs to the technical field of biological feature identification and information security, and particularly relates to an iris feature point comparison method based on a bidirectional minimum Hamming distance.
Background
The iris recognition is deeply researched by scholars at home and abroad in recent years due to the characteristics of high precision, difficulty in counterfeiting, stability and the like, and is expected to be applied to practical applications such as school identity recognition systems, mobile phone APP identity verification and the like, wherein the verification accuracy and speed influence the user experience.
The iris feature point comparison is a feature point pair similarity calculation strategy adopted according to the type of a feature point descriptor after detecting feature points in the iris and generating a binary descriptor (feature vector) for the feature points.
The iris identification algorithm based on the local invariant features of the feature points utilizes the relationship between stable pixel points in the iris and surrounding pixels thereof to carry out coding, and realizes iris verification by comparing a plurality of local features. SIFT detects feature points according to a mode of solving an extreme value by Gaussian difference, a deep and stable feature point set is obtained through a down-sampling and Gaussian fuzzy mode, then key points are represented according to the gradient of neighborhood pixels of the key points, whether the key points belong to a group of feature point pairs or not is judged through calculating Euclidean distance between the feature points, and finally the ratio of the feature point pairs is calculated to complete iris verification. The method is robust to blurring and rotation of the image. Typical SIFT-related documents are for example: region-based SIFT pro ach to iris recognition, Optics and Lasers in Engineering, 2009; SIFT based iris recognition with normalization and enhancement 2013. However, according to the features extracted in the mode, the similarity of the feature points calculated by adopting the traditional Euclidean distance relates to a large number of square and evolution operations, the verification speed is reduced, and more comparison point pairs still exist among non-homologous irises, so that the accuracy of iris verification is influenced. The invention provides a bidirectional minimum Hamming distance comparison method aiming at the comparison problem of iris feature points, and realizes stable and efficient comparison of the iris feature points.
Disclosure of Invention
The invention aims to provide an iris characteristic point comparison method based on bidirectional minimum Hamming distance under the condition of generating binary descriptors for characteristic points extracted in iris texture, so as to improve the iris identification performance based on characteristic point comparison.
The invention specifically comprises the following steps:
step 1, respectively determining the coordinates of the characteristic points of the iris to be verified and the reference iris by using a characteristic point detection algorithm, and coding the characteristic points by using a characteristic point description algorithm to generate a binary descriptor.
And 2, respectively establishing a characteristic point descriptor matrix for all characteristic points of the iris to be verified and the reference iris, wherein the number of rows of the matrix is the number of the detected iris characteristic points, and the number of columns of the matrix is the length of the characteristic point vector. And each row of feature vectors in the feature descriptor matrix is the feature descriptor.
And 3, calculating the Hamming distance between each row of characteristic vectors of the to-be-verified iris characteristic point descriptor matrix and all the row characteristic vectors of the reference iris characteristic point descriptor matrix from top to bottom in sequence, and finding the row of characteristic vectors with the minimum Hamming distance corresponding to each row of characteristic vectors in the to-be-verified iris characteristic point matrix in the reference iris characteristic point matrix.
And 4, calculating the Hamming distance again for the row of feature vectors with the minimum distance from the feature vector of the iris feature point descriptor matrix to be verified in the reference iris feature point descriptor matrix and all the row feature vectors of the iris feature point descriptor matrix to be verified, and if the Hamming distance between the latest row of feature vectors and the original row vector in the iris feature point descriptor matrix to be verified is still minimum, comparing the iris feature point pair successfully.
The invention has the following beneficial effects:
the invention determines the Hamming distance of the iris characteristic point pair according to the binary system characteristics of the characteristic points, avoids the complex calculation of the traditional Euclidean distance and realizes the rapid comparison of the iris characteristic points; compared with the one-way Hamming distance, the method greatly reduces the comparison quantity of the characteristic points between the non-homologous irises and improves the performance of iris verification.
Drawings
Fig. 1 is a schematic diagram of an iris feature point comparison method based on bidirectional minimum hamming distance designed by the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
As shown in fig. 1, the iris feature point comparison method based on the bidirectional minimum hamming distance specifically includes the following steps:
step 1, respectively determining the coordinates of the characteristic points of the iris to be verified and the reference iris by using a characteristic point detection algorithm, and coding the characteristic points by using a characteristic point description algorithm to generate a binary characteristic descriptor.
Step 2, respectively establishing a characteristic point descriptor matrix for all characteristic points of the iris to be verified and the reference iris, namely an iris characteristic point descriptor matrix X to be verified and a reference iris characteristic point descriptor matrix Y, wherein the iris image characteristic point descriptor matrix X to be verified is { X ═ X }1，x2，…，xpP is more than or equal to 3, and the descriptor matrix of the reference iris characteristic point is Y ═ Y1，y2，…，yqQ is more than or equal to 3, and p and q are the number of characteristic points detected in the two irises respectively; x is the number ofi(1. ltoreq. i.ltoreq.p) is any one of the feature point descriptors of X, yj(1. ltoreq. j. ltoreq. q) is any one of the feature point descriptors in Y.
And 3, calculating the Hamming distance between each row of characteristic vectors of the to-be-verified iris characteristic point descriptor matrix and all the row characteristic vectors of the reference iris characteristic point descriptor matrix from top to bottom in sequence, and finding the row of characteristic vectors with the minimum Hamming distance corresponding to each row of characteristic vectors in the to-be-verified iris characteristic point descriptor matrix in the reference iris characteristic point descriptor matrix.
HDdistance(xi，yj) Denotes xiTo yjThe hamming distance of; similarly HDdistance (y)j，xi) Denotes yjTo xiHamming distance of.
And 4, calculating the Hamming distance again for the row of feature vectors with the minimum distance from the feature vector of the iris feature point descriptor matrix to be verified in the reference iris feature point descriptor matrix and all the row feature vectors of the iris feature point descriptor matrix to be verified, and if the Hamming distance between the latest row of feature vectors and the original row vector in the iris feature point descriptor matrix to be verified is still minimum, comparing the iris feature point pair successfully.
When HDdistance (x)i，yj)≤HDdistance(xi，yr) R is not less than 1 and not more than q and r is not equal to j and HDdistance (y)j，xi)≤HDdistance(yj，xr) And r is more than or equal to 1 and less than or equal to p, and r is not equal to i, the comparison of the two characteristic points is correct. And finally, according to the formula of rate m/min (p, q), the ratio of the characteristic point pairs is obtained, and the iris verification is completed.
Where min (p, q) is the smaller of the number of feature points detected in the two irises.
Claims (4)
1. An iris feature point comparison method based on bidirectional minimum Hamming distance is characterized by comprising the following steps:
step 1, respectively determining the coordinates of the characteristic points of the iris to be verified and the reference iris by using a characteristic point detection algorithm, and coding the characteristic points by using a characteristic point description algorithm to generate a binary descriptor;
step 2, respectively establishing a characteristic point descriptor matrix for all characteristic points of the iris to be verified and the reference iris, wherein the number of rows of the matrix is the number of the detected iris characteristic points, and the number of columns of the matrix is the length of the characteristic point vector; each line of feature vectors in the feature point descriptor matrix is a feature point descriptor;
step 3, calculating the Hamming distance between each row of characteristic vectors of the to-be-verified iris characteristic point descriptor matrix and all the row characteristic vectors of the reference iris characteristic point descriptor matrix from top to bottom in sequence, and finding a row of characteristic vectors with the minimum Hamming distance corresponding to each row of characteristic vectors in the to-be-verified iris characteristic point matrix in the reference iris characteristic point matrix;
and 4, calculating the Hamming distance again for the row of feature vectors with the minimum distance from the feature vector of the iris feature point descriptor matrix to be verified in the reference iris feature point descriptor matrix and all the row feature vectors of the iris feature point descriptor matrix to be verified, and if the Hamming distance between the latest row of feature vectors and the original row vector in the iris feature point descriptor matrix to be verified is still minimum, comparing the iris feature point pair successfully.
2. The method for comparing iris feature points based on bidirectional minimum Hamming distance as claimed in claim 1, wherein the step 2 specifically operates as follows:
respectively establishing a characteristic point descriptor matrix for all characteristic points of the iris to be verified and the reference iris, namely an iris characteristic point descriptor matrix X to be verified and a reference iris characteristic point descriptor matrix Y, wherein the iris image characteristic point descriptor matrix X is equal to { X ═ X1，x2，…，xpP is more than or equal to 3, and the descriptor matrix of the reference iris characteristic point is Y ═ Y1，y2，…，yqQ is more than or equal to 3, and p and q are the number of characteristic points detected in the two irises respectively; x is the number ofi(1. ltoreq. i.ltoreq.p) is any one of the feature point descriptors of X, yj(1. ltoreq. j. ltoreq. q) is any one of the feature point descriptors in Y.
3. The method for comparing iris feature points based on bidirectional minimum Hamming distance as claimed in claim 2, wherein the step 3 is specifically operated as follows:
calculating the Hamming distance between each row of characteristic vectors of the to-be-verified iris characteristic point descriptor matrix and all the row characteristic vectors of the reference iris characteristic point descriptor matrix from top to bottom in sequence, and finding out a row of characteristic vectors with the minimum Hamming distance corresponding to each row of characteristic vectors in the to-be-verified iris characteristic point descriptor matrix from the reference iris characteristic point descriptor matrix;
HDdistance(xi，yj) Denotes xiTo yjThe hamming distance of; similarly HDdistance (y)j，xi) Denotes yjTo xiHamming distance of.
4. The method for comparing iris feature points based on bidirectional minimum Hamming distance as claimed in claim 3, wherein the step 4 comprises the following operations:
calculating the Hamming distance again for the row of characteristic vectors which is the smallest in distance with the characteristic vector of the iris characteristic point descriptor matrix to be verified in the reference iris characteristic point descriptor matrix and all the row characteristic vectors of the iris characteristic point descriptor matrix to be verified, and if the Hamming distance between the latest row of characteristic vectors in the distance and the original row vector in the iris characteristic point descriptor matrix to be verified is still the smallest, the iris characteristic point pair is successfully compared;
when HDdistance (x)i，yj)≤HDdistance(xi，yr) R is not less than 1 and not more than q and r is not equal to j and HDdistance (y)j，xi)≤HDdistance(yj，xr) R is more than or equal to 1 and less than or equal to p, and r is not equal to i, the comparison of the two characteristic points is correct; finally, according to a formula of rate m/min (p, q), the ratio of the characteristic point pairs is solved, and iris verification is completed;
where min (p, q) is the smaller of the number of feature points detected in the two irises.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011504929.XA CN112580506A (en) | 2020-12-18 | 2020-12-18 | Iris characteristic point comparison method based on bidirectional minimum Hamming distance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011504929.XA CN112580506A (en) | 2020-12-18 | 2020-12-18 | Iris characteristic point comparison method based on bidirectional minimum Hamming distance |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112580506A true CN112580506A (en) | 2021-03-30 |
Family
ID=75136051
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011504929.XA Pending CN112580506A (en) | 2020-12-18 | 2020-12-18 | Iris characteristic point comparison method based on bidirectional minimum Hamming distance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112580506A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101034434A (en) * | 2007-04-10 | 2007-09-12 | 杭州电子科技大学 | Identification recognizing method based on binocular iris |
| CN108010045A (en) * | 2017-12-08 | 2018-05-08 | 福州大学 | Visual pattern characteristic point error hiding method of purification based on ORB |
| CN111344703A (en) * | 2017-11-24 | 2020-06-26 | 三星电子株式会社 | User authentication device and method based on iris recognition |
| CN111833249A (en) * | 2020-06-30 | 2020-10-27 | 电子科技大学 | UAV image registration and splicing method based on bidirectional point characteristics |
-
2020
- 2020-12-18 CN CN202011504929.XA patent/CN112580506A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101034434A (en) * | 2007-04-10 | 2007-09-12 | 杭州电子科技大学 | Identification recognizing method based on binocular iris |
| CN111344703A (en) * | 2017-11-24 | 2020-06-26 | 三星电子株式会社 | User authentication device and method based on iris recognition |
| CN108010045A (en) * | 2017-12-08 | 2018-05-08 | 福州大学 | Visual pattern characteristic point error hiding method of purification based on ORB |
| CN111833249A (en) * | 2020-06-30 | 2020-10-27 | 电子科技大学 | UAV image registration and splicing method based on bidirectional point characteristics |
Non-Patent Citations (2)
| Title |
|---|
| 叶志坚 等: "一种改进的FREAK算法的图像特征点匹配", 《广东工业大学学报》, vol. 34, no. 6, 15 November 2017 (2017-11-15), pages 37 - 42 * |
| 陈天华 等: "实时鲁棒的特征点匹配算法", 《中国图象图形学报》, vol. 21, no. 9, 16 September 2016 (2016-09-16), pages 1213 - 1220 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111368683B (en) | Face image feature extraction method and face recognition method based on modular constraint CenterFace | |
| CN112016402B (en) | Self-adaptive method and device for pedestrian re-recognition field based on unsupervised learning | |
| CN103218621B (en) | The recognition methods of multiple dimensioned vehicle in a kind of life outdoor videos monitoring | |
| Huang et al. | DeepDiff: Learning deep difference features on human body parts for person re-identification | |
| CN110633289A (en) | License plate number determining method, server and parking management system | |
| CN106503694A (en) | Digit recognition method based on eight neighborhood feature | |
| Kabbai et al. | Image matching based on LBP and SIFT descriptor | |
| CN114862925A (en) | Image registration method, device and system based on SIFT and storage medium | |
| CN108830283B (en) | Image feature point matching method | |
| CN114943754A (en) | Image registration method, system and storage medium based on SIFT | |
| Yu et al. | A multi-task learning CNN for image steganalysis | |
| CN113743443B (en) | Image evidence classification and recognition method and device | |
| CN105224619B (en) | A kind of spatial relationship matching process and system suitable for video/image local feature | |
| CN112580506A (en) | Iris characteristic point comparison method based on bidirectional minimum Hamming distance | |
| Wang et al. | Text detection algorithm based on improved YOLOv3 | |
| CN115830357A (en) | Template matching method, system, device and storage medium | |
| CN112288045B (en) | Seal authenticity distinguishing method | |
| CN113554036A (en) | Characteristic point extraction and matching method for improving ORB algorithm | |
| Al-Sahaf et al. | Binary image classification using genetic programming based on local binary patterns | |
| CN116030511A (en) | Three-dimensional face recognition method based on ToF | |
| An et al. | Captcha recognition algorithm based on the relative shape context and point pattern matching | |
| CN109190741B (en) | Secret information management method based on QR (quick response) code and terminal | |
| Zhao et al. | Research on Feature Matching of an Improved ORB Algorithm | |
| Song et al. | Point symbol recognition algorithm based on improved generalized hough transform and nonlinear mapping | |
| Bouamra et al. | Off-line signature verification using multidirectional run-length features |
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 |