CN118522041B

CN118522041B - Small area fingerprint image matching method, device, equipment and storage medium

Info

Publication number: CN118522041B
Application number: CN202410973607.1A
Authority: CN
Inventors: 谢勇辉; 秦进; 陈汉钦; 阎瑶; 李天保; 杨伟文
Original assignee: Shenzhen Magic Information Technology Co ltd
Current assignee: Shenzhen Magic Information Technology Co ltd
Priority date: 2024-07-19
Filing date: 2024-07-19
Publication date: 2024-11-12
Anticipated expiration: 2044-07-19
Also published as: CN118522041A

Abstract

The present application relates to the technical field of image processing, and discloses a small-area fingerprint image matching method, device, equipment and storage medium. The method comprises: obtaining at least two fingerprint images to be matched; preprocessing the fingerprint images to be matched to obtain a ridge map containing ridges; extracting ridge point features of the ridges to obtain ridge point features; extracting body features of the ridges to obtain ridge body features; performing ridge alignment operations on the two ridge maps based on the ridge point features and ridge body features to obtain an alignment area; performing ridge point matching operations on the two ridge maps based on the overall position deviation or local position deviation of the ridge points in the alignment area to obtain a ridge point matching result; obtaining a matching result of the two fingerprint images to be matched based on the ridge point matching result. The embodiments of the present application can improve the matching accuracy of small-area fingerprint images.

Description

Small-area fingerprint image matching method, device, equipment and storage medium

Technical Field

The application relates to the technical field of image processing amount, in particular to a small-area fingerprint image matching method, a device, equipment and a storage medium.

Background

The biological recognition technology is an important means for realizing the identification by comparing the special biological characteristics of people, wherein the fingerprint recognition technology is the mainstream recognition technology of the market.

In the related art, fingerprint identification technologies include an optical method, a capacitance method, an ultrasonic method, and the like, and the optical method is widely applied to imaging, sensing, storage, and other fields due to a high response speed. The optical fingerprint identification technology mainly comprises the steps of fingerprint acquisition, pretreatment (such as gray processing, filtering enhancement and the like) of acquired fingerprint patterns, feature point extraction of the pretreated fingerprints, and fingerprint identification by comparing the extracted feature points.

However, the small-area fingerprint image has insufficient fingerprint detail information due to the small area of the image, and influences the accuracy of fingerprint feature comparison, so that the accuracy of fingerprint identification is influenced.

Disclosure of Invention

The application aims to provide a small-area fingerprint image matching method, device, equipment and storage medium, aiming at improving the matching precision of small-area fingerprint images.

The embodiment of the application provides a small-area fingerprint image matching method, which comprises the following steps:

Acquiring at least two fingerprint images to be matched;

preprocessing the fingerprint image to be matched to obtain a ridge line graph containing ridge lines;

Extracting the ridge line point characteristics of the ridge line to obtain ridge line point characteristics;

Extracting the line body characteristics of the ridge line to obtain ridge line body characteristics;

performing ridge line alignment operation on the two ridge line graphs according to the ridge line point characteristics and the ridge line body characteristics to obtain an alignment region; the alignment area is an area where the two ridge line graphs are partially overlapped;

Performing ridge point matching operation on the two ridge line graphs according to the integral position deviation of the ridge line points in the alignment area or the local position deviation of the local position with the highest alignment coincidence degree of the ridge line points in the alignment area when the integral position deviation of the ridge line points in the alignment area exceeds a preset deviation range, so as to obtain a ridge line point matching result;

and obtaining the matching results of the two fingerprint images to be matched according to the ridge line point matching results.

In some embodiments, the ridge point feature includes a ridge point category feature and a ridge point association feature, and the ridge point feature extracting is performed on the ridge line to obtain a ridge point feature, including:

extracting ridge line points in the ridge line graph;

Classifying the ridge line points according to the neighborhood colors of the ridge line points and the positions of the ridge line points in the ridge line graph to obtain the ridge line point category characteristics;

tracking the ridge line according to the ridge line point category characteristics to obtain a tracking result, and associating the ridge line with the ridge line point according to the tracking result to obtain the ridge line point association characteristics.

In some embodiments, the ridge line body features include a line length feature, a curvature feature, a line width feature, and a line type feature, and the ridge line body feature extracting is performed on the ridge line to obtain a ridge line body feature, including:

performing Freeman chain code coding on the ridge line to obtain ridge line chain code representation;

Performing mapping accumulation operation on the ridge line chain code representation to obtain the line length characteristic;

calculating the curvature of the ridge line according to the relative position of the ridge line point on the ridge line to obtain the curvature characteristic;

Taking the sampling points on the ridge line as reference, and carrying out line width detection on the ridge line along the direction perpendicular to the ridge line to obtain the line width characteristics;

Classifying the ridge lines according to target ridge line point characteristics to obtain the linear characteristics; the target ridgeline point feature is a ridgeline point feature of a ridgeline point located at a start-end position of the ridgeline;

And when the linear characteristic represents that the ridge line is a target ridge line type, performing interval sampling on the ridge line, and generating a corresponding ridge line descriptor according to an interval sampling result.

In some embodiments, the ridge line body features include a line length feature, a curvature feature, a line width feature, and a line type feature, and performing ridge line alignment operation on the two ridge line graphs according to the ridge line point feature and the ridge line body feature to obtain an alignment region, including:

when the ridge lines accord with a first alignment condition, carrying out similarity matching and rigidity transformation matching on the ridge lines according to the line length characteristics, and determining a plurality of pairs of ridge lines with the highest matching degree to obtain a first matching result; the first alignment condition is that an initial ridgeline point and/or a final ridgeline point of the ridgeline are formed by non-line body interruptions or are formed by line body bifurcations;

When the ridge line accords with a second alignment condition, according to the curvature characteristic, inflection point positioning is carried out on the ridge line, line body superposition matching and rigidity transformation matching are carried out on the ridge line by utilizing the inflection point obtained by positioning, and a plurality of pairs of ridge lines with the highest matching degree are determined, so that a second matching result is obtained; the second alignment condition is that the maximum curvature of the ridge line is greater than a preset maximum curvature threshold;

when the ridge lines do not meet the first alignment condition and the second alignment condition, performing rigid transformation matching and transformation angle estimation on the ridge lines according to the curvature characteristics and the line width characteristics, rotating the ridge lines to the estimated transformation angle, performing line body superposition matching, and determining a plurality of pairs of ridge lines with the highest matching degree to obtain a third matching result;

And aligning the two ridge line graphs according to the first matching result, the second matching result or the third matching result to form the alignment area.

In some embodiments, the performing similarity matching and rigid transformation matching on the ridge lines according to the line length features, and determining several pairs of ridge lines with the highest matching degree, to obtain a first matching result, includes:

calculating the length deviation of the line body between the ridge lines according to the line length characteristics;

When the length deviation of the line body is smaller than a preset line length deviation threshold value, similarity matching is carried out on the ridge line descriptors of the ridge line, and descriptor similarity is obtained;

When the similarity of the descriptors is larger than a preset similarity threshold, performing rigid transformation matching on the ridge line, and solving a rotation matrix and a translation parameter of the ridge line by a least square method;

When the rotation matrix is smaller than a preset rotation threshold value and the translation parameter is smaller than a preset translation threshold value, selecting a plurality of pairs of ridge lines, closest to the preset rotation threshold value, of the rotation matrix and closest to the preset translation threshold value of the translation parameter, as the first matching result;

And/or according to the curvature characteristic, performing inflection point positioning on the ridge line, performing line body coincidence matching and rigid transformation matching on the ridge line by using the inflection point obtained by positioning, determining a plurality of pairs of ridge lines with highest matching degree, and obtaining a second matching result, wherein the method comprises the following steps:

determining the maximum curvature of the ridge line according to the curvature characteristics;

calculating the maximum curvature deviation between the ridge lines when the maximum curvature is larger than a preset maximum curvature threshold value;

When the maximum curvature deviation is smaller than a preset curvature deviation threshold, a chain code derivative integration method is adopted to locate the inflection point of the ridge line, and the inflection point is obtained through locating;

calculating the line weight matching degree of the ridge line according to the quantity of the inflection points and the relative included angles;

When the line body coincidence matching degree is smaller than a preset line body coincidence degree threshold value, carrying out rigid transformation matching on the ridge lines, and determining a plurality of pairs of ridge lines with the highest matching degree to obtain a second matching result;

And/or performing rigid transformation matching and transformation angle estimation on the ridge line according to the curvature characteristic and the line width characteristic, performing line body superposition matching after rotating the ridge line to the estimated transformation angle, and determining a plurality of pairs of ridge lines with the highest matching degree to obtain a third matching result, wherein the method comprises the following steps:

calculating the average curvature and the average linewidth of the ridge line according to the curvature characteristic and the linewidth characteristic;

When the average curvature is larger than a preset average curvature threshold value and the average line width is larger than a preset average line width threshold value, carrying out rigid transformation matching on the ridge line, solving a rotation matrix of the ridge line by a least square method, and calculating translation parameters of the ridge line by a distance field matching method;

And when the translation parameter is smaller than the preset translation threshold, selecting a plurality of pairs of ridge lines, the translation parameter of which is closest to the preset translation threshold, as the third matching result.

In some embodiments, the performing a ridge point matching operation on the two ridge maps according to the global position deviation of the ridge points in the alignment area or the local position deviation of the local position with the highest alignment coincidence degree of the ridge points in the alignment area when the global position deviation of the ridge points in the alignment area exceeds a preset deviation range, to obtain a ridge point matching result includes:

calculating the overall position deviation of the ridgeline points in the alignment area;

Judging whether the integral position deviation exceeds a preset deviation range or not;

if the ridge point matching result does not exceed the ridge point matching result, outputting a ridge point matching result passing through the ridge point matching;

If the local position deviation of the ridge line points exceeds the preset position deviation, dividing a reference area containing the ridge line points, and performing image blocking on one of the two ridge line graphs to obtain a plurality of ridge line blocks;

Recursion of each ridge line block by taking the reference area as a reference to obtain a target mapping transformation relation; the target mapping transformation relation is a rigid transformation relation which enables the mapping position deviation between the ridge line points in the reference area or the ridge line graph and the ridge line points in the mapped ridge line graph to be minimum after the ridge line points in the reference area or the ridge line graph are mapped to the other one of the two ridge line graphs;

And outputting a ridge point matching result passing through the ridge point matching when the mapping position deviation is smaller than a preset mapping deviation threshold value.

In some embodiments, recursively extracting each ridge tile based on the reference region to obtain a target mapping transformation relationship includes:

Constructing a map-bias problem; the map-offset problem describes a dependency between the map position offset and a map transformation relationship;

Calculating the rigid transformation relation of the ridge line points corresponding to the two ridge line graphs in the reference area to obtain a target mapping transformation relation of the reference area;

Performing block recursion operation by taking the reference area as an initial block; the block recursion operation is to solve a mapping transformation relationship when the mapping position deviation is minimized by using a target mapping relationship of an adjacent block as an initial solution of the mapping-deviation problem for the current ridge block according to a preset recursion direction, and the adjacent block is a ridge block adjacent to the current ridge block in the recursion direction or the reference region as the operation of the target mapping transformation relationship.

The embodiment of the application also provides a small-area fingerprint image matching device, which comprises:

the first module is used for acquiring at least two fingerprint images to be matched;

the second module is used for preprocessing the fingerprint image to be matched to obtain a ridge line graph containing ridge lines;

the third module is used for extracting the ridge line point characteristics of the ridge line to obtain ridge line point characteristics;

a fourth module, configured to extract a ridge line body feature from the ridge line, to obtain a ridge line body feature;

A fifth module, configured to perform a ridge line alignment operation on the two ridge line graphs according to the ridge line point feature and the ridge line body feature, to obtain an alignment area; the alignment area is an area where the two ridge line graphs are partially overlapped;

A sixth module, configured to perform a ridge point matching operation on two ridge maps according to an overall position deviation of the ridge points in the alignment area, or a local position deviation of a local position with a highest alignment overlapping degree of the ridge points in the alignment area when the overall position deviation of the ridge points in the alignment area exceeds a preset deviation range, so as to obtain a ridge point matching result;

And a seventh module, configured to obtain a matching result of the two fingerprint images to be matched according to the ridge line point matching result.

The embodiment of the application also provides electronic equipment, which comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the small-area fingerprint image matching method when executing the computer program.

The embodiment of the application also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program realizes the small-area fingerprint image matching method when being executed by a processor.

The application has the beneficial effects that: the method comprises the steps of performing ridge point feature extraction and line body feature extraction on a ridge line graph obtained by preprocessing a fingerprint image to be matched to obtain ridge point features and ridge line body features, performing ridge line alignment operation on the two ridge line graphs by simultaneously considering the ridge line point features and the ridge line body features to obtain an alignment area where the two ridge line graphs are partially overlapped, performing ridge point matching operation on the two ridge line graphs by simultaneously considering the integral position deviation of the ridge points in the alignment area and the local position deviation of the local position with the highest ridge point alignment overlapping degree in the alignment area to obtain a ridge point matching result, and obtaining a matching result of the two fingerprint images to be matched according to the ridge point matching result. Because the ridge line point characteristics and the ridge line body characteristics are extracted at the same time, characteristic elements required by fingerprint image matching are expanded, ridge line alignment operation and ridge line point matching operation are sequentially carried out, the ridge line alignment operation simultaneously considers the ridge line point characteristics and the ridge line body characteristics, the ridge line point matching operation simultaneously considers the integral position deviation of the ridge line points in the alignment area and the local position deviation of the local position with the highest alignment coincidence degree of the ridge line points in the alignment area, multiple matching and comparison can be carried out on the fingerprint images to be matched step by step, the defects of single fingerprint image matching mechanism and single type of matching characteristics are overcome, and the matching precision of the small-area fingerprint images is improved.

Drawings

Fig. 1 is a flowchart of a small-area fingerprint image matching method according to an embodiment of the present application.

Fig. 2 is a flowchart of a specific method of step S103 according to an embodiment of the present application.

Fig. 3 is a flowchart of a specific method of step S104 according to an embodiment of the present application.

Fig. 4 is a flowchart of a specific method of step S105 according to an embodiment of the present application.

Fig. 5 is a flowchart of a specific method of step S106 according to an embodiment of the present application.

Fig. 6 is a schematic structural diagram of a small-area fingerprint image matching device according to an embodiment of the present application.

Fig. 7 is a schematic diagram of a hardware structure of an electronic device according to an embodiment of the present application.

Fig. 8 is a schematic diagram of a matching effect of two ridge line graphs according to an embodiment of the present application.

Fig. 9 is a schematic diagram of a selection effect of a reference area according to an embodiment of the present application.

Fig. 10 is a schematic diagram illustrating the dividing effect of the ridge line block according to the embodiment of the present application.

Detailed Description

The present application will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present application more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

It should be noted that although functional block division is performed in the apparatus schematic and logical order is shown in the flowchart, in some cases, the steps shown may be performed in a different order than block division in the apparatus or in the flowchart. The terms first, second and the like in the description and in the claims and figures, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of the application only and is not intended to be limiting of the application.

Referring to fig. 1, fig. 1 is a flowchart of a small-area fingerprint image matching method according to an embodiment of the present application. In some embodiments of the present application, the method in fig. 1 may specifically include, but is not limited to, steps S101 to S107, and these seven steps are described in detail below in conjunction with fig. 1.

Step S101, at least two fingerprint images to be matched are obtained.

Step S102, preprocessing the fingerprint image to be matched to obtain a ridge line graph containing ridge lines.

And step S103, ridge line point feature extraction is carried out on the ridge line, and ridge line point features are obtained.

And step S104, extracting the line body characteristics of the ridge line to obtain the ridge line body characteristics.

And step S105, performing ridge line alignment operation on the two ridge line graphs according to the ridge line point characteristics and the ridge line body characteristics to obtain an alignment region.

The alignment area is an area where two ridge line graphs partially overlap.

And S106, performing ridge point matching operation on the two ridge line graphs according to the integral position deviation of the ridge line points in the alignment area or the local position deviation of the local position with the highest alignment coincidence degree of the ridge line points in the alignment area when the integral position deviation of the ridge line points in the alignment area exceeds a preset deviation range, so as to obtain a ridge point matching result.

And step S107, obtaining a matching result of the two fingerprint images to be matched according to the ridge line point matching result.

In step S101, a fingerprint may be acquired by a fingerprint acquisition device to obtain a fingerprint image to be matched, the acquired fingerprint image to be matched is stored in a fingerprint database, and at least two fingerprint images to be matched may be acquired by calling the fingerprint database and the currently acquired fingerprint image to be matched. For example, when a fingerprint is used for checking in and checking in, a plurality of fingerprint images are stored on the attendance machine, when a certain employee checks in, own fingerprint can be input on the attendance machine, the attendance machine collects the fingerprint to obtain a fingerprint image, the currently input fingerprint image and at least one prestored fingerprint image are obtained, and at least two fingerprint images to be matched are obtained. It can be understood that the fingerprint images to be matched contain fingerprint information obtained by collecting a fingerprint of one finger, and the fingerprint information in the two fingerprint images to be matched can be derived from the same finger or from different fingers.

In step S102, the fingerprint image to be matched is preprocessed, which may be that the fingerprint image to be matched is successively subjected to noise filtering, direction field estimation, ridge line enhancement, front background segmentation, frequency field and quality field estimation, gabor filtering enhancement, binarization and refinement, so as to obtain a ridge line map including ridge lines. It can be understood that the protruding lines visible on the surface of the finger are generally called ridges, the ridges are called valleys between the ridges, the ridges are one of the important features of the fingerprint image, the ridge map is a binary image, the fingerprint ridges have been converted into unit pixel widths, and the gray values of the pixel points on the ridge map are 0 or 1, and 1 represents the ridge points.

In step S103, ridge line point feature extraction is performed on ridge lines, which may be to traverse each ridge line point on the ridge line graph, perform classification marking on the ridge line points according to the positions on the ridge lines, and then divide each ridge line point according to the classification marking result, so that one ridge line point is divided into one ridge line, and obtain ridge line point features of the ridge line points. For example, the ridge line points may be classified into on-line points, end points, bifurcation points, and boundary points according to positions on the ridge line, wherein the number of neighborhood with a gray value of 1 is 1 and is not at the boundary of the ridge line, the number of neighborhood with a gray value of 1 is 1 and is at the boundary point of the boundary of the ridge line, the number of neighborhood with a gray value of 1 is 2 and is on-line points, the number of neighborhood with a gray value of 1 is greater than 2 and is bifurcation points, one ridge line may be divided between two end points, one end point and one boundary point, one end point and one bifurcation point, and one boundary point and one bifurcation point, and the ridge line points within the range are divided into the ridge line.

In step S104, the ridge line feature extraction may be performed on the ridge line by representing the ridge line by a ridge line point set composed of ridge line points, the ridge line point set including direction information and coordinate information of each ridge line point, and the ridge line feature extraction may be performed on various line features of the ridge line by using the direction information and coordinate information of each ridge line point included in the ridge line, thereby obtaining the ridge line feature. For example, the ridge line body feature may be a feature including a line length feature, a curvature feature, a line width feature, and a line shape feature, may be a feature of accumulating and summing coordinate information of each ridge line point to obtain a line length feature of the ridge line, may be a feature of calculating a line body curvature of coordinate information and direction information of adjacent several ridge line points to obtain a curvature feature of the ridge line, may be a feature of determining a tangential direction of the ridge line on the current ridge line point based on direction information of the ridge line point, and may be a feature of determining a line shape of the ridge line based on the ridge line point features of the first and second ridge line points by calculating a width of the ridge line on the tangential direction.

In step S105, the alignment operation of the two ridge lines is performed on the ridge lines according to the rigid transformation method, after traversing the ridge lines in the two ridge lines, the priority ranking is performed on the ridge lines according to the ridge line point feature and the ridge line body feature, the ridge lines with high feature stability and high distinguishing power are set to have higher priority, the alignment operation is performed on the ridge line with the highest current priority preferentially, the alignment operation is performed on the ridge lines sequentially according to the priority order, if the preset number of alignment ridge lines meet the corresponding alignment condition during the alignment operation, the alignment operation is ended, otherwise, the alignment operation is continued until the preset number of alignment ridge lines meet the corresponding alignment condition, and the local overlapping areas of the two ridge lines after alignment of the ridge lines meeting the alignment condition are marked with areas, so as to obtain the alignment area. And if the preset number of pairs of ridge lines do not meet the corresponding alignment condition after the alignment of all the ridge lines is finished, the ridge lines of the two ridge lines are not matched, and a matching result of the mismatch of the fingerprint images to be matched is output. For example, it may be that the ridge lines whose type of the set end ridge line point is the end point or the bifurcation point have the highest priority, the ridge lines whose maximum curvature other than the set highest priority is larger than the preset maximum curvature threshold have the next highest priority, the ridge lines whose next highest priority is set to have the lowest priority, if the ridge lines having the highest priority are both present, the ridge lines whose highest priority are preferentially aligned, the alignment is ended and it is determined that the ridge lines of the two ridge lines are alignable, if there is a preset number of the ridge lines satisfying the corresponding alignment condition, the ridge lines whose highest priority are not aligned, if there is no ridge line having the highest priority or the ridge line having the highest priority is not aligned, the alignment is preferentially performed on the ridge lines having the next highest priority, if there is a preset number of the ridge lines satisfying the corresponding alignment condition, the ridge lines whose two ridge lines whose lowest priority are not aligned are ended and it is determined that the ridge lines whose highest priority are alignable, if there is only the ridge lines having the lowest priority or the ridge lines whose highest priority are not aligned, the ridge lines whose corresponding to be the preset number of the ridge lines whose highest priority are not aligned, the matching condition is output, and if there is a preset number of the ridge lines whose corresponding alignment condition is not satisfied.

In step S106, the ridge point matching operation is performed on the two ridge line graphs, which may be performed by considering the global position deviation of the ridge line points in the alignment area and the local position deviation of the local position with the highest alignment coincidence degree in the alignment area at the same time, so as to perform the ridge point matching operation on the two ridge line graphs, when the global position deviation does not exceed the preset deviation range, the ridge point matching result passing through the ridge point matching of the two ridge line graphs is obtained, when the global position deviation exceeds the preset deviation range, the ridge point matching operation is performed on the two ridge line graphs according to the local position deviation of the local position with the highest alignment coincidence degree in the alignment area, if the local position deviation of the local position with the highest alignment coincidence degree in the alignment area exceeds the preset deviation range, the ridge point matching result is obtained, if the local position deviation of the local position with the highest alignment coincidence degree in the alignment area does not exceed the preset deviation range, the ridge line in the two ridge line graphs is transformed according to the global position deviation of the corresponding ridge line in the two ridge line graphs, and whether the ridge line in the two ridge line graphs are matched according to the global position deviation results is determined.

In step S107, when a ridge point matching result that the ridge points of the two ridge line patterns pass is obtained, a matching result that the two fingerprint images to be matched match is obtained, and when a ridge point matching result that the ridge points of the two ridge line patterns do not pass is obtained, a matching result that the two fingerprint images to be matched do not match is obtained.

In steps S101 to S107 of the embodiment of the present application, ridge point feature extraction and line feature extraction are performed on a ridge line graph obtained by preprocessing a fingerprint image to be matched, so as to obtain a ridge point feature and a ridge line feature, ridge line alignment operation is performed on the two ridge line graphs while considering the ridge point feature and the ridge line feature, so as to obtain an alignment region where the two ridge line graphs partially overlap, and then ridge point matching operation is performed on the two ridge line graphs while considering the integral position deviation of the ridge point in the alignment region and the local position deviation of the local position where the ridge point alignment overlap degree is highest in the alignment region, so as to obtain a ridge point matching result, and according to the ridge point matching result, a matching result of the two fingerprint images to be matched is obtained. Because the ridge line point characteristics and the ridge line body characteristics are extracted at the same time, characteristic elements required by fingerprint image matching are expanded, ridge line alignment operation and ridge line point matching operation are sequentially carried out, the ridge line alignment operation simultaneously considers the ridge line point characteristics and the ridge line body characteristics, the ridge line point matching operation simultaneously considers the integral position deviation of the ridge line points in the alignment area and the local position deviation of the local position with the highest alignment coincidence degree of the ridge line points in the alignment area, multiple matching and comparison can be carried out on the fingerprint images to be matched step by step, the defects of single fingerprint image matching mechanism and single type of matching characteristics are overcome, and the matching precision of the small-area fingerprint images is improved.

Referring to fig. 2, in a specific embodiment, step S103 may specifically include, but is not limited to, steps S201 to S203, which are described in detail below in conjunction with fig. 2.

Step S201, ridge line points in the ridge line map are extracted.

Step S202, classifying the ridge points according to the neighborhood colors of the ridge points and the positions of the ridge points in the ridge map, and obtaining the ridge point category characteristics.

And step S203, tracking the ridge line according to the ridge line point category characteristics to obtain a tracking result, and associating the ridge line with the ridge line point according to the tracking result to obtain a ridge line point association characteristic.

In step S201, ridge line points in the ridge line graph are extracted, that is, the ridge line points may be acquired at intervals along the length direction of the ridge line according to the gray values of the pixel points in the ridge line graph, and the ridge line points are acquired once between intervals along the length direction of the ridge line, so as to obtain a plurality of required ridge line points.

In step S202, the ridge points are classified according to their neighborhood colors and positions in the ridge map, which may be that eight neighborhood algorithms are used to determine gray values of eight neighborhoods of the ridge points, and the ridge points are classified according to the number of gray values of 1 in the eight neighborhoods of the ridge points, so as to obtain the ridge point class characteristics. Specifically, the ridge line point category feature includes an on-line point, an end point, a bifurcation point and a boundary point, wherein the number of neighborhoods with a gray value of 1 is 1 and is not at the boundary of the ridge line graph, the number of neighborhoods with a gray value of 1 is 1 and is at the boundary point of the boundary of the ridge line graph, the number of neighborhoods with a gray value of 1 is 2 and is the on-line point, and the number of neighborhoods with a gray value of 1 is greater than 2 and is the bifurcation point.

In step S203, each ridge line in the ridge line graph may be extracted by using a ridge line tracking algorithm, ridge line tracking may be performed according to the ridge line point category characteristics of each ridge line point, and the ridge line obtained by dividing and the ridge line point on the ridge line may be associated to obtain a ridge line point association characteristic. Specifically, the ridge line tracking is to randomly start tracking from a ridge line point which is not tracked and has the ridge line point category characteristic of an endpoint or a boundary point in a ridge line diagram, randomly start the ridge line tracking from a bifurcation point if the ridge line point which is not tracked is not provided with the endpoint or the boundary point, identify adjacent ridge line points of the current tracking point to identify the number of the untracked ridge line points in the eight neighbors of the current tracking point, mark the current tracking point as the tracked point if the number of the untracked ridge line points is 1, move the tracking position to the adjacent ridge line point, mark the current tracking point as the tracked point if the number of the untracked ridge line points is more than 1, move the tracking position to one of the untracked adjacent ridge line points of the current tracking point if the number of the untracked ridge line points is 0, finish the current tracking, determine the ridge line points contained in the current tracking line, start the next ridge line tracking until all the ridge lines are tracked until the next ridge line points exist.

Referring to fig. 3, in a specific embodiment, the ridge line body features include a line length feature, a curvature feature, a line width feature, and a line type feature, and step S104 may specifically include, but is not limited to, steps S301 to S306, which are described in detail below in conjunction with fig. 3.

Step S301, freeman chain code encoding is carried out on the ridge line, and ridge line chain code representation is obtained.

Step S302, mapping and accumulating operation is carried out on the ridge line chain code representation, and line length characteristics are obtained.

Step S303, calculating the curvature of the ridge line according to the relative position of the ridge line point on the ridge line, and obtaining the curvature characteristic.

And step S304, detecting the line width of the ridge line along the direction perpendicular to the ridge line by taking the sampling point on the ridge line as a reference, and obtaining the line width characteristic.

Step S305, classifying the ridge lines according to the target ridge line point characteristics to obtain linear characteristics.

The target ridgeline point feature is a ridgeline point feature of a ridgeline point located at a start and end positions of the ridgeline.

And step S306, when the linear characteristic representing ridge line is the target ridge line type, performing interval sampling on the ridge line, and generating a corresponding ridge line descriptor according to an interval sampling result.

In step S301, the Freeman chain code encoding is performed on the ridge lines, that is, after a plurality of ridge lines are tracked and extracted, the ridge line points of each ridge line are encoded by using the eight-direction Freeman chain code, the ridge line points are ordered according to the positions of the ridge line points on the ridge lines, and the position coordinates of the initial ridge line point of the ridge line and the direction information of each ridge line point are recorded, so that the ridge line chain code representation is obtained. For example, S may be used to represent the position coordinates of the starting ridgeline point, and for the ith ridgeline point on the ridgeline, S may be usedTo represent the ridge point encoded representation of the ridge pointCharacterizing the direction in which the i-th ridgeline point points to the i + 1-th ridgeline point,The ridge chain code representation for a particular ridge can be expressed asN is the number of ridgeline points in the ridgeline.

In step S302, the calculation formula for performing the mapping accumulation operation on the ridge chain code representation is:

，

Wherein, In the case of a line length feature,And (5) the chain code mapping result of the ith ridge line point.

In step S303, the curvature of the ridge line may be calculated by traversing the ridge line points at the non-start position and the non-end position on the ridge line, and calculating the curvature by combining the two adjacent or separated ridge line points of a certain length and the position coordinates of the ridge line points, thereby obtaining the curvature characteristic. Specifically, for the ith ridge line point in the ridge lineIn other words, the ridge point and two ridge points spaced from the ridge point by m (m is a positive integer, for example, 5) are combinedAndIs subjected to curvature calculation, wherein,Is the position coordinates of (a)，Is the position coordinates of (a)，Is the position coordinates of (a)The calculation formula for calculating the curvature of the ridge line is:

，

Wherein, Is that、AndThe area of the triangle formed by the two connecting lines,Is the curvature characteristic of the ridge line at the i-th ridge line point position.

In step S304, the line width of the ridge line may be detected by obtaining the ridge line direction of the ridge line through a direction field, in the ridge line diagram, the ridge line point is verified along the vertical direction of the ridge line direction, when the non-ridge line pixel point is encountered, the obtained line segment length is the ridge line width of the point, the local line width characteristic of the ridge line is obtained, and after traversing each ridge line point on the ridge line or a plurality of preselected ridge line points serving as samples, the integral line width characteristic of the ridge line is obtained.

In step S305, the ridge lines are classified according to the target ridge line point feature, and may be classified according to the ridge line point features of the first and last two ridge line points to determine the line feature of the ridge line. Specifically, when the ridge line point feature of the ridge line point located at the start and end positions of the ridge line is the target ridge line point feature, the line feature of the ridge line is the target ridge line feature, and when the ridge line point feature of the ridge line point located at the start and end positions of the ridge line is the non-target ridge line point feature, the line feature of the ridge line is the non-target ridge line feature. For example, when the end point and the bifurcation point are the target ridge point feature, the start ridge point or the end ridge point of the ridge is the end point or bifurcation point, the linear feature of the ridge is the target ridge feature, whereas the linear feature of the ridge is the non-target ridge feature.

In step S306, the ridge lines are sampled at intervals, corresponding ridge line descriptors are generated according to the sampling results at intervals, the ridge lines may be sampled and marked from the starting ridge line point of the ridge lines at certain intervals, a circular area is constructed by taking each sampling point as the center of a circle and taking a preset parameter as the radius, the direction obtained by the sampling point in the direction field is taken as the horizontal direction of the area, the sampling point descriptors are obtained by performing specific calculation and statistics on the gray values in the circular area, the circular area constructed by all the sampling points is called as the supporting area of the ridge line descriptors, and then the ridge line descriptors are generated by performing histogram statistics on the gradient in the circular area constructed by each sampling point. Because of the selection rule in the horizontal direction of the region and the characteristics of the fingerprint ridge, the gradient histogram statistics of this embodiment is performed in the vicinity of the horizontal direction, with the gradient histogram statistics of this embodiment having 8 columns in total, with the gradient angle range of each column statistics being [0 °,30 ° ], [ 60 ° ], [120 °,150 ° ], [150 °,180 ° ], [180 °,210 ° ], [210 °,240 ° ], [300 ° ], 330 ° ] and [330 ° ], with each sampling point statistics obtaining an 8-dimensional statisticFor each of the dimensional componentsCalculate its sample mean valueThen, binary quantization is carried out on each dimension through the mean value to obtain a sampling point descriptorWhereinThe method comprises the following steps:

，

Therefore, each sampling point can obtain an 8-bit binary sampling point descriptor, and the sampling point descriptors are connected in series to obtain the ridge line descriptor of the ridge line.

Referring to fig. 4, in a specific embodiment, the ridge line body features include a line length feature, a curvature feature, a line width feature, and a line type feature, and step S105 may specifically include, but is not limited to, steps S401 to S404, which are described in detail below in connection with fig. 4.

And S401, when the ridge lines meet the first alignment condition, performing similarity matching and rigidity transformation matching on the ridge lines according to the line length characteristics, and determining a plurality of pairs of ridge lines with the highest matching degree to obtain a first matching result.

And step S402, when the ridge line meets a second alignment condition, positioning an inflection point of the ridge line according to the curvature characteristic, performing line body superposition matching and rigidity transformation matching on the ridge line by utilizing the inflection point obtained by positioning, and determining a plurality of pairs of ridge lines with the highest matching degree to obtain a second matching result.

And S403, when the ridge lines do not meet the first alignment condition and the second alignment condition, carrying out rigid transformation matching and transformation angle estimation on the ridge lines according to the curvature characteristic and the line width characteristic, rotating the ridge lines to the transformation angle obtained by estimation, carrying out line body superposition matching, and determining a plurality of pairs of ridge lines with the highest matching degree to obtain a third matching result.

Step S404, aligning the two ridge line graphs according to the first matching result, the second matching result or the third matching result to form an aligned area.

The first alignment condition is that an initial ridge line point and/or a final ridge line point of the ridge line are formed by non-line body interruption or line body bifurcation, and the second alignment condition is that the maximum curvature of the ridge line is larger than a preset maximum curvature threshold value.

In step S401, for a ridge line that meets the first alignment condition, an initial ridge line point and/or a termination ridge line point is formed by a non-line body break or formed by a line body bifurcation, that is, the initial ridge line point and/or the termination ridge line point of the ridge line are used as end points or bifurcation points, when the ridge line alignment operation is performed on two such ridge lines, according to the line length characteristics of the two ridge lines, the line length deviation between the two ridge lines is calculated, when the line length deviation meets the preset line length deviation condition, similarity matching is performed on the ridge line descriptors of the two ridge lines, when the result of similarity matching of the ridge line descriptors meets the preset descriptor similarity condition, rigid transformation matching is performed on the two ridge lines, and rotation matrix and translation parameters required by the alignment of the two ridge lines are solved, and after traversing multiple pairs of ridge lines, a plurality of optimal groups of rotation matrix and translation parameters are selected from the rotation matrix and translation parameters, so as to obtain the first matching result.

In step S401 of some embodiments, the method may include the steps of:

calculating the length deviation of the line body between ridge lines according to the line length characteristics;

when the similarity of the descriptors is larger than a preset similarity threshold, performing rigid transformation matching on the ridge line, and solving a rotation matrix and translation parameters of the ridge line by a least square method;

and when the rotation matrix is smaller than the preset rotation threshold and the translation parameter is smaller than the preset translation threshold, selecting a plurality of pairs of ridge lines of which the rotation matrix is closest to the preset rotation threshold and the translation parameter is closest to the preset translation threshold as a first matching result.

Specifically, for the ridge lines meeting the first alignment condition, when the deviation of the line body lengths of the two ridge lines is smaller than a preset line length deviation threshold, the similarity of the ridge lines can be measured by using the descriptor similarity of the ridge line descriptors to judge whether the two ridge lines are matched, and when the descriptor similarity is larger than the preset similarity threshold, the two ridge lines are primarily matched. For example, two ridgelinesAndThe calculation formula of the descriptor similarity of (2) is as follows:

，

Wherein, In order to describe the degree of similarity of the sub-sets,Is a ridge lineIs described as a ridge line description of (c),Is a ridge lineIs described as a ridge line description of (c),Is a ridge lineIs used for the length of the line,Is a ridge lineIs used for the length of the line,As an exclusive-or operator,Is a ridge lineIs the ridgeline point descriptor of the ith ridgeline point,Is a ridge lineIs the ridgeline point descriptor of the ith ridgeline point.

In some embodiments, the ridge points at the start and end positions of the ridge are all endpoints and/or bifurcation points, in some scenes needing 360-degree identification, ridge descriptors in two directions of the ridge need to be calculated, namely, a ridge descriptor obtained by sampling from an initial ridge point and a ridge descriptor obtained by sampling from a termination ridge point in the opposite direction, and in calculating the ridge similarity, the ridge similarity needs to be calculated twice.

When the similarity of the descriptors is larger than a preset similarity threshold, constructing and calculating a rigid transformation relation of the two ridge lines, wherein the expression of the rigid transformation relation is as follows:

，

Wherein, AndAnd respectively acquiring two ridge line point sets of two ridge lines, wherein R is a rotation matrix in rigid transformation and T is a translation parameter in rigid transformation.

And combining a calculation formula of the descriptor similarity and an expression of the rigid transformation relation to obtain an overdetermined equation set of rotation matrix and translation parameters, and solving a least square solution of the equation set to obtain the rigid transformation relation of the pair of matched ridge lines, namely the pair of ridge lines are matched. And if the two ridge line graphs are only matched with ridge lines which are not more than the preset number of pairs and meet the first alignment condition, taking the calculated rotation matrix and the translation parameter as a first matching result. If the number of matched ridge lines is not greater than the preset number, checking whether the rigidity transformation relation obtained by matching each pair of ridge lines is compatible, judging whether the rigidity transformation relation is compatible by calculating the deviation of a rotation matrix and translation parameters of the rigidity transformation relation, if the deviation of the rotation matrix and the translation parameters exceeds a certain preset threshold, judging that the rigidity transformation relation is incompatible, otherwise, the rigidity transformation relation with the most compatible matched ridge lines is found, and taking the corresponding rotation matrix and translation parameters as a first matching result. If no ridge line conforming to the first alignment condition is matched, checking the two ridge line graphs, judging whether the two ridge line graphs contain a preset number of ridge lines conforming to the first alignment condition and whether the two ridge line graphs contain ridge lines conforming to the first alignment condition in a preset image area, for example, judging whether the two ridge line graphs contain at least four ridge lines conforming to the first alignment condition and each quadrant area of a four-quadrant coordinate system constructed by the center of the ridge line graphs contains at least one ridge line conforming to the first alignment condition, and if at least one ridge line graph is yes, obtaining a matching result of mismatch of the fingerprint images to be matched.

In step S402, for the ridge line meeting the second alignment condition, the ridge line is matched by using the maximum curvature of the ridge line, the ridge line is approximately matched by calculating all inflection points of the ridge line, the line body overlapping matching is performed on the ridge line, so as to quickly calculate the overlapping portion of the ridge line, after the overlapping portion of the ridge line is matched, the corresponding inflection point correspondence is converted into a rotation matrix and translation parameters required for rigidly transforming and matching the ridge line, and after traversing multiple pairs of ridge lines, the optimal rotation matrix and translation parameters are selected from the rotation matrix and translation parameters, so as to obtain a second matching result.

In step S402 of some embodiments, the method may include the steps of:

Calculating the maximum curvature deviation between ridge lines when the maximum curvature is larger than a preset maximum curvature threshold value;

When the maximum curvature deviation is smaller than a preset curvature deviation threshold value, a chain code derivative integration method is adopted to locate inflection points of the ridge line, and the inflection points are located and obtained;

according to the quantity of inflection points and the relative included angle, calculating the line weight matching degree of the ridge line;

And when the line weight matching degree is smaller than a preset line weight matching degree threshold, carrying out rigid transformation matching on the ridge lines, and determining a plurality of pairs of ridge lines with the highest matching degree to obtain a second matching result.

Specifically, for two ridge lines meeting the second alignment conditionAndRidge lineAndMaximum curvatures are respectivelyAndOnly when the maximum curvature deviatesWhen the curvature is smaller than a certain preset maximum curvature threshold value, the ridge line is alignedAndMatching is performed. Judging the ridge lineAndIf there is a match, the inflection point on the ridge is first located by the integral method of the chain code derivative, where the chain code derivativeThe definition is as follows:

，

Wherein, In the form of a differential code,As a derivative of the chain code,AndAre all the ridge line point codes,Representing a modulo operation. Calculating the integral of the derivative of the chain code starting from the second ridgeline point of the ridgeline, from the second ridgeline point to the firstChain code derivative integration of individual ridgeline pointsThe method comprises the following steps:

，

When (when) When it is, then determine the firstThe ridge line points are inflection points, and when the inflection points are extracted, the initial position of the chain code derivative cumulative summation becomes the inflection point position plus 1.

Assume that、For the position of the first two inflection points of the ridge, then:

，

After all inflection points of the ridge line conforming to the second alignment condition are calculated, ridge line matching can be approximated by inflection point matching, assuming that 、、、……、For all inflection points sequentially found on a ridge line, a sequence of inflection point lengths of the ridge line is definedAnd inflection point included angle sequenceThe following are provided:

，

Wherein, For the i-th inflection interval length,Is the relative angle of the ith inflection point. Thereby, the ridge line can be calculatedAndIs provided with the overlapped sections ofTo the extent of the sequence of overlapping segments that need to be checked,、And expressing the respective initial positions, and calculating the difference degree of the overlapped sections according to the following formula:

，

Wherein T is the line weight coincidence matching degree of the coincident sections, As a parameter of the weight-bearing element,Is the relative angle of the ith inflection point of the ridge line a,Is the relative angle of the ith inflection point of the ridge line b,For the ith inflection interval length of the ridge line a,Is the ith inflection interval length of the ridge line b. The preset line weight coincidence degree threshold is T _th, and when T is less than T _th, the ridge line is determinedAndOn the contrary, is determined as a ridge lineAndIs not matched. If the ridge lineAndOn the coincidence section matching of the curve, namely corresponding inflection points form a corresponding relation, the curve is converted into a least square solution problem of an overdetermined equation set for solving a rotation matrix and translation parameters of the curve by carrying out rigid transformation matching on the curve, thereby solving the curveAndAnd according to the rigid transformation relation between the two ridge lines, according to rotation matrixes and translation parameters required by solving alignment of the two ridge lines, traversing a plurality of pairs of ridge lines, and selecting a plurality of optimal groups of rotation matrixes and translation parameters from the rotation matrixes and translation parameters to obtain a second matching result. If no ridge line meeting the second alignment condition is matched, checking the two ridge line graphs, judging whether the two ridge line graphs contain a preset number of ridge lines meeting the second alignment condition, for example, judging whether the number proportion of the ridge lines meeting the second alignment condition in the two ridge line graphs exceeds 70%, if at least one ridge line graph is yes, the ridge lines of the two ridge line graphs are not matched, and a matching result of the mismatch of the fingerprint images to be matched can be obtained.

In step S403, for the ridge lines that do not meet the first alignment condition and the second alignment condition, according to the curvature characteristic and the line width characteristic, the ridge line with a larger difference between the curvature characteristic and the line width characteristic is filtered, then the ridge line is subjected to rigid transformation matching and transformation angle estimation, the angle for performing the required rigid transformation matching is estimated first, then the ridge line is rotated to the estimated transformation angle after the corresponding rotation angle is estimated, the translation parameter to be checked is calculated by an enumeration method, and the line body overlapping matching is performed on the ridge line by using the calculated translation parameter, so as to determine a plurality of pairs of ridge lines with the highest matching degree, and a third matching result is obtained.

In step S403 of some embodiments, the method may include the steps of:

Calculating the average curvature and the average line width of the ridge line according to the curvature characteristics and the line width characteristics;

When the average curvature is larger than a preset average curvature threshold value and the average line width is larger than a preset average line width threshold value, carrying out rigid transformation matching on the ridge line, solving a rotation matrix of the ridge line by a least square method, and calculating translation parameters of the ridge line by adopting a distance field matching method;

And when the translation parameter is smaller than the preset translation threshold, selecting a plurality of pairs of ridge lines with the translation parameter closest to the preset translation threshold as a third matching result.

According to the curvature characteristic and the line width characteristic, calculating the average curvature and the average line width of the ridge lines, filtering the matching between the obviously dissimilar ridge lines by utilizing the calculated average curvature and average line width, and then preferentially estimating the angle of the rigidity transformation, wherein each ridge line can be basically seen as a straight line, the matching of each possible matched ridge line is calculated, and the rotation angle between the two small-area fingerprint images which are matched currently can be solved by counting the rotation angle histogram. After the rotation angle is estimated, considering the condition limit that the overlapping area needs to reach the preset threshold value, calculating the translation parameter to be checked by adopting an enumeration method, and filtering out a large amount of parameter space which does not need to be searched by adopting a ridge line pre-pairing method, as shown in fig. 8, in a specific embodiment, after the rotation angle is estimated, rotating both ridge line patterns to the state that the ridge lines are in the horizontal state as a whole, so that the corresponding number of ridge lines in the two ridge line patterns are mutually matched, wherein the ridge line in the middle position in one ridge line pattern can only be matched with the ridge line in the middle position in the other ridge line pattern, for example, the 1 st ridge line in the ridge line pattern on the right lower side can only be matched with the 1 st to 11 st ridge lines in the ridge line pattern on the left upper side, otherwise, the true and false can not be identified due to the fact that the information quantity is too small, thereby determining the value of the translation parameter in the vertical direction. After the rotation angle of rigid deformation and the vertical displacement after the alignment are determined, the displacement in the horizontal direction is quickly solved by matching based on distance fields, firstly, the distance field of one ridge line diagram is calculated by using chamfering distance conversion, then, for each horizontal displacement to be verified, the integral distance deviation obtained in the distance field of the ridge line diagram after the action of the other ridge line diagram is calculated, and the horizontal displacement corresponding to the minimum distance deviation is taken as the optimal solution. And if the minimum distance deviation is larger than a preset translation threshold value, obtaining a matching result of unmatched fingerprint images to be matched.

In step S404, when the first matching result, the second matching result, or the third matching result is obtained, the two ridge line graphs are aligned, so that the ridge lines in the two ridge line graphs that match the first matching result, the second matching result, or the third matching result are aligned, and an alignment area is formed.

Referring to fig. 5, in a specific embodiment, step S106 may specifically include, but is not limited to, steps S501 to S506, and these six steps are described in detail below in conjunction with fig. 5.

In step S501, the overall positional deviation of the ridge line points in the alignment area is calculated.

Step S502, judging whether the overall position deviation exceeds a preset deviation range. If not, executing step S503; if yes, go to step S504.

Step S503, outputting ridge point matching results passing ridge point matching.

In step S504, when the local position deviation of the ridge line point is smaller than the preset position deviation, a reference area including the ridge line point is divided, and one of the two ridge line graphs is subjected to image segmentation to obtain a plurality of ridge line blocks.

In step S505, each ridge line block is recursively mapped with the reference region as a reference to obtain a target mapping transformation relationship.

The target mapping transformation relation is a rigid transformation relation which minimizes the deviation of the mapping position between the ridge line points in the ridge line graph and the reference area or the ridge line points in the ridge line graph after mapping the ridge line points to the other one of the two ridge line graphs.

And S506, outputting a ridge point matching result passing through ridge point matching when the mapping position deviation is smaller than a preset mapping deviation threshold value.

In steps S501 to S506, a distance field measurement method is adopted to calculate the overall position deviation of the ridge line points in the alignment area, when the overall position deviation of the ridge line points in the alignment area exceeds a preset deviation range, a ridge line point matching result that the ridge line points do not pass is output, on the basis that the overall position deviation of the ridge line points in the alignment area does not exceed the preset deviation range, a reference area (the area in the square frame shown in fig. 9 is the reference area) in which the local position deviation of the ridge line points is smaller than the preset position deviation is defined, the local position deviation of the ridge line points in the reference area is smaller than the local position deviation of the ridge line points outside the reference area, image blocking is performed on one of the two ridge line graphs by taking the reference area as a reference, a plurality of ridge line graphs shown in fig. 10 are defined, then, on the basis that the rigid transformation relation between the two ridge line graphs in the reference area is not exceeded, the ridge line points in the two ridge line graphs are mapped to the target transformation relation of the other one of the two ridge line graphs, the ridge line graphs are mapped to the ridge line graphs, the ridge line graphs in the map is mapped to the other ridge line graph, when the ridge line graphs in the target transformation relation is mapped to the other ridge line graphs, the ridge line graphs are mapped to the ridge line graphs, the ridge line graphs in the ridge line graphs are mapped to the ridge line graphs, and the ridge line graph map is not passes, and the ridge line graph is output, and the ridge line graph is obtained, and the ridge line graph is obtained.

In step S505 of some embodiments, the method may include the steps of:

Constructing a map-bias problem;

the map-offset problem describes the dependency between the map position offset and the map transformation relationship;

calculating the rigid transformation relation of the ridge line points corresponding to the two ridge line graphs in the reference area to obtain the target mapping transformation relation of the reference area;

taking the reference area as an initial block, and executing block recursion operation;

The block recursion operation is that according to a preset recursion direction, a mapping transformation relation when the mapping position deviation is minimized is solved by using a target mapping relation of adjacent blocks as an initial solution of a mapping-deviation problem of the current ridge block, and the adjacent blocks are ridge blocks or reference areas adjacent to the current ridge block in the recursion direction as an operation of the target mapping transformation relation.

For example, for ridge maps a and B, ridge map B is partitioned into a plurality of ridge map blocks, and each ridge point in ridge map B only belongs to the ridge map blocks included therein.

Definition of the first embodimentEach ridge line pattern blockThe ridge line point set is as followsEach ridge line block has a transformation relation for mapping the ridge line point set in the ridge line block to the coordinate system of the ridge line diagram A, and the first is setEach ridge line pattern blockThe transformation relation of (2) isFor the firstEach ridge line pattern blockAny one of the ridge pointsThe transformation relation of the points mapped to the coordinate system of the ridge line graph A is that。

For pixel points mapped to the coordinate system of the ridge map AThe error defining the pixel is:

，

Wherein, Is the set of ridge points of ridge map a, v is the pixel point after the ridge point is mapped from ridge map B to ridge map a,Representing pixel pointsAndIs a euclidean distance of (c).

For the first line graph BEach ridge line pattern blockDefining an errorThe method comprises the following steps:

，

When (when) The smaller the representation of the firstEach ridge line pattern blockThe higher the degree of matching.

For the firstEach ridge line pattern blockConstruction map-bias problemWherein the map-offset problem describes a subordinate relation between the map position offset and the map transformation relation, and is a process of obtaining the map transformation relation of the ridge line point in the case that the map position offset is minimum. Map-bias problemThe method comprises the following steps:

Min: ，

Subject to:

，

Wherein, Is a ridge line pointThe coordinate representation under the coordinate system of the ridge graph B,Is a rigid transformationThe mapping-deviation problem is solved at a given initial value) Under the condition of (1), a more accurate solution is obtained by iterating N times through the steepest descent method)。

For two adjacent (left-right or up-down adjacent) ridge line tiles in ridge line map BAndDefinition of the slaveTo the point ofRecursively takeIs a transformation relation of (a)As a means ofIn (a)Initial solution of problem, solving mapping-bias problemThe optimal solution obtainedAnd finally, obtaining the target mapping transformation relation of all the ridge line blocks as the target mapping transformation relation.

In a specific embodiment, the tile recursion operation is performed with the reference area as an initial block, and may be performed with the reference area as an initial block until right recursion is disabled, with the reference area as an initial block, left recursion until left recursion is disabled, with each ridge tile of the row in which the reference area is located as an initial block, and up recursion until up recursion is disabled, and with each ridge tile of the row in which the reference area is located as an initial block, and down recursion is disabled.

Referring to fig. 6, an embodiment of the present application further provides a small-area fingerprint image matching device, which can implement the small-area fingerprint image matching method, where the device includes:

A first module 601, configured to obtain at least two fingerprint images to be matched;

a second module 602, configured to pre-process a fingerprint image to be matched to obtain a ridge line graph including ridge lines;

a third module 603, configured to extract a ridge point feature from a ridge line, to obtain a ridge point feature;

A fourth module 604, configured to extract a ridge line body feature from the ridge line, to obtain a ridge line body feature;

A fifth module 605, configured to perform a ridge line alignment operation on the two ridge line graphs according to the ridge line point feature and the ridge line body feature, to obtain an alignment region; the alignment area is an area where the two ridge line graphs are partially overlapped;

A sixth module 606, configured to perform a ridge point matching operation on the two ridge maps according to the overall position deviation of the ridge points in the alignment area, or the local position deviation of the local position with the highest alignment coincidence degree of the ridge points in the alignment area when the overall position deviation of the ridge points in the alignment area exceeds a preset deviation range, so as to obtain a ridge point matching result;

A seventh module 607, configured to obtain a matching result of the two fingerprint images to be matched according to the ridge line point matching result.

The specific implementation of the small-area fingerprint image matching device is basically the same as the specific embodiment of the small-area fingerprint image matching method, and will not be described herein.

Fig. 7 is a block diagram of an electronic device, according to an example embodiment.

An electronic device 700 according to such an embodiment of the present disclosure is described below with reference to fig. 7. The electronic device 700 shown in fig. 7 is merely an example and should not be construed to limit the functionality and scope of use of embodiments of the present disclosure in any way.

As shown in fig. 7, the electronic device 700 is embodied in the form of a general purpose computing device. Components of electronic device 700 may include, but are not limited to: at least one processing unit 710, at least one memory unit 720, a bus 730 connecting the different system components (including the memory unit 720 and the processing unit 710), a display unit 740, and the like.

Wherein the storage unit stores program code that is executable by the processing unit 710 such that the processing unit 710 performs the steps according to various exemplary embodiments of the present disclosure described in the above-described small area fingerprint image matching method section of the present specification.

The memory unit 720 may include readable media in the form of volatile memory units, such as Random Access Memory (RAM) 7201 and/or cache memory 7202, and may further include Read Only Memory (ROM) 7203.

The storage unit 720 may also include a program/utility 7204 having a set (at least one) of program modules 7205, such program modules 7205 including, but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment.

Bus 730 may be a bus representing one or more of several types of bus structures including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

The electronic device 700 may also communicate with one or more external devices 700' (e.g., keyboard, pointing device, bluetooth device, etc.), one or more devices that enable a user to interact with the electronic device 700, and/or any device (e.g., router, modem, etc.) that enables the electronic device 700 to communicate with one or more other computing devices. Such communication may occur through an input/output (I/O) interface 750. Also, electronic device 700 may communicate with one or more networks such as a Local Area Network (LAN), a Wide Area Network (WAN) and/or a public network, such as the Internet, through network adapter 760. Network adapter 760 may communicate with other modules of electronic device 700 via bus 730. It should be appreciated that although not shown, other hardware and/or software modules may be used in connection with electronic device 700, including, but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, data backup storage systems, and the like.

The embodiment of the application also provides a computer readable storage medium, which stores a computer program, and the computer program realizes the small-area fingerprint image matching method when being executed by a processor.

According to the small-area fingerprint image matching method, device, equipment and storage medium, ridge point feature extraction and line body feature extraction are carried out on a ridge line image which is subjected to pretreatment on a fingerprint image to be matched, so that ridge point features and ridge line body features are obtained, ridge line alignment operation which simultaneously considers the ridge point features and the ridge line body features is firstly carried out on the two ridge line images, an alignment area where the two ridge line images are partially overlapped is obtained, then ridge point matching operation which simultaneously considers the integral position deviation of ridge points in the alignment area and the local position deviation of the local position where the ridge point alignment overlapping degree is highest in the alignment area is carried out on the two ridge line images, a ridge point matching result is obtained, and the matching result of the two fingerprint images to be matched is obtained according to the ridge point matching result. Because the ridge line point characteristics and the ridge line body characteristics are extracted at the same time, characteristic elements required by fingerprint image matching are expanded, ridge line alignment operation and ridge line point matching operation are sequentially carried out, the ridge line alignment operation simultaneously considers the ridge line point characteristics and the ridge line body characteristics, the ridge line point matching operation simultaneously considers the integral position deviation of the ridge line points in the alignment area and the local position deviation of the local position with the highest alignment coincidence degree of the ridge line points in the alignment area, multiple matching and comparison can be carried out on the fingerprint images to be matched step by step, the defects of single fingerprint image matching mechanism and single type of matching characteristics are overcome, and the matching precision of the small-area fingerprint images is improved.

From the above description of embodiments, those skilled in the art will readily appreciate that the example embodiments described herein may be implemented in software, or may be implemented in software in combination with the necessary hardware. Thus, the technical solution according to the embodiments of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.) or on a network, including several instructions to cause a computing device (may be a personal computer, a server, or a network device, etc.) to perform the above-described method according to the embodiments of the present disclosure.

The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

The computer readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, with readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A readable storage medium may also be any readable medium that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Program code for carrying out operations of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of remote computing devices, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., connected via the Internet using an Internet service provider).

It will be appreciated by those skilled in the art that the modules may be distributed in a device according to the embodiments, or may be modified in one or more devices different from the embodiments. The modules of the above embodiments may be combined into one module, or may be further split into a plurality of sub-modules.

From the above description of embodiments, those skilled in the art will readily appreciate that the example embodiments described herein may be implemented in software, or may be implemented in software in combination with the necessary hardware. Thus, the technical solutions according to the embodiments of the present disclosure may be embodied in the form of a software product, which may be stored in a non-volatile storage medium (may be a CD-ROM, a U-disk, a mobile hard disk, etc.) or on a network, and include several instructions to cause a computing device (may be a personal computer, a server, a mobile terminal, or a network device, etc.) to perform the method according to the embodiments of the present disclosure.

Exemplary embodiments of the present disclosure are specifically illustrated and described above. It is to be understood that this disclosure is not limited to the particular arrangements, instrumentalities and methods of implementation described herein; on the contrary, the disclosure is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

Claims

1. A method for matching small-area fingerprint images, comprising:

Acquiring at least two fingerprint images to be matched;

obtaining a matching result of the two fingerprint images to be matched according to the ridge line point matching result;

The ridge line body feature comprises a line length feature, a curvature feature, a line width feature and a line type feature, and the ridge line alignment operation is performed on two ridge line graphs according to the ridge line point feature and the ridge line body feature to obtain an alignment region, and the ridge line alignment operation comprises the following steps:

Aligning the two ridge line graphs according to the first matching result, the second matching result or the third matching result to form the alignment area;

And according to the line length characteristics, performing similarity matching and rigid transformation matching on the ridge lines, determining a plurality of pairs of ridge lines with the highest matching degree, and obtaining a first matching result, wherein the method comprises the following steps:

2. The method for matching small-area fingerprint images according to claim 1, wherein the ridge point features include ridge point category features and ridge point association features, the ridge point feature extraction is performed on the ridge line to obtain ridge point features, including:

extracting ridge line points in the ridge line graph;

3. The method for matching a small-area fingerprint image according to claim 1, wherein the ridge line body features include a line length feature, a curvature feature, a line width feature, and a line type feature, and the performing line body feature extraction on the ridge line to obtain the ridge line body features includes:

4. The method for matching small-area fingerprint images according to claim 1, wherein performing a ridge point matching operation on two ridge maps according to a global position deviation of the ridge points in the alignment area or a local position deviation of a local position with a highest alignment overlapping degree of the ridge points in the alignment area when the global position deviation of the ridge points in the alignment area exceeds a preset deviation range, to obtain a ridge point matching result comprises:

5. The method of claim 4, wherein recursively generating each ridge pattern block with the reference region as a reference to obtain a target mapping transformation relationship, comprises:

6. A small-area fingerprint image matching apparatus, comprising:

a seventh module, configured to obtain a matching result of the two fingerprint images to be matched according to the ridge line point matching result;

7. An electronic device comprising a memory storing a computer program and a processor that when executing the computer program implements the small area fingerprint image matching method of any one of claims 1 to 5.

8. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the small area fingerprint image matching method of any one of claims 1 to 5.