A feature vector based fingerprint recognition encryption system

Abstract

The application provides a feature vector-based fingerprint identification encryption system, which comprises a fingerprint collection module, an image processing module, a feature processing module, an information storage module and a checking module, the fingerprint collection module is used for collecting fingerprint images, the image processing module is used for pre-processing the collected fingerprint images, the feature processing module is used for processing the pre-processed images into feature vectors, the information storage module is used for storing the feature vectors, and the checking module is used for identifying the fingerprint images based on the feature vectors; the system can convert the collected fingerprint images into feature vectors, and then calculate and process the fingerprint images to be verified by using the feature vectors, so that the fingerprint images to be verified can be accurately identified even when the information amount is small, and the efficiency of fingerprint identification is improved.

Description

Technical Field

[0001] This invention relates to the field of graphic data reading, and more specifically to a fingerprint recognition encryption system based on feature vectors. Background Technology

[0002] Fingerprint recognition involves classifying and comparing fingerprints to identify individuals. As a biometric identification technology, fingerprint recognition has matured in the new century and entered the realm of human production and daily life. Fingerprints are the patterns formed by the raised and recessed skin at the tips of human fingers. They are formed before birth and their shape remains unchanged as an individual grows, only their intensity varies. Each person's fingerprint is unique, allowing for good differentiation among numerous details. Fingerprints contain many feature points that provide confirmation of uniqueness, forming the basis of fingerprint recognition. These features are divided into overall features and local features. Overall features include the core point, triangulation points, and the number of ridges; local features are the detailed characteristics of the fingerprint, such as the direction, curvature, and position of the nodes at the feature points—all important indicators for distinguishing different fingerprints. However, this traditional recognition method requires a high level of information from the fingerprints being identified. When the collected fingerprint information is incomplete, multiple collections are often needed for successful identification, reducing efficiency.

[0003] The foregoing description of the background art is intended only to facilitate understanding of the invention. This description does not endorse or acknowledge any common general knowledge in the materials mentioned.

[0004] Many fingerprint recognition systems have been developed. Through extensive research and reference, we found existing systems such as the one disclosed in publication number CN104616001B. These systems generally include a backlight panel to provide incident light to illuminate the finger; an image sensor to acquire fingertip and pulse images; an acquisition module to control the image sensor to acquire fingertip images and to continuously acquire multiple pulse images; a processing module to receive the fingertip and pulse images acquired by the acquisition module; to obtain a fingertip fingerprint image from the fingertip image; to obtain a pulse waveform image from the multiple pulse images; and a recognition module to combine the fingertip fingerprint and pulse waveform images for identification. However, this system still requires a fingerprint image with high information content. When the finger is dirty or the finger is lightly pressed, the fingerprint often cannot be recognized successfully, and the recognition efficiency needs to be improved. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings by proposing a fingerprint recognition encryption system based on feature vectors.

[0006] The present invention adopts the following technical solution:

[0007] A fingerprint recognition encryption system based on feature vectors includes a fingerprint acquisition module, an image processing module, a feature processing module, an information storage module, and a verification module. The fingerprint acquisition module is used to acquire fingerprint images. The image processing module is used to preprocess the acquired fingerprint images. The feature processing module is used to process the preprocessed images into feature vectors. The information storage module is used to store the feature vectors. The verification module recognizes the fingerprint images based on the feature vectors.

[0008] The image processing module divides the fingerprint image into six square images. The feature processing module converts these six images into six matrices: X1, X2, X3, X4, X5, and X6. The elements of these six matrices consist of 0 and 1, where 1 represents ridge pixels in the fingerprint image and 0 represents non-ridge pixels. The feature processing module obtains the comprehensive matrix Y according to the following formula:

[0009] Y = X1 + X2 + X3 + X4 + X5 + X6;

[0010] The feature processing module counts the number of elements with values ​​of 0, 1, 2, 3, 4, 5, and 6 in the comprehensive matrix, which are denoted as n1, n2, n3, n4, n5, n6, and n7, respectively.

[0011] The feature processing module calculates the feature vector stored in the information storage module according to the following formula.

[0012] Where, λ i (i = 1, 2, 3, 4, 5, 6, 7) are the eigenvalues ​​of matrices X1, X2, X3, X4, X5, X6, and Y, respectively. These are the eigenvectors of matrices X1, X2, X3, X4, X5, X6, and Y, respectively.

[0013] The verification module extracts a square image from the acquired fingerprint image and converts it into a matrix Z. By shifting the first column element of matrix Z to the last column, m new matrices Z are obtained. k , (k = 1, 2, 3, ..., m), where m is the number of columns in matrix Z, and the proofreading module will check matrix Z k With feature vectors Multiplying them together yields m vectors. The proofreading module will process m vectors A verification matrix Z0 is constructed and the determinant value D = det(Z0) of the verification matrix Z0 is calculated. When the absolute value of the determinant value D is less than the verification threshold, the fingerprint image passes the verification.

[0014] Furthermore, the system includes an enrollment mode and an unlocking mode. In the enrollment mode, the image acquired by the fingerprint acquisition module is processed by the image processing module and the feature processing module to generate a feature vector, which is then stored in the information storage module. In the unlocking mode, the verification module obtains the feature vector from the information storage module and performs verification analysis on the fingerprint image obtained from the fingerprint acquisition module. Once the verification is successful, the unlocking is successful.

[0015] Furthermore, the image processing module performs specification judgment on the acquired fingerprint image. If the fingerprint image does not meet the specifications, no further operations will be performed on the fingerprint image.

[0016] Furthermore, the image processing module divides the pixels in the fingerprint image into ridge pixels and non-ridge pixels, and sets the connected ridge pixels as a quasi-ridge set. When the width of the quasi-ridge set is less than the width threshold, it is called a ridge set. The image processing module counts the number of ridge sets. When the number of ridge sets is within a standard range, the fingerprint image meets the specifications.

[0017] Furthermore, the image processing module establishes a coordinate system with the fingerprint ring center and divides the fingerprint image into 6 square images according to the coordinate system. The coordinate range of each square image is: {x∈[-3a, -a], y∈[a, 3a]}, {x∈[-a, a], y∈[a, 3a]}, {x∈[a, 3a], y∈[a, 3a]}, {x∈[-3a, -a], y∈[-a, a]}, {x∈[-a, a], y∈[-a, a]}, {x∈[a, 3a], y∈[-a, a]};

[0018] Where 2a is the side length of the square image in the coordinate system.

[0019] The beneficial effects achieved by this invention are:

[0020] This system segments fingerprint images into multiple square images, converts each square image into a matrix, and then processes it into feature vectors. After comprehensive processing of the feature vectors, a final feature vector is obtained. The final feature vector can reflect the characteristics of each square image. When verifying the fingerprint image to be processed, even if the amount of information in the fingerprint image to be processed is small, a value that meets the requirements can still be obtained by processing it with the feature vector. This recognition method can improve the fingerprint recognition efficiency.

[0021] To further understand the features and technical content of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are for reference and illustration only and are not intended to limit the present invention. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the overall structural framework of the present invention;

[0023] Figure 2 This is a schematic diagram of the data entry process of the present invention;

[0024] Figure 3 This is a schematic diagram of the unlocking mode process of the present invention;

[0025] Figure 4 This is a schematic diagram of fingerprint image segmentation according to the present invention;

[0026] Figure 5 This is a schematic diagram illustrating the changes in data analyzed by the proofreading module of this invention. Detailed Implementation

[0027] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can understand the advantages and effects of the present invention from the content disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and various details in this specification can also be modified and changed based on different viewpoints and applications without departing from the spirit of the present invention. Furthermore, the accompanying drawings of the present invention are for simple illustrative purposes only and are not depictions of actual dimensions; this is stated beforehand. The following embodiments will further describe the relevant technical content of the present invention in detail, but the disclosed content is not intended to limit the scope of protection of the present invention.

[0028] Example 1.

[0029] This embodiment provides a fingerprint recognition encryption system based on feature vectors, combined with... Figure 1 The system includes a fingerprint acquisition module, an image processing module, a feature processing module, an information storage module, and a verification module. The fingerprint acquisition module is used to acquire fingerprint images. The image processing module is used to preprocess the acquired fingerprint images. The feature processing module is used to process the preprocessed images into feature vectors. The information storage module is used to store the feature vectors. The verification module recognizes the fingerprint images based on the feature vectors.

[0030] The image processing module divides the fingerprint image into six square images. The feature processing module converts these six images into six matrices: X1, X2, X3, X4, X5, and X6. The elements of these six matrices consist of 0 and 1, where 1 represents ridge pixels in the fingerprint image and 0 represents non-ridge pixels. The feature processing module obtains the comprehensive matrix Y according to the following formula:

[0031] Y = X1 + X2 + X3 + X4 + X5 + X6;

[0032] The feature processing module counts the number of elements with values ​​of 0, 1, 2, 3, 4, 5, and 6 in the comprehensive matrix, which are denoted as n1, n2, n3, n4, n5, n6, and n7, respectively.

[0033] The feature processing module calculates the feature vector stored in the information storage module according to the following formula.

[0034]

[0035] Where, λ i (i = 1, 2, 3, 4, 5, 6, 7) are the eigenvalues ​​of matrices X1, X2, X3, X4, X5, X6, and Y, respectively. These are the eigenvectors of matrices X1, X2, X3, X4, X5, X6, and Y, respectively.

[0036] The verification module extracts a square image from the acquired fingerprint image and converts it into a matrix Z. By shifting the first column element of matrix Z to the last column, m new matrices Z are obtained. k , (k = 1, 2, 3, ..., m), where m is the number of columns in matrix Z, and the proofreading module will check matrix Z k With feature vectors Multiplying them together yields m vectors. The proofreading module will process m vectors A verification matrix Z0 is constructed and the determinant value D = det(Z0) of the verification matrix Z0 is calculated. When the absolute value of the determinant value D is less than the verification threshold, the fingerprint image passes the verification.

[0037] The system includes an input mode and an unlock mode. In the input mode, the image acquired by the fingerprint acquisition module is processed by the image processing module and the feature processing module to generate a feature vector, which is then stored in the information storage module. In the unlock mode, the verification module obtains the feature vector from the information storage module and performs verification analysis on the fingerprint image obtained from the fingerprint acquisition module. Once the verification is successful, the unlock is successful.

[0038] The image processing module performs specification judgment on the acquired fingerprint image. If the fingerprint image does not meet the specifications, no further operation will be performed on the fingerprint image.

[0039] The image processing module divides the pixels in the fingerprint image into ridge pixels and non-ridge pixels. The ridge pixels with connectivity are set as a quasi-ridge set. When the width of the quasi-ridge set is less than the width threshold, it is called a ridge set. The image processing module counts the number of ridge sets. When the number of ridge sets is within a standard range, the fingerprint image meets the specifications.

[0040] The image processing module establishes a coordinate system based on the fingerprint ring center and divides the fingerprint image into 6 square images according to the coordinate system. The coordinate range of each square image is: {x∈[-3a, -a], y∈[a, 3a]}, {x∈[-a, a], y∈[a, 3a]}, {x∈[a, 3a], y∈[a, 3a]}, {x∈[-3a, -a], y∈[-a, a]}, {x∈[-a, a], y∈[-a, a]}, {x∈[a, 3a], y∈[-a, a]};

[0041] Where 2a is the side length of the square image in the coordinate system.

[0042] Example 2.

[0043] This embodiment provides a fingerprint recognition encryption system based on feature vectors, including a fingerprint acquisition module, an image processing module, a feature processing module, an information storage module, and a verification module. The fingerprint acquisition module is used to acquire the user's fingerprint image, the image processing module is used to perform image preprocessing on the acquired fingerprint image, the feature processing module is used to process the preprocessed image into feature vectors, the information storage module is used to store the feature vector information, and the verification module is used to recognize the fingerprint image using the feature vectors.

[0044] The encryption system includes an input mode and an unlock mode, combined with... Figure 2 The operation process of the input mode includes the following steps:

[0045] S1. The acquisition module acquires the user's fingerprint image and sends it to the image processing module;

[0046] S2. The image processing module judges the specifications of the fingerprint image. If it meets the requirements, it proceeds to step S3. If it does not meet the requirements, it returns to step S1 to re-acquire the image.

[0047] S3. The image processing module determines the fingerprint ring center of the fingerprint image and segments the fingerprint image based on the fingerprint ring center position;

[0048] S4. The feature processing module converts each segmented image into a matrix, and processes the matrix to obtain a feature vector.

[0049] S5. The information storage module saves the feature vector;

[0050] Combination Figure 3 The operation process of the unlocking mode includes the following steps:

[0051] S6. The acquisition module acquires the user's fingerprint image and sends it to the verification module;

[0052] S7. The verification module obtains all feature vector information from the information storage module;

[0053] S8. The verification module sequentially verifies and analyzes the feature vector information with the fingerprint image. When one feature vector information passes the verification, the unlocking is successful. When all feature vector information fails the verification, the unlocking fails.

[0054] In step S2, the process by which the image processing module determines the specifications of the fingerprint image includes the following steps:

[0055] S21. Obtain pixels in the fingerprint image whose grayscale value is less than the ridge threshold. These pixels are called ridge pixels.

[0056] S22. Divide the ridge line pixels into at least one quasi-ridge line set based on the connectivity of the ridge line pixels;

[0057] S23. Perform width analysis on each quasi-ridge set. When the maximum width of the quasi-ridge set is less than the width threshold, the quasi-ridge set is called a ridge set.

[0058] S24. Count the number of ridge sets. When the number of ridge sets is within a standard range, the fingerprint image is determined to meet the specifications. The standard range is obtained by those skilled in the art through experimental experience data.

[0059] In step S3, the process of determining the fingerprint ring center includes the following steps:

[0060] S31. Obtain the set of ridge lines from step S2;

[0061] S32. Obtain the annular portion of each ridge set and calculate the center position of the annular portion;

[0062] S33. Calculate the average value of the center positions of all ridge sets to obtain the fingerprint ring center;

[0063] In step S3, the process of segmenting the fingerprint image includes the following steps:

[0064] S34. Draw a straight line through the center point, which divides all ridge set into a left half set and a right half set. The straight line satisfies that the number of pixels in the left half set is equal to the number of pixels in the right half set. Note that a difference of 1 in the number of pixels is considered as equal.

[0065] S35. Establish a coordinate system with the center point as the origin and the straight line as the y-axis;

[0066] S36. Divide the fingerprint image into 6 square images according to the coordinate system, and combine... Figure 4The coordinate range of each square image is: {x∈[-3a, -a], y∈[a, 3a]}, {x∈[-a, a], y∈[a, 3a]}, {x∈[a, 3a], y∈[a, 3a]}, {x∈[-3a, -a], y∈[-a, a]}, {x∈[-a, a], y∈[-a, a]}, {x∈[a, 3a], y∈[-a, a]};

[0067] Where 2a is the side length of the square image in the coordinate system;

[0068] In step S4, the feature processing module processes the square image using the following steps:

[0069] S41. Each square image is converted into a matrix by converting the ridge pixels of the square image to 1 and the non-ridge pixels to 0. The six matrices obtained according to the order of the square images from left to right and from top to bottom are denoted as X1, X2, X3, X4, X5 and X6 respectively.

[0070] S42. Add the above six matrices together to obtain a new matrix Y;

[0071] S43. Count the number of elements in matrix Y with values ​​of 0, 1, 2, 3, 4, 5, and 6, and denote them as n1, n2, n3, n4, n5, n6, and n7 respectively.

[0072] S44. Calculate the eigenvalues ​​and eigenvectors of matrices X1, X2, X3, X4, X5, X6, and Y, respectively. The eigenvalues ​​are denoted as λ1, λ2, λ3, λ4, λ5, λ6, and λ7, and the eigenvectors are denoted as... and

[0073] S45. Calculate the final eigenvector according to the following formula.

[0074]

[0075] Feature vector Recorded in the information storage module;

[0076] The method for finding the eigenvalues ​​and eigenvectors of a matrix X is as follows: Let det|X-λI|=0, calculate the eigenvalues ​​λ, where I is the identity matrix, and then... Calculate the eigenvectors

[0077] Combination Figure 5 In step S8, the process by which the verification module performs verification analysis on the fingerprint image and feature vector includes:

[0078] S81. Extract a 2a*2a square image from the fingerprint image;

[0079] S82. Convert the captured image into matrix Z by converting ridge pixels to 1 and non-ridge pixels to 0.

[0080] S83. Let k = 1, and create matrix Z. k =Z;

[0081] S84, Matrix Z k With feature vectors Multiply to obtain a vector

[0082] S85, Matrix Z k The elements in the first column are moved to the last column to obtain matrix Z. k+1 Let k be incremented by 1;

[0083] S86. Repeat steps S84 and S85 a total of m times to obtain m vectors. Then proceed to step S87, where m is the number of columns in matrix Z;

[0084] S87, Given m vectors Construct an m*m verification matrix Z0, and calculate the determinant value of matrix Z0, D = det(Z0);

[0085] S88. When the absolute value of the determinant value D is less than the verification threshold, the fingerprint image passes the verification.

[0086] All matrices mentioned above are m*m matrices, and all vectors mentioned above are m*1 vectors.

[0087] The content disclosed above is only a preferred and feasible embodiment of the present invention, and is not intended to limit the scope of protection of the present invention. Therefore, all equivalent technical changes made based on the content of the present invention specification and drawings are included within the scope of protection of the present invention. Furthermore, the elements therein can be updated as technology develops.

Claims

1. A feature vector based fingerprinting encryption system, characterized in that, The system comprises a fingerprint collection module, an image processing module, a feature processing module, an information storage module and a collation module, the fingerprint collection module is used for collecting a fingerprint image, the image processing module is used for pre-processing the collected fingerprint image, the feature processing module is used for processing the pre-processed image into a feature vector, the information storage module is used for storing the feature vector, and the collation module is used for identifying the fingerprint image based on the feature vector. The image processing module divides the fingerprint image into six square pictures, the feature processing module converts the six pictures into six matrices X1, X2, X3, X4, X5 and X6, the elements in the six matrices are composed of 0 and 1, wherein 1 represents a ridge pixel point in the fingerprint image, and 0 represents a non-ridge pixel point in the fingerprint image, and the feature processing module obtains a comprehensive matrix Y according to the following formula: Y = X1 + X2 + X3 + X4 + X5 + X6. The feature processing module counts the number of elements with values of 0, 1, 2, 3, 4, 5 and 6 in the comprehensive matrix, and records them as n1, n2, n3, n4, n5, n6 and n7 respectively. The feature processing module calculates the feature vector saved in the information storage module according to the following formula where λ i (i = 1, 2, 3, 4, 5, 6, 7) are eigenvalues of matrices X1, X2, X3, X4, X5, X6, and Y, respectively, are eigenvectors of matrices X1, X2, X3, X4, X5, X6, and Y, respectively. The proofreading module cuts a square picture from the collected fingerprint image and converts the square picture into a matrix Z, and obtains m new matrices Z by moving the first column elements in the matrix Z to the last column one by one k , (k=1, 2, 3, ···, m), wherein m is the column number of the matrix Z, the proofreading module multiplies the matrix Z k with a characteristic vector to obtain m vectors The proofreading module constructs a verification matrix Z0 from the m vectors and calculates the determinant value D=det(Z0) of the verification matrix Z0, and when the absolute value of the determinant value D is less than a proofreading threshold, the fingerprint image passes the proofreading.

2. A feature vector based fingerprinting encryption system as claimed in claim 1, wherein, The system comprises an entry mode and an unlocking mode, in the entry mode, the image collected by the fingerprint collection module is processed by the image processing module and the feature processing module to generate a feature vector which is saved in the information storage module, in the unlocking mode, the collation module obtains the feature vector from the information storage module to collate and analyze the fingerprint image obtained from the fingerprint collection module, and the unlocking is successful after the collation is passed.

3. A feature vector based fingerprinting encryption system as claimed in claim 2, wherein, The image processing module performs specification judgment on the collected fingerprint image, and does not perform subsequent operation on the fingerprint image when the fingerprint image does not conform to the specification.

4. A feature vector based fingerprinting encryption system as claimed in claim 3, wherein, The image processing module divides the pixel points in the fingerprint image into ridge pixel points and non-ridge pixel points, sets the ridge pixel points with connectivity as a quasi-ridge set, and when the width of the quasi-ridge set is less than a width threshold, the quasi-ridge set is called a ridge set, the image processing module counts the number of ridge sets, and when the number of ridge sets is within a standard range, the fingerprint image conforms to the specification.

5. A feature vector based fingerprinting encryption system as claimed in claim 4, characterized in that, The image processing module establishes a coordinate system based on the center of the fingerprint loop, and divides the fingerprint image into six square pictures according to the coordinate system, and the coordinate range of each square picture is: {x∈[-3a,-a],y∈[a,3a]}, {x∈[-a,a],y∈[a,3a]}, {x∈[a,3a],y∈[a,3a]}, {x∈[-3a,-a],y∈[-a,a]}, {x∈[-a,a],y∈[-a,a]}, {x∈[a,3a],y∈[-a,a]}; Wherein, 2a is the side length of the square picture in the coordinate system.

Images