An iris image matching method, device, equipment and medium
By using the covariance calculation of the target Mahalanobis distance and gradient parameters in iris image matching, combined with Gaussian filter denoising, the calculation error problem in the iris image matching process is solved, improving the matching accuracy and the security of the safe deposit box.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA CONSTRUCTION BANK
- Filing Date
- 2022-09-06
- Publication Date
- 2026-06-05
AI Technical Summary
The existing iris image matching process has large calculation errors, which affects the security of safe deposit boxes.
By determining the Mahalanobis distance of multiple targets based on feature points of the iris image to be detected, and combining the Mahalanobis distance of the standard iris image, the matching degree of the iris image is calculated using gradient parameters and covariance operations. Gaussian filter denoising and feature extraction are then used to improve matching accuracy.
It improves the accuracy of iris image matching and the reliability of the safe deposit box, reduces calculation errors, and enhances security.
Smart Images
Figure CN115761870B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of image processing, specifically to a method, apparatus, device, and medium for matching iris images. Background Technology
[0002] In the management of safe deposit boxes, biometric features such as fingerprint recognition and facial recognition are typically used as input parameters for opening the safe. This involves matching the user's real-time fingerprint (or facial image) with pre-stored reference fingerprints (or facial images), and opening the safe if the match is found. However, these fingerprint and facial recognition methods have relatively low security.
[0003] In recent years, iris recognition has been increasingly used in safe deposit box services. This involves comparing the iris image input by the user with a pre-stored standard iris image as a reference, and deciding whether to open the safe deposit box based on the matching result. However, existing methods for comparing the user-input iris image with the pre-stored standard iris image use Euclidean distance as the comparison parameter, which results in a relatively large calculation error. Summary of the Invention
[0004] This application provides an iris image matching method, apparatus, device, and medium to solve the problem of large calculation errors in the iris image matching process.
[0005] Firstly, an iris image matching method includes:
[0006] Based on each feature point of the iris image to be detected, multiple target Mahalanobis distances are determined, wherein each target Mahalanobis distance is determined based on the gradient parameters corresponding to any two target feature points of the iris image to be detected.
[0007] The degree of matching between the iris image to be detected and the pre-stored standard iris image is determined based on the sum of multiple target Mahalanobis distances and the sum of multiple standard Mahalanobis distances. The multiple standard Mahalanobis distances are determined based on the standard iris image.
[0008] In this embodiment, multiple target Mahalanobis distances are determined based on each feature point of the iris image to be detected. It should be noted that each target Mahalanobis distance is determined based on the gradient parameters corresponding to any two target feature points of the iris image to be detected. The matching degree between the iris image to be detected and the pre-stored standard iris image is determined based on the sum of multiple target Mahalanobis distances and the sum of multiple standard Mahalanobis distances. The standard Mahalanobis distance is determined based on the standard iris image. The above method of determining the iris image by Mahalanobis distance takes into account the inherent relationship between feature points, improves the matching accuracy between the iris image to be detected and the pre-stored standard iris image, and improves the reliability of the safe deposit box usage process.
[0009] In one possible embodiment, the target Mahalanobis distance is determined as follows:
[0010] The target difference matrix is transposed to obtain the transposed matrix. The target difference matrix is determined based on the first target feature matrix and the second target feature matrix. The first target feature matrix includes the first gradient parameter of the first target feature point and the second gradient parameter of the first target neighbor point. The second target feature matrix includes the third gradient parameter of the second target feature point and the fourth gradient parameter of the second target neighbor point. The first target neighbor point is the feature point with the closest distance to the first target feature point, and the second target neighbor point is the feature point with the closest distance to the second target feature point.
[0011] Perform covariance operation on the target difference matrix to obtain the covariance matrix;
[0012] The target Mahalanobis distance between the first target feature point and the second target feature point is determined based on the transpose matrix, covariance matrix, and weighting coefficients. The weighting coefficients are determined based on the radius of the iris image to be detected in polar coordinates.
[0013] In one possible embodiment, the gradient parameters include the gradient magnitude and the gradient direction, and the gradient magnitude corresponding to the target feature point is determined by the following method:
[0014] Based on the polar coordinates of the target feature points in the polar coordinate system Determine the first reference polar coordinates respectively Second reference polar coordinates Third reference polar coordinates and the fourth reference polar coordinates
[0015] The first reference modulus is determined based on the first and second reference polar coordinates. The second reference modulus is determined based on the third and fourth reference polar coordinates.
[0016] Based on the first reference modulus and the second reference modulus, determine the gradient modulus corresponding to the target feature point.
[0017] The gradient direction corresponding to the target feature point is determined in the following way:
[0018] Determine the unit radius vector e of the target feature point in polar coordinates. r and unit angle vector e θ ;
[0019] The gradient direction corresponding to the target feature point is determined based on the unit radius vector, unit angle vector, and radius.
[0020] Where r is the radius of the iris image to be detected corresponding to the target feature point in the polar coordinate system, θ is the center of the circle of the iris image to be detected corresponding to the target feature point in the polar coordinate system, and N is a natural number.
[0021] In one possible embodiment, before determining the multiple target Mahalanobis distances, the method further includes:
[0022] A Gaussian filter is used to denoise the iris image to be detected;
[0023] Feature extraction is performed on the denoised iris image to be detected to obtain various feature points.
[0024] In one possible embodiment, the degree of matching between the iris image to be detected and a pre-stored standard iris image is determined based on the sum of multiple target Mahalanobis distances and the sum of multiple standard Mahalanobis distances, including:
[0025] The distance difference is determined based on the sum of the Mahalanobis distances of multiple targets and the sum of the standard Mahalanobis distances.
[0026] If the distance difference is not greater than the preset distance difference threshold, then the iris image to be detected is determined to match the pre-stored standard iris image.
[0027] If the distance difference is greater than the preset distance difference threshold, it is determined that the iris image to be detected does not match the pre-stored standard iris image.
[0028] Secondly, this application provides an iris image matching device, the device comprising:
[0029] The determination module is used to determine multiple target Mahalanobis distances based on each feature point of the iris image to be detected, wherein each target Mahalanobis distance is determined based on the gradient parameters corresponding to any two target feature points of the iris image to be detected.
[0030] The matching module is used to determine the degree of matching between the iris image to be detected and the pre-stored standard iris image based on the sum of multiple target Mahalanobis distances and the sum of multiple standard Mahalanobis distances, wherein the multiple standard Mahalanobis distances are determined based on the standard iris image.
[0031] In one possible embodiment, the target Mahalanobis distance is determined as follows:
[0032] The target difference matrix is transposed to obtain the transposed matrix. The target difference matrix is determined based on the first target feature matrix and the second target feature matrix. The first target feature matrix includes the first gradient parameter of the first target feature point and the second gradient parameter of the first target neighbor point. The second target feature matrix includes the third gradient parameter of the second target feature point and the fourth gradient parameter of the second target neighbor point. The first target neighbor point is the feature point with the closest distance to the first target feature point, and the second target neighbor point is the feature point with the closest distance to the second target feature point.
[0033] Perform covariance operation on the target difference matrix to obtain the covariance matrix;
[0034] The target Mahalanobis distance between the first target feature point and the second target feature point is determined based on the transpose matrix, covariance matrix, and weighting coefficients. The weighting coefficients are determined based on the radius of the iris image to be detected in polar coordinates.
[0035] In one possible embodiment, the gradient parameters include the gradient magnitude and the gradient direction, and the gradient magnitude corresponding to the target feature point is determined by the following method:
[0036] Based on the polar coordinates of the target feature points in the polar coordinate system Determine the first reference polar coordinates respectively Second reference polar coordinates Third reference polar coordinates and the fourth reference polar coordinates
[0037] The first reference modulus is determined based on the first and second reference polar coordinates. The second reference modulus is determined based on the third and fourth reference polar coordinates.
[0038] Based on the first reference modulus and the second reference modulus, determine the gradient modulus corresponding to the target feature point.
[0039] The gradient direction corresponding to the target feature point is determined in the following way:
[0040] Determine the unit radius vector e of the target feature point in polar coordinates. r and unit angle vector e θ ;
[0041] The gradient direction corresponding to the target feature point is determined based on the unit radius vector, unit angle vector, and radius.
[0042]
[0043] Where r is the radius of the iris image to be detected corresponding to the target feature point in the polar coordinate system, θ is the center of the circle of the iris image to be detected corresponding to the target feature point in the polar coordinate system, and N is a natural number.
[0044] In one possible embodiment, before determining the multiple target Mahalanobis distances, the method further includes:
[0045] A Gaussian filter is used to denoise the iris image to be detected;
[0046] Feature extraction is performed on the denoised iris image to be detected to obtain various feature points.
[0047] In one possible embodiment, the matching degree between the iris image to be detected and a pre-stored standard iris image is determined based on the sum of multiple target Mahalanobis distances and the sum of multiple standard Mahalanobis distances. The matching module is used for:
[0048] The distance difference is determined based on the sum of the Mahalanobis distances of multiple targets and the sum of the standard Mahalanobis distances.
[0049] If the distance difference is not greater than the preset distance difference threshold, then the iris image to be detected is determined to match the pre-stored standard iris image.
[0050] If the distance difference is greater than the preset distance difference threshold, it is determined that the iris image to be detected does not match the pre-stored standard iris image.
[0051] Thirdly, this application provides an electronic device, comprising:
[0052] Memory, used to store program instructions;
[0053] A processor is configured to invoke program instructions stored in the memory and execute the steps of the method described in any one of the first aspects according to the obtained program instructions.
[0054] Fourthly, this application provides a computer-readable storage medium storing a computer program, the computer program including program instructions, which, when executed by a computer, cause the computer to perform the method described in any one of the first aspects.
[0055] Fifthly, this application provides a computer program product comprising: computer program code, which, when run on a computer, causes the computer to perform the method described in any one of the first aspects. Attached Figure Description
[0056] Figure 1 This is a system architecture diagram for matching iris images provided in an embodiment of this application;
[0057] Figure 2 A flowchart illustrating a method for matching iris images provided in this application embodiment;
[0058] Figure 3 A flowchart for determining the target Mahalanobis distance is provided in this application embodiment;
[0059] Figure 4 A flowchart for determining whether an iris image to be detected matches a standard iris image is provided in an embodiment of this application;
[0060] Figure 5 A structural diagram of an iris image matching device provided in an embodiment of this application;
[0061] Figure 6 This is a structural diagram of an electronic device provided in an embodiment of this application. Detailed Implementation
[0062] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. Unless otherwise specified, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that shown here.
[0063] The terms "first" and "second" in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the term "comprising" and any variations thereof are intended to cover non-exclusive protection. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices. The term "multiple" in this application can mean at least two, for example, two, three, or more, and the embodiments of this application do not impose limitations.
[0064] The data collection, dissemination, and use in this application all comply with relevant national laws and regulations.
[0065] Before introducing the iris image matching method provided in the embodiments of this application, the technical background of the embodiments of this application will be described in detail below for ease of understanding.
[0066] When using iris recognition to open a safe deposit box, it's necessary to compare the user-input iris image with a pre-stored standard iris image as a reference, and decide whether to open the safe deposit box based on the matching result. However, existing methods use Euclidean distance as a comparison parameter, resulting in significant calculation errors.
[0067] To address the problem of large calculation errors in the iris image matching process, the preferred embodiments of this disclosure will be described in detail below with reference to the accompanying drawings.
[0068] See Figure 1 As shown in this embodiment of the disclosure, the system includes at least one electronic device. Figure 1 In this process, the electronic device processes the iris image to be detected. At the same time, a standard iris image is pre-stored inside the electronic device. The electronic device matches the iris image to be detected with the standard iris image, which will be described in detail below.
[0069] See Figure 2 As shown in the embodiments of this disclosure, the specific process of an iris image matching method is as follows:
[0070] Step 201: Based on each feature point of the iris image to be detected, determine multiple target Mahalanobis distances, wherein each target Mahalanobis distance is determined based on the gradient parameters corresponding to any two target feature points of the iris image to be detected.
[0071] Considering that different safe deposit boxes will capture different iris images during use via acquisition devices (e.g., cameras), and that these images will inevitably contain noise, the determination of the Mahalanobis distances for multiple targets also includes:
[0072] 1) Use a Gaussian filter to denoise the iris image to be detected.
[0073] Typically, the noise in the iris image to be detected is white noise. Therefore, during the process, a Gaussian filter is used to denoise the iris image to be detected. That is, the iris image to be detected is input into the Gaussian filter as an input parameter, and the noise contained therein is removed by the Gaussian filter. Thus, the output of the Gaussian filter is the denoised iris image to be detected. It should be noted that the iris image to be detected is usually elliptical, and accordingly, it will be represented in polar coordinates.
[0074] Suppose that the iris image to be detected, which is used as the input parameter, is O(x,y), and the expression of the Gaussian filter is g(x,y). Then, the denoising process is to multiply the above O(x,y) by g(x,y) to obtain the denoised iris image to be detected, H(x,y), where H(x,y) = O(x,y) * g(x,y).
[0075] 2) Extract features from the denoised iris image to be detected to obtain each feature point.
[0076] During the implementation process, after obtaining the denoised iris image to be detected, further feature extraction is performed. For example, a scale-invariant feature transformation algorithm is used for feature extraction. After feature extraction, multiple feature points corresponding to the denoised iris image to be detected are obtained.
[0077] Additionally, it should be noted that when the acquisition device captures images of the human eye, factors such as the shooting angle, brightness, and distance may cause the iris image to shift or rotate. Therefore, before feature extraction, the iris image to be detected can be normalized, that is, the circular iris image to be detected is normalized into a rectangular image, and the iris image to be detected is transformed from polar coordinates to rectangular coordinates.
[0078] After obtaining the above-mentioned feature points, the Mahalanobis distance is determined based on the feature points included therein.
[0079] For details, please refer to Figure 3 As shown, the target Mahalanobis distance is determined in the following way:
[0080] Step 2011: Transpose the target difference matrix to obtain the transpose matrix. The target difference matrix is determined based on the first target feature matrix and the second target feature matrix. The first target feature matrix includes the first gradient parameter of the first target feature point and the second gradient parameter of the first target neighbor point. The second target feature matrix includes the third gradient parameter of the second target feature point and the fourth gradient parameter of the second target neighbor point. The first target neighbor point is the feature point with the closest distance to the first target feature point, and the second target neighbor point is the feature point with the closest distance to the second target feature point.
[0081] It should be noted that, in order to determine the Mahalanobis distance, the transpose matrix and covariance matrix are first constructed in this embodiment. Therefore, based on the aforementioned multiple feature points, the first target feature point, the first target neighbor point, the second target feature point, and the second target neighbor point need to be determined. The first target feature point and the second target feature point are any two of the aforementioned multiple feature points. The first target neighbor point is the feature point closest to the first target feature point, and the second target neighbor point is the feature point closest to the second target feature point; the distance here is the straight-line distance between the two points.
[0082] After determining the first target feature point, the first target neighbor point, the second target feature point, and the second target neighbor point, the gradient parameters corresponding to each of these points are calculated. The first target feature matrix is constructed using the first gradient parameter of the first target feature point and the second gradient parameter of the first target neighbor point, and the second target feature matrix is constructed using the third gradient parameter of the second target feature point and the fourth gradient parameter of the second target neighbor point.
[0083] During implementation, the target difference matrix is obtained by subtracting the second target feature matrix from the first target feature matrix, and then the target difference matrix is transposed to obtain the transpose matrix. Assume the first target feature matrix is t. a The second target feature matrix is t b Therefore, the target difference matrix is (t a -t b ) T .
[0084] The following section provides a detailed explanation of how the gradient parameters are calculated:
[0085] I. Gradient parameters include gradient magnitude and gradient direction. The gradient magnitude corresponding to the target feature point is determined in the following way:
[0086] (1) Based on the polar coordinates of the target feature points in the polar coordinate system Determine the first reference polar coordinates respectively Second reference polar coordinates Third reference polar coordinates and the fourth reference polar coordinates
[0087] During implementation, one of the aforementioned feature points will be used as the target feature point for illustrative purposes. Once the polar coordinates of this target feature point in the polar coordinate system are determined... Then, using the above polar coordinates Extend the range as a reference, for example, extend r to r+N, rN, and extend θ to θ-N and θ+N.
[0088] Furthermore, the first reference polar coordinates are obtained. Second reference polar coordinates Third reference polar coordinates and the fourth reference polar coordinates
[0089] Here's a supplementary explanation: in the above formula, r is the radius of the iris image to be detected corresponding to the target feature point in the polar coordinate system, θ is the center of the circle of the iris image to be detected corresponding to the target feature point in the polar coordinate system, and N is a natural number.
[0090] (2) Determine the first reference magnitude based on the first and second reference polar coordinates. The second reference modulus is determined based on the third and fourth reference polar coordinates.
[0091] During implementation, through formulas That is, the first reference polar coordinates and the second reference polar coordinates determine the first reference magnitude, and, through the formula That is, the second reference modulus is determined by the third and fourth reference polar coordinates.
[0092] (3) Determine the gradient magnitude corresponding to the target feature point based on the first reference magnitude and the second reference magnitude.
[0093] Reuse formula That is, the first reference modulus value and the second reference modulus value are used to determine the gradient modulus corresponding to the target feature point.
[0094] II. Determine the gradient direction corresponding to the target feature point using the following method:
[0095] [1] Determine the unit radius vector e of the target feature point in the polar coordinate system. r and unit angle vector e θ .
[0096] Similarly, the target feature point here is one of the aforementioned feature points; that is, after calculating the gradient magnitude of a feature point, the gradient direction is calculated. Before calculating the gradient direction corresponding to the target feature point, the unit radius vector e of the target feature point in the polar coordinate system is determined. r and unit angle vector e θ .
[0097] [2] Determine the gradient direction of the target feature point based on the unit radius vector, unit angle vector and radius.
[0098] During implementation, through formulas That is, differentiate the unit radius vector and the unit angle vector, and multiply the differentiated result with the reciprocal of the radius r to calculate the gradient direction corresponding to the target feature point.
[0099] Step 2012: Perform covariance operation on the target difference matrix to obtain the covariance matrix.
[0100] After obtaining the transpose matrix, it is also necessary to determine the covariance matrix, which is obtained by performing a covariance operation on the aforementioned target difference matrix. Assume the first target feature matrix is t. a The second target feature matrix is t b Therefore, the covariance matrix is ∑ -1 (t a -t b ).
[0101] Step 2013: Determine the target Mahalanobis distance between the first target feature point and the second target feature point based on the transpose matrix, covariance matrix and weight coefficients, wherein the weight coefficients are determined based on the radius of the iris image to be detected in the polar coordinate system.
[0102] Continuing with the previous example, after determining the transpose matrix, covariance matrix, and weight coefficients using the first and second target feature points, the formula can be used to... Calculate the target Mahalanobis distance between the first target feature point and the second target feature point.
[0103] It should be noted that F in the above formula r This refers to the weighting coefficient, meaning that both the first and second target feature points correspond to a weighting coefficient. Specifically, the weighting coefficient is determined based on the radius of the iris image to be detected in polar coordinates. Typically, the radius can be divided according to a preset rule, and each segment corresponds to a weighting coefficient value. For example, when the radius is 10, the radius is divided into segments 1, 2, 3, 4…9, 10, with each segment corresponding to a weighting coefficient of 1, 2, 3, 4…9, 10. If both the first and second target feature points are in the 3rd segment, the corresponding weighting coefficient is 3. If one target feature point is in the 3rd segment and the other in the 4th segment, the larger weighting coefficient is used.
[0104] Step 202: Based on the sum of multiple target Mahalanobis distances and the sum of multiple standard Mahalanobis distances, determine the degree of matching between the iris image to be detected and the pre-stored standard iris image, wherein the multiple standard Mahalanobis distances are determined based on the standard iris image.
[0105] See Figure 4 As shown, the steps for calculating the matching degree mentioned above specifically include:
[0106] Step 2021: Determine the distance difference based on the sum of multiple target Mahalanobis distances and the sum of multiple standard Mahalanobis distances.
[0107] During implementation, after obtaining the target Mahalanobis distances corresponding to every two feature points among all feature points, i.e., multiple target Mahalanobis distances, the sum of the above target Mahalanobis distances is calculated, i.e., the sum of multiple target Mahalanobis distances.
[0108] Similarly, for standard iris images, multiple standard Mahalanobis distances are calculated using the same method. The specific calculation process is similar to that of the target Mahalanobis distance and will not be repeated here. The obtained standard Mahalanobis distances are then summed to obtain the sum of the multiple standard Mahalanobis distances.
[0109] During implementation, the difference between the sum of multiple target Mahalanobis distances and the sum of multiple standard Mahalanobis distances is calculated to determine the distance difference value.
[0110] Step 2022: If the distance difference is not greater than the preset distance difference threshold, then the iris image to be detected is determined to match the pre-stored standard iris image.
[0111] During implementation, if the distance difference determined above is not greater than the preset distance difference threshold, that is, the distance difference is within the range of the preset distance difference threshold, then the iris image to be detected is determined to match the pre-stored standard iris image.
[0112] Step 2023: If the distance difference is greater than the preset distance difference threshold, it is determined that the iris image to be detected does not match the pre-stored standard iris image.
[0113] During implementation, if the distance difference determined above is greater than the preset distance difference threshold, that is, the distance difference is not within the range of the preset distance difference threshold, it is determined that the iris image to be detected does not match the pre-stored standard iris image.
[0114] It should be noted that the aforementioned distance difference threshold is the average distance difference between the representation derived from historical iris images or neural networks and the pre-stored standard iris images.
[0115] Based on the same inventive concept, embodiments of this application provide an iris image matching device, see reference. Figure 5 As shown, the device includes:
[0116] The determination module 501 is used to determine multiple target Mahalanobis distances based on each feature point of the iris image to be detected, wherein each target Mahalanobis distance is determined based on the gradient parameters corresponding to any two target feature points of the iris image to be detected.
[0117] The matching module 502 is used to determine the degree of matching between the iris image to be detected and the pre-stored standard iris image based on the sum of multiple target Mahalanobis distances and the sum of multiple standard Mahalanobis distances, wherein the multiple standard Mahalanobis distances are determined based on the standard iris image.
[0118] In one possible embodiment, the target Mahalanobis distance is determined as follows:
[0119] The target difference matrix is transposed to obtain the transposed matrix. The target difference matrix is determined based on the first target feature matrix and the second target feature matrix. The first target feature matrix includes the first gradient parameter of the first target feature point and the second gradient parameter of the first target neighbor point. The second target feature matrix includes the third gradient parameter of the second target feature point and the fourth gradient parameter of the second target neighbor point. The first target neighbor point is the feature point with the closest distance to the first target feature point, and the second target neighbor point is the feature point with the closest distance to the second target feature point.
[0120] Perform covariance operation on the target difference matrix to obtain the covariance matrix;
[0121] The target Mahalanobis distance between the first target feature point and the second target feature point is determined based on the transpose matrix, covariance matrix, and weighting coefficients. The weighting coefficients are determined based on the radius of the iris image to be detected in polar coordinates.
[0122] In one possible embodiment, the gradient parameters include the gradient magnitude and the gradient direction, and the gradient magnitude corresponding to the target feature point is determined by the following method:
[0123] Based on the polar coordinates of the target feature points in the polar coordinate system Determine the first reference polar coordinates respectively Second reference polar coordinates Third reference polar coordinates and the fourth reference polar coordinates
[0124] The first reference modulus is determined based on the first and second reference polar coordinates. The second reference modulus is determined based on the third and fourth reference polar coordinates.
[0125] Based on the first reference modulus and the second reference modulus, determine the gradient modulus corresponding to the target feature point.
[0126] The gradient direction corresponding to the target feature point is determined in the following way:
[0127] Determine the unit radius vector e of the target feature point in polar coordinates. r and unit angle vector eθ ;
[0128] The gradient direction corresponding to the target feature point is determined based on the unit radius vector, unit angle vector, and radius.
[0129] Where r is the radius of the iris image to be detected corresponding to the target feature point in the polar coordinate system, θ is the center of the circle of the iris image to be detected corresponding to the target feature point in the polar coordinate system, and N is a natural number.
[0130] In one possible embodiment, before determining the multiple target Mahalanobis distances, the method further includes:
[0131] A Gaussian filter is used to denoise the iris image to be detected;
[0132] Feature extraction is performed on the denoised iris image to be detected to obtain various feature points.
[0133] In one possible embodiment, the matching degree between the iris image to be detected and a pre-stored standard iris image is determined based on the sum of multiple target Mahalanobis distances and the sum of multiple standard Mahalanobis distances. The matching module 502 is used for:
[0134] The distance difference is determined based on the sum of the Mahalanobis distances of multiple targets and the sum of the standard Mahalanobis distances.
[0135] If the distance difference is not greater than the preset distance difference threshold, then the iris image to be detected is determined to match the pre-stored standard iris image.
[0136] If the distance difference is greater than the preset distance difference threshold, it is determined that the iris image to be detected does not match the pre-stored standard iris image.
[0137] Based on the same inventive concept, embodiments of this application provide an electronic device that can realize the iris image matching function discussed above. (See also...) Figure 6 As shown, the device includes a memory 601 and a processor 602, wherein the memory 601 stores program code that, when executed by the processor, causes the processor 602 to perform the steps in the iris image matching method according to various exemplary embodiments of this application described above.
[0138] Based on the same inventive concept, embodiments of this application provide a computer-readable storage medium. The computer program product includes computer program code, which, when executed on a computer, causes the computer to perform any of the iris image matching methods discussed above. Since the principle by which the above-described computer-readable storage medium solves the problem is similar to that of the iris image matching method, the implementation of the above-described computer-readable storage medium can be found in the implementation of the method; repeated details will not be elaborated further.
[0139] Based on the same inventive concept, this application also provides a computer program product, which includes computer program code that, when run on a computer, causes the computer to execute any of the iris image matching methods discussed above. Since the principle by which the above computer program product solves the problem is similar to that of the iris image matching method, the implementation of the above computer program product can be referred to the implementation of the method, and repeated details will not be elaborated further.
[0140] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can 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, etc.) containing computer-usable program code.
[0141] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can 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, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0142] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0143] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of user-operated steps to be executed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0144] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.
Claims
1. A method for matching iris images, characterized in that, The method includes: Based on each feature point of the iris image to be detected and the gradient parameters of each feature point, multiple target Mahalanobis distances are determined, wherein each target Mahalanobis distance is determined based on the gradient parameters corresponding to any two target feature points of the iris image to be detected. The degree of matching between the iris image to be detected and the pre-stored standard iris image is determined based on the sum of multiple target Mahalanobis distances and the sum of multiple standard Mahalanobis distances, wherein the multiple standard Mahalanobis distances are determined based on the standard iris image; The target Mahalanobis distance is determined in the following manner: The target difference matrix is transposed to obtain the transposed matrix, wherein the target difference matrix is determined based on the first target feature matrix and the second target feature matrix. The first target feature matrix includes the first gradient parameter of the first target feature point and the second gradient parameter of the first target neighbor point. The second target feature matrix includes the third gradient parameter of the second target feature point and the fourth gradient parameter of the second target neighbor point. The first target neighbor point is the feature point with the closest distance to the first target feature point, and the second target neighbor point is the feature point with the closest distance to the second target feature point. Perform covariance operation on the target difference matrix to obtain the covariance matrix; Based on the transpose matrix, the covariance matrix, and the weighting coefficients, the target Mahalanobis distance between the first target feature point and the second target feature point is determined, wherein the weighting coefficients are determined based on the radius of the iris image to be detected in polar coordinates.
2. The method as described in claim 1, characterized in that, The gradient parameters include the gradient magnitude and gradient direction. The gradient magnitude corresponding to the target feature point is determined in the following way: Based on the polar coordinates of the target feature points in the polar coordinate system Determine the first reference polar coordinates respectively Second reference polar coordinates Third reference polar coordinates and the fourth reference polar coordinates ; The first reference modulus is determined based on the first and second reference polar coordinates. The second reference modulus is determined based on the third and fourth reference polar coordinates. ; Based on the first reference modulus and the second reference modulus, the gradient modulus corresponding to the target feature point is determined. ; The gradient direction corresponding to the target feature point is determined in the following way: Determine the unit radius vector of the target feature point in polar coordinates. and unit angle vector ; The gradient direction corresponding to the target feature point is determined based on the unit radius vector, the unit angle vector, and the radius. ; Where r is the radius of the iris image to be detected corresponding to the target feature point in the polar coordinate system. Let N be the center of the circle in the polar coordinate system of the iris image to be detected corresponding to the target feature point, where N is a natural number.
3. The method as described in claim 1, characterized in that, Before determining the Mahalanobis distances of multiple targets, the method further includes: The iris image to be detected is denoised using a Gaussian filter. Feature extraction is performed on the denoised iris image to be detected to obtain each of the aforementioned feature points.
4. The method according to any one of claims 1 to 3, characterized in that, The determination of the matching degree between the iris image to be detected and the pre-stored standard iris images based on the sum of multiple target Mahalanobis distances and the sum of multiple standard Mahalanobis distances includes: The distance difference is determined based on the sum of the multiple target Mahalanobis distances and the sum of multiple standard Mahalanobis distances; If the distance difference is not greater than a preset distance difference threshold, then the iris image to be detected is determined to match a pre-stored standard iris image; If the distance difference is greater than a preset distance difference threshold, it is determined that the iris image to be detected does not match the pre-stored standard iris image.
5. An iris image matching device, characterized in that, include: The determination module is used to determine multiple target Mahalanobis distances based on each feature point of the iris image to be detected and the gradient parameters of each feature point, wherein each target Mahalanobis distance is determined based on the gradient parameters corresponding to any two target feature points of the iris image to be detected. A matching module is used to determine the degree of matching between the iris image to be detected and a pre-stored standard iris image based on the sum of multiple target Mahalanobis distances and the sum of multiple standard Mahalanobis distances, wherein the multiple standard Mahalanobis distances are determined based on the standard iris image; The target Mahalanobis distance is determined in the following manner: The target difference matrix is transposed to obtain the transposed matrix, wherein the target difference matrix is determined based on the first target feature matrix and the second target feature matrix. The first target feature matrix includes the first gradient parameter of the first target feature point and the second gradient parameter of the first target neighbor point. The second target feature matrix includes the third gradient parameter of the second target feature point and the fourth gradient parameter of the second target neighbor point. The first target neighbor point is the feature point with the closest distance to the first target feature point, and the second target neighbor point is the feature point with the closest distance to the second target feature point. Perform covariance operation on the target difference matrix to obtain the covariance matrix; Based on the transpose matrix, the covariance matrix, and the weighting coefficients, the target Mahalanobis distance between the first target feature point and the second target feature point is determined, wherein the weighting coefficients are determined based on the radius of the iris image to be detected in polar coordinates.
6. An electronic device, characterized in that, include: Memory, used to store program instructions; A processor is configured to invoke program instructions stored in the memory and execute the steps of the method according to any one of claims 1-4.
7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, the computer program including program instructions that, when executed by a computer, cause the computer to perform the method as described in any one of claims 1-4.
8. A computer program product, characterized in that, The computer program product includes: computer program code, which, when run on a computer, causes the computer to perform the method described in any one of claims 1-4.