CN111753719A - Fingerprint identification method and device - Google Patents
Fingerprint identification method and device Download PDFInfo
- Publication number
- CN111753719A CN111753719A CN202010586734.8A CN202010586734A CN111753719A CN 111753719 A CN111753719 A CN 111753719A CN 202010586734 A CN202010586734 A CN 202010586734A CN 111753719 A CN111753719 A CN 111753719A
- Authority
- CN
- China
- Prior art keywords
- matched
- subgraph
- sub
- matching
- affine transformation
- 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 30
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims abstract description 164
- 230000009466 transformation Effects 0.000 claims abstract description 164
- 239000011159 matrix material Substances 0.000 claims abstract description 78
- 239000013598 vector Substances 0.000 claims abstract description 50
- 238000012216 screening Methods 0.000 claims abstract description 16
- 238000004364 calculation method Methods 0.000 claims description 18
- 238000004590 computer program Methods 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 14
- 238000010586 diagram Methods 0.000 description 16
- 238000005516 engineering process Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000011840 criminal investigation Methods 0.000 description 2
- 238000011835 investigation Methods 0.000 description 2
- 238000012790 confirmation Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000013519 translation 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
- G06V40/12—Fingerprints or palmprints
- G06V40/1347—Preprocessing; Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/42—Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
- Collating Specific Patterns (AREA)
Abstract
The application relates to the technical field of computers, in particular to a fingerprint identification method and a fingerprint identification device, which are used for identifying and obtaining each fingerprint feature point to be matched in a fingerprint image to be matched and respectively obtaining a subgraph to be matched containing the fingerprint feature point to be matched from the fingerprint image to be matched; respectively matching the subgraph characteristic vectors of the subgraphs to be matched according to the calculated subgraph characteristic vectors of the subgraphs to be matched to obtain the subgraphs with the subgraph characteristics meeting the similarity condition; according to the subgraphs meeting the similarity condition obtained by corresponding matching of the subgraphs to be matched, calculating an affine transformation matrix of the subgraphs, and screening out the subgraphs meeting the matching condition according to the affine transformation matrix; and determining the fingerprint matching result of the fingerprint image to be matched according to the screened sub-images meeting the matching conditions, so that the characteristic points which do not meet the matching conditions are filtered through the affine transformation matrix, and the accuracy of fingerprint identification can be improved.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a fingerprint identification method and apparatus.
Background
At present, a fingerprint identification system is widely applied, for example, fingerprint identification can be applied to civil investigation and criminal investigation application scenes, in the whole fingerprint identification system, fingerprint matching is a very important link of the fingerprint identification system, in the prior art, fingerprint matching can be realized by comparing a fingerprint image to be matched with a fingerprint image in a fingerprint database one by one, however, the method in the prior art may cause mismatching, and when mismatching occurs, the accuracy of fingerprint identification can be improved.
Disclosure of Invention
The embodiment of the application provides a fingerprint identification method and a fingerprint identification device, so that the accuracy of fingerprint identification is improved.
The embodiment of the application provides the following specific technical scheme:
a fingerprint identification method, comprising:
identifying and obtaining each fingerprint feature point to be matched in the fingerprint image to be matched, and respectively obtaining a sub-image to be matched containing the fingerprint feature point to be matched from the fingerprint image to be matched;
respectively matching the subgraph characteristic vectors of the subgraphs to be matched according to the calculated subgraph characteristic vectors of the subgraphs to be matched to obtain the subgraphs with the subgraph characteristics meeting the similarity condition;
according to the subgraphs meeting the similarity condition obtained by corresponding matching of the subgraphs to be matched, calculating an affine transformation matrix of the subgraphs, and screening out the subgraphs meeting the matching condition according to the affine transformation matrix;
and determining the fingerprint matching result of the fingerprint image to be matched according to the screened sub-images meeting the matching condition.
Optionally, if the similarity condition is the highest similarity, calculating an affine transformation matrix of the subgraph according to the subgraph meeting the similarity condition obtained by correspondingly matching the subgraphs to be matched, where the similarity condition includes:
randomly selecting a plurality of preset subgraphs to be matched and subgraph pairs with the highest similarity obtained by corresponding matching, and respectively calculating to obtain candidate affine transformation matrixes;
respectively carrying out affine transformation on each subgraph to be matched according to any candidate affine transformation matrix aiming at each candidate affine transformation matrix obtained through calculation, determining the subgraph corresponding to each subgraph to be matched after the affine transformation, if the first distance between the subgraph after the affine transformation and the subgraph with the highest similarity obtained through corresponding matching is smaller than a first distance threshold value, determining that the matching condition is met between the corresponding subgraph to be matched and the subgraph with the highest similarity obtained through matching, and determining the sum of the number of the subgraphs to be matched meeting the matching condition and the sum of the corresponding first distances;
and taking the candidate affine transformation matrix corresponding to the maximum sum of the number of the sub-images to be matched meeting the matching conditions in the candidate affine transformation matrixes obtained by calculation and the minimum sum of the corresponding first distances as the affine transformation matrix.
Optionally, screening out sub-graphs meeting the matching condition according to the affine transformation matrix specifically includes:
according to the affine transformation matrix, performing affine transformation on the sub-images to be matched respectively, and determining sub-images corresponding to the sub-images to be matched after the affine transformation;
and if the second distance between the sub-image after the affine transformation and the sub-image with the highest similarity obtained correspondingly is smaller than a preset second distance threshold, determining that the sub-image to be matched and the sub-image with the highest similarity obtained by matching meet the matching condition, and screening out the sub-image meeting the matching condition.
Optionally, matching the subgraph feature to obtain the subgraph whose subgraph feature meets the similarity condition according to the computed subgraph feature vector of each subgraph to be matched, specifically including:
and respectively aiming at each subgraph to be matched, finding out the subgraph class which is the same as the subgraph class according to the subgraph class obtained by the identification of any subgraph to be matched, respectively comparing the subgraph class with the central subgraph feature vector of each subgraph class according to the calculated subgraph feature vector of the subgraph to be matched, determining the subgraph class meeting the similarity condition, and matching the subgraph class meeting the similarity condition to obtain the subgraph with the subgraph feature meeting the similarity condition from the subgraph class meeting the similarity condition.
Optionally, determining a fingerprint matching result of the fingerprint image to be matched according to the screened sub-images meeting the matching condition, specifically including:
searching fingerprint images associated with the sub-images meeting the matching conditions according to the screened sub-images meeting the matching conditions;
respectively calculating the ratio of the number of subgraphs meeting the matching condition in any one fingerprint image to the total number of subgraphs of any one fingerprint image, wherein the subgraphs meet the matching condition, aiming at the searched fingerprint images which are related, and taking the ratio as the matching score of any one fingerprint image and the fingerprint image to be matched;
and according to the determined matching scores, determining a fingerprint image which is finally matched with the fingerprint image to be matched from the searched fingerprint images which are associated.
A fingerprint recognition device, comprising:
the identification module is used for identifying and obtaining each fingerprint feature point to be matched in the fingerprint image to be matched and respectively obtaining sub-images to be matched containing the fingerprint feature points to be matched from the fingerprint image to be matched;
the matching module is used for matching and obtaining the subgraph of which the subgraph characteristics meet the similarity condition according to the calculated subgraph characteristic vectors of the subgraphs to be matched;
the processing module is used for calculating an affine transformation matrix of each subgraph according to the subgraph meeting the similarity condition obtained by corresponding matching of each subgraph to be matched, and screening out the subgraphs meeting the matching condition according to the affine transformation matrix;
and the determining module is used for determining the fingerprint matching result of the fingerprint image to be matched according to the screened sub-images meeting the matching condition.
Optionally, if the similarity condition is the highest similarity, when the affine transformation matrix of the subgraph is calculated according to the subgraph meeting the similarity condition obtained by correspondingly matching each subgraph to be matched, the processing module is specifically configured to:
randomly selecting a plurality of preset subgraphs to be matched and subgraph pairs with the highest similarity obtained by corresponding matching, and respectively calculating to obtain candidate affine transformation matrixes;
respectively carrying out affine transformation on each subgraph to be matched according to any candidate affine transformation matrix aiming at each candidate affine transformation matrix obtained through calculation, determining the subgraph corresponding to each subgraph to be matched after the affine transformation, if the first distance between the subgraph after the affine transformation and the subgraph with the highest similarity obtained through corresponding matching is smaller than a first distance threshold value, determining that the matching condition is met between the corresponding subgraph to be matched and the subgraph with the highest similarity obtained through matching, and determining the sum of the number of the subgraphs to be matched meeting the matching condition and the sum of the corresponding first distances;
and taking the candidate affine transformation matrix corresponding to the maximum sum of the number of the sub-images to be matched meeting the matching conditions in the candidate affine transformation matrixes obtained by calculation and the minimum sum of the corresponding first distances as the affine transformation matrix.
Optionally, when a sub-graph satisfying the matching condition is screened out according to the affine transformation matrix, the processing module is specifically configured to:
according to the affine transformation matrix, performing affine transformation on the sub-images to be matched respectively, and determining sub-images corresponding to the sub-images to be matched after the affine transformation;
and if the second distance between the sub-image after the affine transformation and the sub-image with the highest similarity obtained correspondingly is smaller than a preset second distance threshold, determining that the sub-image to be matched and the sub-image with the highest similarity obtained by matching meet the matching condition, and screening out the sub-image meeting the matching condition.
Optionally, the matching module is specifically configured to:
and respectively aiming at each subgraph to be matched, finding out the subgraph class which is the same as the subgraph class according to the subgraph class obtained by the identification of any subgraph to be matched, respectively comparing the subgraph class with the central subgraph feature vector of each subgraph class according to the calculated subgraph feature vector of the subgraph to be matched, determining the subgraph class meeting the similarity condition, and matching the subgraph class meeting the similarity condition to obtain the subgraph with the subgraph feature meeting the similarity condition from the subgraph class meeting the similarity condition.
Optionally, the determining module is specifically configured to:
searching fingerprint images associated with the sub-images meeting the matching conditions according to the screened sub-images meeting the matching conditions;
respectively calculating the ratio of the number of subgraphs meeting the matching condition in any one fingerprint image to the total number of subgraphs of any one fingerprint image, wherein the subgraphs meet the matching condition, aiming at the searched fingerprint images which are related, and taking the ratio as the matching score of any one fingerprint image and the fingerprint image to be matched;
and according to the determined matching scores, determining a fingerprint image which is finally matched with the fingerprint image to be matched from the searched fingerprint images which are associated.
An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the above fingerprint identification method when executing the program.
A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned fingerprint identification method.
In the embodiment of the application, each fingerprint feature point to be matched of the fingerprint image to be matched is identified and obtained, subgraphs to be matched containing the fingerprint feature point to be matched are obtained from the fingerprint image to be matched respectively, subgraphs with subgraph features meeting similarity conditions are obtained by matching according to each calculated subgraph feature vector of each subgraph to be matched respectively, an affine transformation matrix of the subgraph is calculated according to the subgraph meeting the similarity conditions obtained by corresponding matching of each subgraph to be matched, the subgraph meeting the matching conditions is screened out according to the affine transformation matrix, the fingerprint matching result of the fingerprint image to be matched is determined according to each screened subgraph meeting the matching conditions, thus, the subgraph feature vector of the subgraph to be matched is compared with the subgraph feature vector of each subgraph of each fingerprint image to be matched, the subgraph matched with the subgraph to be matched can be determined from each subgraph, and then fingerprint identification is carried out on the fingerprint image to be matched, after each matched subgraph is obtained, affine transformation is carried out on each matched subgraph, the subgraph which is wrongly matched is filtered, and then the fingerprint matching result of the fingerprint image to be matched is determined according to the matched subgraph, so that the accuracy in fingerprint identification can be improved.
Drawings
FIG. 1 is a flowchart of a fingerprint matching method based on affine transformation according to an embodiment of the present application;
FIG. 2 is a feature point classification diagram in the embodiment of the present application;
FIG. 3 is a schematic diagram of an affine transformation of a fingerprint in an embodiment of the present application;
FIG. 4 is a flow chart of another fingerprint identification method in an embodiment of the present application;
FIG. 5 is a schematic diagram of a fingerprint identification device according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of an electronic device in an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
At present, the fingerprint identification technology is the most widely applied biometric identification technology, and the identification of an operation or an operated person is mainly performed according to the information of lines, detail features and the like of human fingerprints, so the application of the fingerprint identification system is very wide, for example, the fingerprint identification can be applied to the application scenes of civil investigation and criminal investigation, the application field of the fingerprint identification technology is increasingly wide in the whole fingerprint identification system, the fingerprint matching is a very important link of the fingerprint identification system, in the prior art, the fingerprint matching can be realized by comparing the fingerprint image to be matched with the fingerprint image in the fingerprint database one by one, and then the fingerprint image matched with the fingerprint image to be matched is obtained, however, the method in the prior art is not the fingerprint image matched with each other originally because the condition of error matching can occur during the fingerprint identification, may be erroneously matched and, when an erroneous match occurs, the accuracy of fingerprint recognition may be reduced.
In the embodiment of the application, each fingerprint feature point to be matched in the fingerprint image to be matched is identified and obtained, a subgraph to be matched containing the fingerprint feature point to be matched is obtained from the fingerprint image to be matched, the subgraph with subgraph features meeting the similarity condition is obtained by matching according to the calculated subgraph feature vector of the subgraph to be matched, the subgraph meeting the similarity condition is obtained by corresponding matching according to the subgraph to be matched, an affine transformation matrix of the subgraph is calculated, the subgraph meeting the matching condition is screened out according to the affine transformation matrix, the fingerprint matching result of the fingerprint image to be matched is determined according to the screened subgraph meeting the matching condition, thus, each subgraph to be matched is compared with each subgraph of each fingerprint image respectively, the matched subgraph is further determined, the identification of the fingerprint can be realized, and through affine transformation, when mismatching occurs, the mismatching subgraphs can be filtered, and the accuracy of fingerprint identification is further improved.
Based on the foregoing embodiment, referring to fig. 1, a flowchart of a fingerprint matching method based on affine transformation in the embodiment of the present application is specifically included:
step 100: and identifying and obtaining each fingerprint feature point to be matched in the fingerprint image to be matched, and respectively obtaining the subgraph to be matched containing the fingerprint feature point to be matched from the fingerprint image to be matched.
In the embodiment of the application, each to-be-matched fingerprint feature point of a to-be-matched fingerprint image is obtained through identification, meanwhile, the position coordinate of each to-be-matched fingerprint feature point on the to-be-matched fingerprint image can be obtained through identification, the to-be-matched fingerprint image is intercepted according to the position coordinate of each to-be-matched fingerprint feature point on the to-be-matched fingerprint image, a fingerprint area with a preset size and containing the fingerprint feature points is intercepted, a corresponding intercepted fingerprint area is obtained, then, each intercepted fingerprint area is used as a to-be-matched sub-image corresponding to the to-be-matched fingerprint feature point, namely, the to-be-matched sub-image in the embodiment of the application contains the fingerprint image corresponding to the to-be-matched fingerprint feature point, and also contains a part of fingerprint image of the to-be-matched fingerprint.
The preset size is set according to actual requirements, for example, the preset size is 64 × 64, which is not limited in the embodiment of the present application.
Further, when the densities of the subgraph to be matched and the subgraph of the fingerprint image in the fingerprint database are not consistent, each corresponding subgraph to be matched is generated based on the binary graph, then, the density width corresponding to each subgraph to be matched is determined, when the density width is determined not to meet the preset density width range, the density calibration is carried out on the subgraph to be matched, then, a preset neural network model is adopted to determine corresponding image characteristic information, the density calibration is carried out on the subgraph to be matched which does not meet the conditions, the subgraph to be matched which corresponds to each fingerprint characteristic point is adopted instead of directly adopting the binary graph to carry out the subsequent matching process, the redundant information is reduced, the influence of the redundant information in the fingerprint image on the fingerprint matching result is avoided, and the consistency of the densities of the subgraph to be matched and the subgraph of the fingerprint image in the fingerprint database is ensured, thereby improving the accuracy of fingerprint identification.
Step 110: and respectively matching the subgraph characteristic vectors of the subgraphs to be matched according to the calculated subgraph characteristic vectors of the subgraphs to be matched to obtain the subgraphs with the subgraph characteristics meeting the similarity condition.
In the embodiment of the application, for each sub-image to be matched, a sub-image class the same as the sub-image class is found according to a sub-image class obtained by identifying any one sub-image to be matched, the sub-image class is compared with the central sub-image feature vector of each sub-image class according to the calculated sub-image feature vector of the sub-image to be matched, the sub-image class meeting the similarity condition is determined, and the sub-image with the sub-image feature meeting the similarity condition is obtained by matching from the sub-image class meeting the similarity condition.
For example, the sub-graph class with the highest similarity may be determined, and the sub-graph class with the highest similarity is obtained through matching, which is not limited in the embodiment of the present application.
Step 120: and calculating an affine transformation matrix of the subgraph according to the subgraphs meeting the similarity condition obtained by correspondingly matching each subgraph to be matched, and screening the subgraphs meeting the matching condition according to the affine transformation matrix.
Step 130: and determining the fingerprint matching result of the fingerprint image to be matched according to the screened sub-images meeting the matching condition.
In the embodiment of the application, after each subgraph meeting the matching condition is screened out, the fingerprint matching result of the fingerprint image to be matched is determined according to each subgraph meeting the matching condition.
In this embodiment of the present application, before screening out each sub-graph that satisfies the matching condition, a sub-graph whose sub-graph features satisfy the similarity condition is obtained first, and the following example takes the similarity condition as the highest similarity, and the step of obtaining a sub-graph whose sub-graph features satisfy the similarity condition by matching in step 110 in this embodiment of the present application according to the calculated sub-graph feature vector of each sub-graph to be matched is described in detail, which may be specifically divided into the following two ways.
The first mode specifically includes:
s1: and respectively aiming at each subgraph to be matched, finding out the subgraph class which is the same as the subgraph class according to the subgraph class obtained by the identification of any subgraph to be matched, respectively comparing the subgraph class with the central subgraph feature vector of each subgraph class according to the calculated subgraph feature vector of the subgraph to be matched, determining the subgraph class meeting the similarity condition, and matching the subgraph class meeting the similarity condition to obtain the subgraph with the subgraph feature meeting the similarity condition from the subgraph class meeting the similarity condition.
In the embodiment of the application, for each sub-image to be matched, a sub-image feature vector of any sub-image to be matched is compared with sub-image feature vectors of each sub-image of each fingerprint image, distances between the sub-image feature vector of any sub-image to be matched and sub-image feature vectors of each sub-image of each preset fingerprint image are determined, distances smaller than a preset distance threshold value are screened out from the determined distances, sub-images corresponding to the distances smaller than the preset distance threshold value are determined according to the distances smaller than the preset distance threshold value, and the sub-image with the highest sub-image feature similarity is selected from the sub-images to serve as the sub-image matched with the sub-image to be matched.
The second mode specifically includes:
s1: and identifying the subgraph class of the subgraph to be matched.
Wherein the sub-graph type is a center point, an end point, a cross point, an arc point, a triangle point or a hole point.
Because the lines of the fingerprint image are not continuous, smooth and straight, but are interrupted, branched or broken frequently, these break points, branched points and turning points are called fingerprint feature points, and meanwhile, these fingerprint feature points provide confirmation information of fingerprint uniqueness, and the classification of the feature points includes the following several categories, most typically, end points and branched points, as shown in fig. 2, which is a feature point classification diagram in the embodiment of the present application.
Further, in the embodiment of the application, when the to-be-matched fingerprint feature points of the to-be-matched fingerprint image are identified, the to-be-matched fingerprint feature points of the to-be-matched image are respectively input into the trained type identification model, the fingerprint types corresponding to the to-be-matched fingerprint feature points are identified, and the fingerprint types corresponding to the to-be-matched fingerprint feature points are used as the sub-graph types of the to-be-matched sub-graphs.
S2: and respectively searching the subgraph class which is the same as the subgraph class of any subgraph to be matched according to the subgraph class of any subgraph to be matched.
In the embodiment of the application, each sub-graph class in the fingerprint database is searched according to the sub-graph class of the sub-graph to be matched, and the sub-graph class which is the same as the sub-graph class of any sub-graph to be matched is searched.
For example, if the sub-graph category of the sub-graph to be matched is the central point, the sub-graph category of which the sub-graph category is also the central point is found from the fingerprint database.
S3: and comparing the sub-image feature vectors of the sub-images to be matched with the central sub-image feature vectors of the sub-images respectively according to the calculated sub-image feature vectors of the sub-images to be matched, and determining the sub-image class which meets the highest similarity.
In the embodiment of the application, each sub-image class corresponds to one central sub-image feature vector, the sub-image feature vectors of the sub-images to be matched are compared with the central sub-image feature vectors of the sub-image classes with the same sub-image class, and then the sub-image class meeting the highest similarity is determined, so that the calculation amount of fingerprint identification can be reduced, and the fingerprint identification efficiency is improved.
For example, by using the euclidean distance algorithm, in the embodiment of the present application, after each first distance is determined, it is determined whether each determined first distance is smaller than a preset first distance threshold, and if it is determined that a sub-graph class whose first distance is smaller than the preset first distance threshold exists, a sub-graph class corresponding to the smallest first distance is determined from sub-graph classes smaller than the preset first distance threshold, and the sub-graph class is used as a sub-graph class with the highest similarity to the sub-graph to be matched.
S4: and matching to obtain the subgraph with the highest subgraph feature similarity from the subgraph classes meeting the similarity condition.
In the embodiment of the application, after the sub-graph set corresponding to the central sub-graph feature is obtained, the sub-graph features to be matched are respectively compared with the sub-graph features of each sub-graph in the sub-set one by one.
For example, a preset distance algorithm is adopted to respectively calculate a second distance between the sub-graph feature vector of the sub-graph to be matched and the sub-graph feature vector of each sub-graph in the sub-graph class with the highest similarity, after each second distance is obtained, the second distance is smaller than a preset second distance threshold, the sub-graph with the minimum second distance is used as the sub-graph with the highest similarity with the sub-graph feature of any sub-graph to be matched, and the sub-graph is determined to be matched with the sub-graph to be matched.
The sub-graph features at least comprise various information features such as textures, edges, angles and the like.
After obtaining the subgraph with the subgraph characteristics meeting the similarity condition through matching, performing affine transformation on the subgraph meeting the similarity condition, and filtering out subgraphs which do not meet the matching condition, then executing the step 120 to screen out the subgraph meeting the matching condition according to the affine transformation matrix, which specifically comprises the following steps:
s1: and performing affine transformation on the sub-images to be matched respectively according to the affine transformation matrix, and determining the sub-images which correspond to the sub-images to be matched and are subjected to affine transformation.
S3: and if the second distance between the sub-image after the affine transformation and the sub-image with the highest similarity obtained correspondingly is smaller than a preset second distance threshold, determining that the sub-image to be matched and the sub-image with the highest similarity obtained by matching meet the matching condition, and screening the sub-image meeting the matching condition.
In the embodiment of the application, after affine transformation is performed on a subgraph to be matched through a calculated affine transformation matrix to obtain a subgraph after the affine transformation, a second distance between the subgraph after the affine transformation and the subgraph with the highest corresponding obtained similarity is calculated, if the second distance between the subgraph after the affine transformation and the subgraph with the highest corresponding obtained similarity is determined to be smaller than a preset second distance threshold, it is determined that the corresponding subgraph to be matched and the subgraph with the highest matching obtained similarity satisfy a matching condition, namely, the matching is correct, and if the second distance is determined to be larger than or equal to the preset second distance threshold, it is determined that the corresponding subgraph to be matched and the subgraph with the highest matching obtained similarity do not satisfy the matching condition, namely, the matching is wrong, and then the subgraph satisfying the matching condition is screened.
For example, referring to fig. 3, which is a schematic diagram of affine transformation of fingerprints in this embodiment, A, B, C, D are to-be-matched subgraphs of to-be-matched fingerprint images, respectively, after affine transformation is performed on the to-be-matched subgraph A, B, C, D, the obtained subgraphs after affine transformation are a ', B ', C ', and D ', respectively, and the actually obtained subgraphs with the highest similarity are P, B ', C ', and D ', respectively, a second distance between the subgraph after affine transformation and the corresponding obtained subgraph with the highest similarity is calculated, that is, the distances between a ' P, B ' B ', C ', and D ' are calculated, and after second distance calculation, a second distance between the subgraph a ' after affine transformation and the obtained subgraph P with the highest similarity is calculated and is greater than a preset second distance threshold, it is determined that the subgraph a to be matched is incorrect, and the subgraph B ' after affine transformation, B ', C ', and D ' are obtained by affine transformation, and if the second distance between the sub-images B ', C ' and D ' with the highest similarity is smaller than a preset second distance threshold, determining that the sub-image B, C, D to be matched meets the matching condition and the matching is correct.
In this embodiment of the present application, an incorrectly matched sub-graph in each matched sub-graph may be filtered according to an affine transformation matrix of the sub-graph to be matched, and then a sub-graph satisfying a matching condition is determined from the matched sub-graphs, so that, in the process of performing affine transformation, an affine transformation matrix of the sub-graph to be matched needs to be used, and the following example takes a similarity condition as a highest similarity, and details a step of calculating an affine transformation matrix of the sub-graph according to the sub-graph satisfying the similarity condition obtained by corresponding matching of each sub-graph to be matched in step 120 in this embodiment of the present application are specifically described, where:
s1: and randomly selecting a plurality of preset subgraphs to be matched and the subgraph pairs with the highest similarity obtained by corresponding matching, and respectively calculating to obtain candidate affine transformation matrixes.
In the embodiment of the application, a plurality of preset subgraphs to be matched and subgraph pairs with the highest similarity obtained by corresponding matching are randomly selected, and candidate affine transformation matrixes are respectively obtained through calculation according to the plurality of preset subgraphs to be matched and the subgraph pairs with the highest similarity obtained by corresponding matching.
For example, taking the coordinate origin at [1,0] as an example, for a position (x, y) in the coordinate system, the projection of the position relative to the coordinate origin in the [1,0] direction is x, and the projection in the [0,1] direction is y, where the projection means the distance from the intersection of the parallel lines with the coordinate axes (x, y) to the coordinate axis to the origin, that is, (x, y) actually is:
when the coordinate system changes, the coordinates of each fingerprint feature point in the coordinate system also changes, but the position of the fingerprint feature point relative to the new coordinate system (x '-y' coordinate system) is not changed, and is still (x, y), and the basis vector of the new coordinate axis becomes [ cos (theta), sin (theta) ] by taking affine transformation as an example.
The new position and new basis vector are relative to the absolute coordinate system (x-y coordinate system), and the other candidate affine transformation matrices are the same.
S2: respectively carrying out affine transformation on each subgraph to be matched according to any candidate affine transformation matrix aiming at each candidate affine transformation matrix obtained through calculation, determining the subgraph corresponding to each subgraph to be matched after the affine transformation, if the first distance between the subgraph after the affine transformation and the subgraph with the highest similarity obtained through corresponding matching is smaller than a first distance threshold value, determining that the matching condition is met between the corresponding subgraph to be matched and the subgraph with the highest similarity obtained through matching, and determining the sum of the number of the subgraphs to be matched meeting the matching condition and the sum of the corresponding first distances.
Wherein, the affine transformation comprises one or any combination of the following: translation, scaling, flipping, rotation, or miscut.
In the embodiment of the application, after the obtained candidate affine transformation matrix, according to any candidate affine transformation matrix, performing affine transformation on each sub-image to be matched of the fingerprint image to be matched respectively to obtain the sub-image corresponding to each sub-image to be matched after the affine transformation.
For example, each sub-image to be matched is translated through any candidate affine transformation matrix by using an affine transformation relation to obtain a corresponding relation of spatial positions of each sub-image to be matched, and further, the sub-image and the coordinates thereof corresponding to each sub-image to be matched after affine transformation are obtained.
Then, after obtaining sub-images and coordinates thereof corresponding to the sub-images to be matched after affine transformation, respectively calculating first distances between the sub-images corresponding to the sub-images to be matched after affine transformation and the corresponding sub-images to be matched, if the first distances are determined to be smaller than a preset first distance threshold, determining that matching conditions are met between the corresponding sub-images to be matched and the sub-images with the highest similarity obtained by matching, namely the matching is correct, the matched sub-images are the sub-images with correct matching, adding the first distances of the sub-images to be matched meeting the matching conditions to obtain a first distance sum obtained based on the candidate affine transformation matrix, and calculating the number sum of the sub-images to be matched meeting the matching conditions.
S3: and taking the candidate affine transformation matrix corresponding to the maximum sum of the number of the subgraphs to be matched meeting the matching conditions in the candidate affine transformation matrices obtained by calculation and the minimum sum of the corresponding first distances as the affine transformation matrix.
In the embodiment of the application, the candidate affine transformation matrix corresponding to the candidate affine transformation matrix which is obtained by calculation and satisfies the matching condition and has the largest sum of the number of the to-be-matched subgraphs and the smallest sum of the first distances is used as the affine transformation matrix of the to-be-matched fingerprint image.
After obtaining sub-images of which the sub-images to be matched satisfy the matching conditions, determining a fingerprint matching result of the fingerprint image to be matched according to the determined sub-images satisfying the matching conditions, and the following exemplary explanation of the fingerprint matching result of the fingerprint image to be matched determined in the step 130 in the embodiment of the present application specifically includes:
s1: and searching fingerprint images associated with the subgraphs meeting the matching conditions according to the screened subgraphs meeting the matching conditions.
In the embodiment of the application, the fingerprint image associated with each subgraph meeting the matching condition is found according to each screened subgraph meeting the matching condition, each subgraph number of each subgraph meeting the matching condition and the association relationship between the subgraph number and the fingerprint image number.
S2: and respectively calculating the ratio of the number of the subgraphs meeting the matching condition in any one fingerprint image to the total number of the subgraphs contained in any one fingerprint image aiming at the searched related fingerprint images, and taking the ratio as the matching score of any one fingerprint image and the fingerprint image to be matched.
In the embodiment of the application, the ratio between the number of the subgraphs meeting the matching condition in any one fingerprint image and the total number of the subgraphs contained in any one fingerprint image is calculated respectively for each found related fingerprint image, and the calculated ratio is used as the matching score of any one fingerprint image and the fingerprint image to be matched.
For example, assuming that the number of subgraphs satisfying the matching condition and the total number of subgraphs included in the fingerprint image a are 80 and 100, respectively, based on the obtained total number of subgraphs 80 satisfying the matching condition and the total number of subgraphs included in any one of the fingerprint images 100, the ratio between the fingerprint image to be matched and the fingerprint image a is calculated to be 80/100, that is, 0.8, and then the calculated ratio is taken as the matching score of the fingerprint image a and the fingerprint image to be matched.
S3: and according to the determined matching scores, determining a fingerprint image finally matched with the fingerprint image to be matched from the searched associated fingerprint images.
In the embodiment of the application, according to the determined matching scores, the fingerprint image with the highest matching score is determined from the searched associated fingerprint images and is used as the fingerprint image which is finally matched with the fingerprint image to be matched.
Further, in this embodiment of the application, the N fingerprint images with the highest determined matching scores may also be used as corresponding fingerprint matching results.
Wherein N is a preset positive integer.
For example, assuming that the value of N is 2, the similarity between the fingerprint image to be matched and the fingerprint image 1, the fingerprint image 2, and the fingerprint image 3 in the fingerprint database is 0.2, 0.4, and 0.6, respectively, based on the image similarity between the fingerprint image to be matched and the fingerprint image 1, the fingerprint image 2, and the fingerprint image 3, the 2 fingerprint images with the highest value of image similarity are used as the corresponding fingerprint identification results, that is, the fingerprint images 2 and 3 are used as the corresponding fingerprint identification results.
In the embodiment of the application, each fingerprint feature point to be matched in the fingerprint image to be matched is identified and obtained, a subgraph to be matched containing the corresponding fingerprint feature point to be matched is obtained from the fingerprint image to be matched, a subgraph with subgraph features meeting the similarity condition is obtained by matching according to the calculated subgraph feature vector of each subgraph to be matched, an affine transformation matrix of the subgraph is calculated according to the subgraph meeting the similarity condition obtained by corresponding matching of each subgraph to be matched, the subgraph meeting the matching condition is screened out according to the affine transformation matrix, the fingerprint matching result of the fingerprint image to be matched is determined according to each screened subgraph meeting the matching condition, thus, the matched subgraph can be determined from each subgraph by comparing the subgraph feature vector of the subgraph to be matched with the subgraph feature vector of each subgraph in the subgraph class with the same subgraph class, and after each matched subgraph is obtained, affine transformation is carried out on each matched subgraph, then the subgraph which is wrongly matched is filtered, and then the fingerprint matching result of the fingerprint image to be matched is determined according to the subgraph which meets the matching condition, so that the accuracy in fingerprint identification can be improved.
Based on the foregoing embodiment, referring to fig. 4, a flowchart of another fingerprint identification method in the embodiment of the present application is specifically included:
step 400: and identifying and obtaining each fingerprint feature point to be matched in the fingerprint image to be matched, and respectively obtaining the subgraph to be matched containing the fingerprint feature point to be matched from the fingerprint image to be matched.
Step 401: and respectively searching the subgraph class which is the same as the subgraph class according to the subgraph class obtained by the identification of any subgraph to be matched.
Step 402: and comparing the sub-image feature vectors of the sub-images to be matched with the central sub-image feature vectors of the sub-images respectively according to the calculated sub-image feature vectors of the sub-images to be matched, and determining the sub-images meeting the similarity condition.
Step 403: and matching to obtain the subgraph with the subgraph characteristics meeting the similarity condition from the subgraph classes meeting the similarity condition.
Step 404: and randomly selecting a plurality of preset subgraphs to be matched and the subgraph pairs with the highest similarity obtained by corresponding matching, and respectively calculating to obtain candidate affine transformation matrixes.
Step 405: and respectively carrying out affine transformation on each subgraph to be matched according to any candidate affine transformation matrix aiming at each candidate affine transformation matrix obtained by calculation, and determining the subgraph after the affine transformation corresponding to each subgraph to be matched.
Step 406: and if the first distance between the sub-image after the affine transformation and the sub-image which is obtained by corresponding matching and has the highest similarity is smaller than the first distance threshold, determining that the matching condition is met between the corresponding sub-image to be matched and the sub-image which is obtained by matching and has the highest similarity, and determining the sum of the number of the sub-images to be matched which meet the matching condition and the sum of the corresponding first distances.
Step 407: and taking the candidate affine transformation matrix corresponding to the maximum sum of the number of the subgraphs to be matched meeting the matching conditions in the candidate affine transformation matrices obtained by calculation and the minimum sum of the corresponding first distances as the affine transformation matrix.
Step 408: and performing affine transformation on the sub-images to be matched respectively according to the affine transformation matrix, and determining the sub-images which correspond to the sub-images to be matched and are subjected to affine transformation.
Step 409: and if the second distance between the sub-image after the affine transformation and the sub-image with the highest similarity obtained correspondingly is smaller than a preset second distance threshold, determining that the sub-image to be matched and the sub-image with the highest similarity obtained by matching meet the matching condition, and screening the sub-image meeting the matching condition.
Step 410: and searching the fingerprint image associated with each subgraph meeting the matching conditions according to each screened subgraph meeting the matching conditions.
Step 411: and respectively calculating the ratio of the number of the subgraphs meeting the matching condition in any one fingerprint image to the total number of the subgraphs of any one fingerprint image, wherein the subgraphs contain the subgraphs, aiming at the searched fingerprint images which are related, and taking the ratio as the matching score of any one fingerprint image and the fingerprint image to be matched.
Step 412: and according to the determined matching scores, determining a fingerprint image finally matched with the fingerprint image to be matched from the searched associated fingerprint images.
In the embodiment of the application, sub-graphs to be matched and all sub-graphs in a fingerprint database do not need to be compared one by one, identification of fingerprint images to be matched can be achieved only by comparing the sub-graphs to be matched with the sub-graphs with the same class, efficiency of fingerprint identification can be improved, and the sub-graphs which are mistakenly matched in the matched sub-graphs are screened out through affine transformation, so that the sub-graphs meeting matching conditions are obtained, and accuracy of fingerprint identification can be improved.
Based on the same inventive concept, the embodiment of the present application provides a fingerprint identification device, which may be a hardware structure, a software module, or a hardware structure plus a software module. Based on the above embodiments, referring to fig. 5, a schematic structural diagram of a fingerprint identification device in the embodiment of the present application is shown, which specifically includes:
the identification module 500 is configured to identify and obtain each to-be-matched fingerprint feature point in a to-be-matched fingerprint image, and obtain to-be-matched subgraphs including the to-be-matched fingerprint feature points from the to-be-matched fingerprint image respectively;
a matching module 510, configured to match the sub-graph feature vectors of the sub-graphs to be matched to obtain sub-graphs whose sub-graph features satisfy a similarity condition;
the processing module 520 is configured to calculate an affine transformation matrix of each sub-graph according to the sub-graphs meeting the similarity condition obtained by corresponding matching of each sub-graph to be matched, and screen out the sub-graphs meeting the matching condition according to the affine transformation matrix;
and the determining module 530 is configured to determine a fingerprint matching result of the fingerprint image to be matched according to each screened sub-image meeting the matching condition.
Optionally, if the similarity condition is the highest similarity, when the affine transformation matrix of the subgraph is calculated according to the subgraph meeting the similarity condition obtained by correspondingly matching each subgraph to be matched, the processing module 520 is specifically configured to:
randomly selecting a plurality of preset subgraphs to be matched and subgraph pairs with the highest similarity obtained by corresponding matching, and respectively calculating to obtain candidate affine transformation matrixes;
respectively carrying out affine transformation on each subgraph to be matched according to any candidate affine transformation matrix aiming at each candidate affine transformation matrix obtained through calculation, determining the subgraph corresponding to each subgraph to be matched after the affine transformation, if the first distance between the subgraph after the affine transformation and the subgraph with the highest similarity obtained through corresponding matching is smaller than a first distance threshold value, determining that the matching condition is met between the corresponding subgraph to be matched and the subgraph with the highest similarity obtained through matching, and determining the sum of the number of the subgraphs to be matched meeting the matching condition and the sum of the corresponding first distances;
and taking the candidate affine transformation matrix corresponding to the maximum sum of the number of the sub-images to be matched meeting the matching conditions in the candidate affine transformation matrixes obtained by calculation and the minimum sum of the corresponding first distances as the affine transformation matrix.
Optionally, when a sub-graph satisfying the matching condition is screened out according to the affine transformation matrix, the processing module 520 is specifically configured to:
according to the affine transformation matrix, performing affine transformation on the sub-images to be matched respectively, and determining sub-images corresponding to the sub-images to be matched after the affine transformation;
and if the second distance between the sub-image after the affine transformation and the sub-image with the highest similarity obtained correspondingly is smaller than a preset second distance threshold, determining that the sub-image to be matched and the sub-image with the highest similarity obtained by matching meet the matching condition, and screening out the sub-image meeting the matching condition.
Optionally, the matching module 510 is specifically configured to:
and respectively aiming at each subgraph to be matched, finding out the subgraph class which is the same as the subgraph class according to the subgraph class obtained by the identification of any subgraph to be matched, respectively comparing the subgraph class with the central subgraph feature vector of each subgraph class according to the calculated subgraph feature vector of the subgraph to be matched, determining the subgraph class meeting the similarity condition, and matching the subgraph class meeting the similarity condition to obtain the subgraph with the subgraph feature meeting the similarity condition from the subgraph class meeting the similarity condition.
Optionally, the determining module 530 is specifically configured to:
searching fingerprint images associated with the sub-images meeting the matching conditions according to the screened sub-images meeting the matching conditions;
respectively calculating the ratio of the number of subgraphs meeting the matching condition in any one fingerprint image to the total number of subgraphs of any one fingerprint image, wherein the subgraphs meet the matching condition, aiming at the searched fingerprint images which are related, and taking the ratio as the matching score of any one fingerprint image and the fingerprint image to be matched;
and according to the determined matching scores, determining a fingerprint image which is finally matched with the fingerprint image to be matched from the searched fingerprint images which are associated.
Based on the above embodiments, referring to fig. 6, a schematic structural diagram of an electronic device in an embodiment of the present application is shown.
An embodiment of the present application provides an electronic device, which may include a processor 610 (central processing Unit, CPU), a memory 620, an input device 630, an output device 640, and the like, wherein the input device 630 may include a keyboard, a mouse, a touch screen, and the like, and the output device 640 may include a display device, such as a Liquid Crystal Display (LCD), a Cathode Ray Tube (CRT), and the like.
The processor 610 is configured to execute any one of the fingerprinting methods according to the embodiments of the present application by calling the program instructions stored in the memory 620 by the processor 610.
Based on the above embodiments, in the embodiments of the present application, a computer-readable storage medium is provided, on which a computer program is stored, and the computer program, when executed by a processor, implements the fingerprint identification method in any of the above method embodiments.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.
Claims (10)
1. A fingerprint identification method, comprising:
identifying and obtaining each fingerprint feature point to be matched in the fingerprint image to be matched, and respectively obtaining a sub-image to be matched containing the fingerprint feature point to be matched from the fingerprint image to be matched;
respectively matching the subgraph characteristic vectors of the subgraphs to be matched according to the calculated subgraph characteristic vectors of the subgraphs to be matched to obtain the subgraphs with the subgraph characteristics meeting the similarity condition;
according to the subgraphs meeting the similarity condition obtained by corresponding matching of the subgraphs to be matched, calculating an affine transformation matrix of the subgraphs, and screening out the subgraphs meeting the matching condition according to the affine transformation matrix;
and determining the fingerprint matching result of the fingerprint image to be matched according to the screened sub-images meeting the matching condition.
2. The method according to claim 1, wherein if the similarity condition is the highest similarity, calculating an affine transformation matrix of each sub-graph according to the sub-graphs meeting the similarity condition obtained by corresponding matching of the sub-graphs to be matched, specifically comprising:
randomly selecting a plurality of preset subgraphs to be matched and subgraph pairs with the highest similarity obtained by corresponding matching, and respectively calculating to obtain candidate affine transformation matrixes;
respectively carrying out affine transformation on each subgraph to be matched according to any candidate affine transformation matrix aiming at each candidate affine transformation matrix obtained through calculation, determining the subgraph corresponding to each subgraph to be matched after the affine transformation, if the first distance between the subgraph after the affine transformation and the subgraph with the highest similarity obtained through corresponding matching is smaller than a first distance threshold value, determining that the matching condition is met between the corresponding subgraph to be matched and the subgraph with the highest similarity obtained through matching, and determining the sum of the number of the subgraphs to be matched meeting the matching condition and the sum of the corresponding first distances;
and taking the candidate affine transformation matrix corresponding to the maximum sum of the number of the sub-images to be matched meeting the matching conditions in the candidate affine transformation matrixes obtained by calculation and the minimum sum of the corresponding first distances as the affine transformation matrix.
3. The method according to claim 2, wherein the screening out the subgraph satisfying the matching condition according to the affine transformation matrix specifically comprises:
according to the affine transformation matrix, performing affine transformation on the sub-images to be matched respectively, and determining sub-images corresponding to the sub-images to be matched after the affine transformation;
and if the second distance between the sub-image after the affine transformation and the sub-image with the highest similarity obtained correspondingly is smaller than a preset second distance threshold, determining that the sub-image to be matched and the sub-image with the highest similarity obtained by matching meet the matching condition, and screening out the sub-image meeting the matching condition.
4. The method of claim 1, wherein the sub-graphs whose sub-graph features satisfy the similarity condition are obtained by matching according to the calculated sub-graph feature vectors of the sub-graphs to be matched, respectively, and specifically includes:
and respectively aiming at each subgraph to be matched, finding out the subgraph class which is the same as the subgraph class according to the subgraph class obtained by the identification of any subgraph to be matched, respectively comparing the subgraph class with the central subgraph feature vector of each subgraph class according to the calculated subgraph feature vector of the subgraph to be matched, determining the subgraph class meeting the similarity condition, and matching the subgraph class meeting the similarity condition to obtain the subgraph with the subgraph feature meeting the similarity condition from the subgraph class meeting the similarity condition.
5. The method according to claim 4, wherein determining the fingerprint matching result of the fingerprint image to be matched according to the screened sub-images meeting the matching condition specifically comprises:
searching fingerprint images associated with the sub-images meeting the matching conditions according to the screened sub-images meeting the matching conditions;
respectively calculating the ratio of the number of subgraphs meeting the matching condition in any one fingerprint image to the total number of subgraphs of any one fingerprint image, wherein the subgraphs meet the matching condition, aiming at the searched fingerprint images which are related, and taking the ratio as the matching score of any one fingerprint image and the fingerprint image to be matched;
and according to the determined matching scores, determining a fingerprint image which is finally matched with the fingerprint image to be matched from the searched fingerprint images which are associated.
6. A fingerprint recognition device, comprising:
the identification module is used for identifying and obtaining each fingerprint feature point to be matched in the fingerprint image to be matched and respectively obtaining sub-images to be matched containing the fingerprint feature points to be matched from the fingerprint image to be matched;
the matching module is used for matching and obtaining the subgraph of which the subgraph characteristics meet the similarity condition according to the calculated subgraph characteristic vectors of the subgraphs to be matched;
the processing module is used for calculating an affine transformation matrix of each subgraph according to the subgraph meeting the similarity condition obtained by corresponding matching of each subgraph to be matched, and screening out the subgraphs meeting the matching condition according to the affine transformation matrix;
and the determining module is used for determining the fingerprint matching result of the fingerprint image to be matched according to the screened sub-images meeting the matching condition.
7. The apparatus according to claim 6, wherein if the similarity condition is the highest similarity, when the affine transformation matrix of the subgraph is calculated according to the subgraphs meeting the similarity condition obtained by corresponding matching of the subgraphs to be matched, the processing module is specifically configured to:
randomly selecting a plurality of preset subgraphs to be matched and subgraph pairs with the highest similarity obtained by corresponding matching, and respectively calculating to obtain candidate affine transformation matrixes;
respectively carrying out affine transformation on each subgraph to be matched according to any candidate affine transformation matrix aiming at each candidate affine transformation matrix obtained through calculation, determining the subgraph corresponding to each subgraph to be matched after the affine transformation, if the first distance between the subgraph after the affine transformation and the subgraph with the highest similarity obtained through corresponding matching is smaller than a first distance threshold value, determining that the matching condition is met between the corresponding subgraph to be matched and the subgraph with the highest similarity obtained through matching, and determining the sum of the number of the subgraphs to be matched meeting the matching condition and the sum of the corresponding first distances;
and taking the candidate affine transformation matrix corresponding to the maximum sum of the number of the sub-images to be matched meeting the matching conditions in the candidate affine transformation matrixes obtained by calculation and the minimum sum of the corresponding first distances as the affine transformation matrix.
8. The apparatus of claim 7, wherein when a sub-graph satisfying a matching condition is screened out according to the affine transformation matrix, the processing module is specifically configured to:
according to the affine transformation matrix, performing affine transformation on the sub-images to be matched respectively, and determining sub-images corresponding to the sub-images to be matched after the affine transformation;
and if the second distance between the sub-image after the affine transformation and the sub-image with the highest similarity obtained correspondingly is smaller than a preset second distance threshold, determining that the sub-image to be matched and the sub-image with the highest similarity obtained by matching meet the matching condition, and screening out the sub-image meeting the matching condition.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the steps of the method of any of claims 1-5 are implemented when the program is executed by the processor.
10. A computer-readable storage medium having stored thereon a computer program, characterized in that: the computer program when executed by a processor implementing the steps of the method of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010586734.8A CN111753719A (en) | 2020-06-24 | 2020-06-24 | Fingerprint identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010586734.8A CN111753719A (en) | 2020-06-24 | 2020-06-24 | Fingerprint identification method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111753719A true CN111753719A (en) | 2020-10-09 |
Family
ID=72678418
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010586734.8A Pending CN111753719A (en) | 2020-06-24 | 2020-06-24 | Fingerprint identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111753719A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112560893A (en) * | 2020-11-13 | 2021-03-26 | 贝壳技术有限公司 | Picture texture matching method and device, electronic medium and storage medium |
| CN113094657A (en) * | 2021-04-30 | 2021-07-09 | 汪知礼 | Software use limiting method and system based on fingerprint identification |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000311180A (en) * | 1999-03-11 | 2000-11-07 | Fuji Xerox Co Ltd | Method for feature set selection, method for generating video image class stastic model, method for classifying and segmenting video frame, method for determining similarity of video frame, computer-readable medium, and computer system |
| US20130094724A1 (en) * | 2011-10-18 | 2013-04-18 | King Fahd University Of Petroleum And Minerals | Method of performing fingerprint matching |
| CN103310453A (en) * | 2013-06-17 | 2013-09-18 | 北京理工大学 | Rapid image registration method based on sub-image corner features |
| CN105718920A (en) * | 2016-02-26 | 2016-06-29 | 深圳芯启航科技有限公司 | Fingerprint identification method and fingerprint identification device |
| CN105718852A (en) * | 2015-11-24 | 2016-06-29 | 深圳芯启航科技有限公司 | Fingerprint image processing method and apparatus |
| CN106919944A (en) * | 2017-01-20 | 2017-07-04 | 南京航空航天大学 | A kind of wide-angle image method for quickly identifying based on ORB algorithms |
| KR20170088721A (en) * | 2016-01-25 | 2017-08-02 | 조선대학교산학협력단 | Method for representing graph-based block-minutiae for fingerprint recognition and the fingerprint recognition system by using the same |
| CN107169479A (en) * | 2017-06-26 | 2017-09-15 | 西北工业大学 | Intelligent mobile equipment sensitive data means of defence based on fingerprint authentication |
| CN108596197A (en) * | 2018-05-15 | 2018-09-28 | 汉王科技股份有限公司 | A kind of seal matching process and device |
| US20190122024A1 (en) * | 2017-10-25 | 2019-04-25 | Synaptics Incorporated | Systems and methods for biometric recognition |
-
2020
- 2020-06-24 CN CN202010586734.8A patent/CN111753719A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2000311180A (en) * | 1999-03-11 | 2000-11-07 | Fuji Xerox Co Ltd | Method for feature set selection, method for generating video image class stastic model, method for classifying and segmenting video frame, method for determining similarity of video frame, computer-readable medium, and computer system |
| US20130094724A1 (en) * | 2011-10-18 | 2013-04-18 | King Fahd University Of Petroleum And Minerals | Method of performing fingerprint matching |
| CN103310453A (en) * | 2013-06-17 | 2013-09-18 | 北京理工大学 | Rapid image registration method based on sub-image corner features |
| CN105718852A (en) * | 2015-11-24 | 2016-06-29 | 深圳芯启航科技有限公司 | Fingerprint image processing method and apparatus |
| KR20170088721A (en) * | 2016-01-25 | 2017-08-02 | 조선대학교산학협력단 | Method for representing graph-based block-minutiae for fingerprint recognition and the fingerprint recognition system by using the same |
| CN105718920A (en) * | 2016-02-26 | 2016-06-29 | 深圳芯启航科技有限公司 | Fingerprint identification method and fingerprint identification device |
| CN106919944A (en) * | 2017-01-20 | 2017-07-04 | 南京航空航天大学 | A kind of wide-angle image method for quickly identifying based on ORB algorithms |
| CN107169479A (en) * | 2017-06-26 | 2017-09-15 | 西北工业大学 | Intelligent mobile equipment sensitive data means of defence based on fingerprint authentication |
| US20190122024A1 (en) * | 2017-10-25 | 2019-04-25 | Synaptics Incorporated | Systems and methods for biometric recognition |
| CN108596197A (en) * | 2018-05-15 | 2018-09-28 | 汉王科技股份有限公司 | A kind of seal matching process and device |
Non-Patent Citations (1)
| Title |
|---|
| 张志伟;: "自动指纹识别系统中关于指纹特征算法的研究", 吉林化工学院学报, no. 01 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112560893A (en) * | 2020-11-13 | 2021-03-26 | 贝壳技术有限公司 | Picture texture matching method and device, electronic medium and storage medium |
| CN113094657A (en) * | 2021-04-30 | 2021-07-09 | 汪知礼 | Software use limiting method and system based on fingerprint identification |
| CN113094657B (en) * | 2021-04-30 | 2024-08-09 | 深圳市群索科技有限公司 | Software use limiting method and system based on fingerprint identification |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wolfson | On curve matching | |
| US6658147B2 (en) | Reshaping freehand drawn lines and shapes in an electronic document | |
| Türetken et al. | Reconstructing curvilinear networks using path classifiers and integer programming | |
| CN109740606B (en) | Image identification method and device | |
| US9619733B2 (en) | Method for generating a hierarchical structured pattern based descriptor and method and device for recognizing object using the same | |
| US20070058856A1 (en) | Character recoginition in video data | |
| US9092697B2 (en) | Image recognition system and method for identifying similarities in different images | |
| EP2064652A1 (en) | Method of image processing | |
| CN110852311A (en) | Three-dimensional human hand key point positioning method and device | |
| CN111191584B (en) | Face recognition method and device | |
| JP2019102061A5 (en) | ||
| JP2019102061A (en) | Text line segmentation method | |
| US20200005078A1 (en) | Content aware forensic detection of image manipulations | |
| CN111753719A (en) | Fingerprint identification method and device | |
| JP2018142189A (en) | Program, distance measuring method, and distance measuring device | |
| CN111199558A (en) | Image matching method based on deep learning | |
| US8548259B2 (en) | Classifier combination for optical character recognition systems utilizing normalized weights and samples of characters | |
| CN113936288A (en) | Inclined text direction classification method and device, terminal equipment and readable storage medium | |
| CN110490157B (en) | Character evaluation method, character learning method, device, equipment and storage medium | |
| CN102375994A (en) | Method and device for detecting and reducing correctness of order of strokes of written Chinese character | |
| CN112287763A (en) | Image processing method, apparatus, device and medium | |
| CN109934858B (en) | Image registration method and device | |
| CN116311391A (en) | High-low precision mixed multidimensional feature fusion fingerprint retrieval method | |
| JP5414631B2 (en) | Character string search method, character string search device, and recording medium | |
| Nurhaida et al. | Determining the Number of Batik Motif Object based on Hierarchical Symmetry Detection Approach |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20201009 |
|
| WD01 | Invention patent application deemed withdrawn after publication |